SC 14D9/A

SECURITIES AND EXCHANGE COMMISSION

WASHINGTON, DC 20549

SCHEDULE 14D-9

SOLICITATION/RECOMMENDATION STATEMENT
PURSUANT TO SECTION 14(d)(4) OF THE
SECURITIES EXCHANGE ACT OF 1934

(Amendment No. 31)

PEOPLESOFT, INC.

(Name of Subject Company)

PEOPLESOFT, INC.

(Name of Person Filing Statement)

Common Stock, Par Value $0.01 Per Share

(Title of Class of Securities)

712713106

(CUSIP Number of Class of Securities)

Craig Conway
President and Chief Executive Officer
PeopleSoft, Inc.
4460 Hacienda Drive, Pleasanton, California 94588-8618
(925) 225-3000

(Name, Address and Telephone Number of Person Authorized to Receive
Notice and Communications on Behalf of the Person Filing Statement)

COPIES TO:

Douglas D. Smith, Esq.
Gibson, Dunn & Crutcher LLP
One Montgomery Street
San Francisco, California 94104
(415) 393-8200

     o Check the box if the filing relates solely to preliminary communications made before the commencement of a tender offer

1

Purpose of Amendment

     The purpose of this amendment is to amend and supplement Items 6 and 8 in the Solicitation/Recommendation Statement on Schedule
14D-9 previously filed by PeopleSoft, Inc. (“PeopleSoft” or the “Company”) on June 12, 2003 and subsequently amended, and to add additional exhibits to Item 9 and amend the exhibit index accordingly.

2

Item 6. Interest in Securities of the Subject Company

     Item 6 is hereby amended and supplemented as follows:

     Except as described below and except as disclosed in the Schedule 14D-9 previously filed by the Company, as subsequently amended, no transactions with respect to the Common Stock of the Company have been effected by the Company or, to the Company’s best knowledge, by any of its executive officers, directors, affiliates or subsidiaries during the past 60 days.

				Number Shares
	Date of		Nature of	of Common
Name	Transaction		Transaction	Stock		Purchase/Sale Price
David A. Duffield		6/30/04	Disposition(1)		25,000	$18.52

(1) Pursuant to Rule 10b5-1(c) trading plan, providing for automatic transactions upon establishment of a written contract, plan or instructions
      under conditions specified in the Rule.

Item 8. Additional Information

     Litigation Matters

     On June 24, 2004, Oracle Corporation and Pepper Acquisition Corp. filed an amended complaint for declaratory and injunctive relief in the Court of Chancery of the State of Delaware in and for New Castle County, a redacted copy of which is filed as exhibit (a)(104) to this Schedule 14D-9. The redacted portions of the complaint were filed with the Court under seal pursuant to Court rules.

     Weblog Posting

     On June 26, 28 and 30, and July 1, 2004, PeopleSoft posted to its website at www.peoplesoft.com additional installments in a weblog that periodically will be updated during the course of the antitrust trial. A copy of the additional installments is attached as exhibit (e)(23) and exhibit (e)(23)(a) to this Schedule 14D-9.

3

Item 9. Materials to Be Filed as Exhibits

Exhibit No.	Document
(a)(1)	Press release issued by PeopleSoft on June 12, 2003 (1)
(a)(2)	Press release issued by PeopleSoft on June 6, 2003 (incorporated by reference to PeopleSoft’s Schedule 14D-9C filed with the SEC on June 7, 2003) (1)
(a)(3)	Letter, dated June 13, 2003, to PeopleSoft’s stockholders (2)
(a)(4)	Letter to customers issued June 16, 2003 (incorporated by reference to PeopleSoft’s June 16, 2003 425 filing) (3)
(a)(5)	Investor presentation materials (incorporated by reference to PeopleSoft’s June 17, 2003 425 filing) (3)
(a)(6)	Press release issued by CRN (incorporated by reference to PeopleSoft’s June 19, 2003 425 filing) (4)
(a)(7)	Press release issued by CNET News.com (incorporated by reference to PeopleSoft’s June 19, 2003 425 filing) (4)
(a)(8)	Transcript of conference call held by PeopleSoft (incorporated by reference to PeopleSoft’s June 13, 2003 425 filing) (4)
(a)(9)	Complaint filed by PeopleSoft in the Superior Court of the State of California, County of Alameda (4)
(a)(10)	Press release issued by PeopleSoft (incorporated by reference to PeopleSoft’s June 16, 2003 425 filing) (4)
(a)(11)	Press release issued by ComputerWeekly.com (incorporated by reference to PeopleSoft’s June 19, 2003 425 filing) (4)
(a)(12)	Press release issued by The Motley Fool (incorporated by reference to PeopleSoft’s June 19, 2003 425 filing) (4)
(a)(13)	Press release issued by the Higher Education User Group (incorporated by reference to PeopleSoft’s June 19, 2003 425 filing) (4)
(a)(14)	Text of information posted on PeopleSoft’s website (incorporated by reference to PeopleSoft’s June 18, 2003 425 filing) (4)
(a)(15)	Press release issued by PeopleSoft (incorporated by reference to PeopleSoft’s June 18, 2003 425 filing) (4)
(a)(16)	Press release issued by the Distributors & Manufacturers’ User Group (incorporated by reference to PeopleSoft’s June 19, 2003 425 filing) (4)
(a)(17)	Press release issued by the Connecticut Attorney General’s Office (4)
(a)(18)	Press release issued by PeopleSoft on June 20, 2003 (4)
(a)(19)	Investor presentation materials (incorporated by reference to PeopleSoft’s June 23, 2003 425 filing) (5)
(a)(20)	Letter to PeopleSoft employees (incorporated by reference to PeopleSoft’s June 23, 2003 425 filing) (5)
(a)(21)	Press release issued by eWeek (incorporated by reference to PeopleSoft’s June 23, 2003 425 filing) (5)
(a)(22)	Press release issued by the Healthcare Industry User Group (incorporated by reference to PeopleSoft’s June 24, 2003 425 filing) (5)
(a)(23)	Advertisement placed by PeopleSoft (incorporated by reference to PeopleSoft’s June 23, 2003 425 filing) (5)
(a)(24)	Advertisement placed by PeopleSoft (incorporated by reference to PeopleSoft’s June 24, 2003 425 filing) (5)
(a)(25)	Text of International Customer Advisory Board’s e-mail posted on PeopleSoft’s website (incorporated by reference to PeopleSoft’s June 24, 2003 425 filing) (5)
(a)(26)	Press release issued by Quest (incorporated by reference to PeopleSoft’s June 24, 2003 425 filing) (5)

4

(a)(27)	Advertisement placed by PeopleSoft (incorporated by reference to PeopleSoft’s June 25, 2003 425 filing) (5)
(a)(28)	Transcript of TriNet webcast posted on PeopleSoft’s website (incorporated by reference to PeopleSoft’s June 25, 2003 425 filing) (5)
(a)(29)	Transcript of CNBC webcast posted on PeopleSoft’s website (incorporated by reference to PeopleSoft’s June 25, 2003 425 filing) (5)
(a)(30)	Press release issued by the International Customer Advisory Board (incorporated by reference to PeopleSoft’s June 23, 2003 425 filing) (5)
(a)(31)	Press release issued by PeopleSoft (incorporated by reference to PeopleSoft’s July 1, 2003 425 filing) (5)
(a)(32)	Press release issued by PeopleSoft (incorporated by reference to PeopleSoft’s July 1, 2003 425 filing) (5)
(a)(33)	Press release issued by PeopleSoft (incorporated by reference to PeopleSoft’s July 2, 2003 425 filing) (5)
(a)(34)	Transcript of conference call held by PeopleSoft (incorporated by reference to PeopleSoft’s July 2, 2003 425 filing) (5)
(a)(35)	Advertisement placed by PeopleSoft on July 2, 2003 (incorporated by reference to PeopleSoft’s July 2, 2003 425 filing) (5)
(a)(36)	Press release issued by PeopleSoft (incorporated by reference to PeopleSoft’s July 7, 2003 425 filing) (6)
(a)(37)	Press release issued by InformationWeek (incorporated by reference to PeopleSoft’s July 8, 2003 425 filing) (6)
(a)(38)	Press release issued by PeopleSoft (incorporated by reference to PeopleSoft’s July 14, 2003 425 filing) (6)
(a)(39)	Press release issued by PeopleSoft (incorporated by reference to PeopleSoft’s July 16, 2003 425 filing) (6)
(a)(40)	Press release issued by CRMDaily.com (incorporated by reference to PeopleSoft’s July 16, 2003 425 filing) (6)
(a)(41)	Press release issued by PeopleSoft (incorporated by reference to PeopleSoft’s July 18, 2003 425 filing) (6)
(a)(42)	Advertisement placed by PeopleSoft (incorporated by reference to PeopleSoft’s July 22, 2003 425 filing) (6)
(a)(43)	Advertisement placed by PeopleSoft (incorporated by reference to PeopleSoft’s July 24, 2003 425 filing) (6)
(a)(44)	Press release issued by PeopleSoft (incorporated by reference to PeopleSoft’s July 24, 2003 425 filing) (6)
(a)(45)	Press release issued by PeopleSoft (incorporated by reference to PeopleSoft’s July 28, 2003 425 filing) (7)
(a)(46)	Press release issued by International Customer Advisory Board and Quest (incorporated by reference to PeopleSoft’s July 29, 2003 425 filing) (7)
(a)(47)	Press release issued by PeopleSoft (incorporated by reference to PeopleSoft’s August 13, 2003 425 filing) (7)
(a)(48)	Redacted version of First Amended Complaint filed by PeopleSoft in the Superior Court of the State of California, County of Alameda (7)
(a)(49)	Press release issued by PeopleSoft on August 29, 2003 (9)
(a)(50)	Press release issued by PeopleSoft on September 4, 2003 (9)
(a)(51)	Unredacted version of First Amended Complaint filed by PeopleSoft in the Superior Court of the State of California, County of Alameda (9)
(a)(52)	Transcript of PeopleSoft Analyst Day conference held by PeopleSoft on September 4, 2003 (9)
(a)(53)	PeopleSoft Analyst Day Power Point presentation materials (9)
(a)(54)	PeopleSoft Analyst Day reconciliation of Non-GAAP to GAAP financial measures (9)

5

(a)(55)	Excerpts from transcript of conference call held by PeopleSoft on October 23, 2003 (10)
(a)(56)	Order entered by the Superior Court of California, County of Alameda (11)
(a)(57)	Press release issued by PeopleSoft on November 17, 2003 (12)
(a)(58)	Press release issued by PeopleSoft on December 2, 2003 (13)
(a)(59)	Second Amended Complaint filed by PeopleSoft in the Superior Court of the State of California, County of Alameda (14)
(a)(60)	Text of background information provided to PeopleSoft sales force to respond to customer inquiries following Oracle statements of November 24, 2003 (15)
(a)(61)	Press release issued by PeopleSoft on December 19, 2003 (16)
(a)(62)	Press release issued by PeopleSoft on January 12, 2004 (16)
(a)(63)	E-mail sent to PeopleSoft employees (17)
(a)(64)	Letter to the Wall Street Journal dated January 22, 2004 (17)
(a)(65)	Press release issued by PeopleSoft on January 24, 2004 (17)
(a)(66)	Press release issued by PeopleSoft on January 30, 2004 (17)
(a)(67)	Excerpts from transcript of conference call held by PeopleSoft on January 29, 2004 (17)
(a)(68)	Order entered by the Superior Court of California, County of Alameda (17)
(a)(69)	Oracle and Pepper Acquisition Corp. Notice of Demurrer and Demurrer (17)
(a)(70)	Oracle and Pepper Acquisition Corp. Notice of Motion to Strike and Motion to Strike (17)
(a)(71)	Advertisement placed by PeopleSoft on February 3, 2004 (17)
(a)(72)	Press release issued by PeopleSoft on February 4, 2004 (18)
(a)(73)	Message sent to PeopleSoft employees (18)
(a)(74)	Press release issued by PeopleSoft on February 9, 2004 (19)
(a)(75)	Letter to employees dated February 9, 2004 (19)
(a)(76)	Letter to customers dated February 9, 2004 (19)
(a)(77)	E-mail sent to PeopleSoft employees on February 25, 2004 (20)
(a)(78)	E-mail sent to PeopleSoft employees on February 27, 2004 (20)
(a)(79)	Investor presentation materials (20)
(a)(80)	Press release issued by PeopleSoft on February 10, 2004 (20)
(a)(81)	Press release issued by U.S. Department of Justice on February 26, 2004 (20)
(a)(82)	Press release issued by PeopleSoft on February 26, 2004 (20)
(a)(83)	Order entered by the Superior Court of California, County of Alameda, overruling Defendants’ Demurrer to Plaintiffs’ Second Amended Complaint (20)
(a)(84)	Order entered by the Superior Court of California, County of Alameda, denying Defendants’ Motion to Strike Portions of Plaintiffs’ Second Amended Complaint (20)
(a)(85)	Letter to stockholders dated March 3, 2004 (20)
(a)(86)	OneVoice e-mail newsletter sent to PeopleSoft employees on March 4, 2004(21)
(a)(87)	OneVoice e-mail newsletter sent to PeopleSoft employees on March 9, 2004(21)
(a)(88)	OneVoice e-mail newsletter sent to PeopleSoft employees on March 12, 2004(21)
(a)(89)	Press release issued by PeopleSoft on March 12, 2004(21)
(a)(90)	Press release issued by PeopleSoft on March 12, 2004(21)
(a)(91)	Press release issued by PeopleSoft on March 19, 2004(22)
(a)(92)	Presentation given at PeopleSoft’s 2004 Annual Meeting of Stockholders(22)
(a)(93)	Transcript of PeopleSoft’s 2004 Annual Meeting of Stockholders(22)
(a)(94)	Press release issued by PeopleSoft on March 25, 2004(22)
(a)(95)	Transcript of conference call held by PeopleSoft on January 29,2004 (incorporated by reference to Exhibit 99.2 to PeopleSoft’s February 4, 2004 Form 8-K)
(a)(96)	Redacted version of the Cross-complaint filed by Oracle in the Superior Court of the State of California, County of Alameda(23)
(a)(97)	Press release issued by the Michigan Attorney General’s Office on April 7, 2004(23)
(a)(98)	Press release issued by the Ohio Attorney General’s Office on April 9, 2004(23)
(a)(99)	Excerpts from transcript of conference call held by PeopleSoft on April 22, 2004(24)
(a)(100)	Press release issued by PeopleSoft on May 14, 2004(25)
(a)(101)	Excerpts from transcript of PeopleSoft’s 2004 Leadership Summit News Conference held on May 18, 2004(25)
(a)(102)	Press release issued by PeopleSoft on May 26, 2004(26)
(a)(103)	Press release issued by PeopleSoft on May 26, 2004(26)
(a)(104)	Redacted version of the Amended Complaint for Declaratory and Injunctive Relief filed by Oracle and Pepper Acquisition Corp. in the Delaware Court of Chancery
(e)(1)	Excerpts from PeopleSoft’s Definitive Proxy Statement dated April 28, 2003 relating to the 2003 Annual Meeting of Stockholders (1)
(e)(2)	Employment Agreement, dated May 10, 1999, by and between Craig Conway and PeopleSoft, Inc., (incorporated by reference to Exhibit 10.47 filed with PeopleSoft’s Annual Report on Form 10-K for the year ended December 31, 1999) (1)
(e)(3)	Employment Contract, dated as of January 1, 2000, with addendums thereto dated as of January 1, 2000, and January 1, 2001, by and between Guy Dubois and PeopleSoft France S.A. (incorporated by reference to Exhibit 10.45 filed with PeopleSoft’s Annual Report on Form 10-K for the year ended December 31, 2001) (1)
(e)(4)	Executive Severance Policy – Executive Vice Presidents, effective as of January 1, 2003 (8)
(e)(5)	Executive Severance Policy – Senior Vice Presidents, effective as of January 1, 2003 (8)
(e)(6)	Terms of Customer Assurance Program (revised) (10)
(e)(6)(i)	Terms of Customer Assurance Program (replacement version) + (11)
(e)(7)	Form of letter sent to customers (11)
(e)(8)	Terms of Customer Assurance Program (11)
(e)(9)	Amendment No. 1 to the Bylaws of PeopleSoft (11)
(e)(10)	Terms of Customer Assurance Program (extension term) (12)
(e)(11)	Employment Agreement, dated January 30, 2004, by and between Craig Conway and PeopleSoft, Inc. (17)
(e)(12)	Excerpts from PeopleSoft’s Definitive Proxy Statement dated February 20, 2004 relating to the 2004 Annual Meeting of Stockholders (22)
(e)(13)	Memorandum of understanding regarding settlement of stockholder class actions(26)
(e)(14)	Amendment to memorandum of understanding regarding settlement of stockholder class actions(27)
(e)(15)	White paper dated February 1, 2004(28)
(e)(16)	June 15, 2004 weblog postings(28)
(e)(17)	Second amendment to memorandum of understanding regarding settlement of stockholder class actions(28)
(e)(18)	Amended Executive Severance Policy—Executive Vice Presidents, amended as of June 14, 2004(29)
(e)(19)	Amended Executive Severance Policy—Senior Vice Presidents, amended as of June 14, 2004(29)
(e)(20)	June 16 and 17, 2004 weblog postings(29)
(e)(21)	Stipulation and Agreement of Compromise, Settlement and Release dated June 17, 2004(29)
(e)(22)	June 20, 21 and 23, 2004 weblog postings(30)
(e)(23)	June 26, 28 and 30, and July 1, 2004 weblog postings
(e)(23)(a)	Chart referred to in the “The Government Rests” weblog posting

(1)	Previously filed as an exhibit to PeopleSoft’s Schedule 14D-9 filed with the SEC June 12, 2003.
(2)	Previously filed as an exhibit to PeopleSoft’s Amendment No. 1 to Schedule 14D-9 filed with the SEC June 13, 2003.
(3)	Previously filed as an exhibit to PeopleSoft’s Amendment No. 2 to Schedule 14D-9 filed with the SEC June 17, 2003.
(4)	Previously filed as an exhibit to PeopleSoft’s Amendment No. 3 to Schedule 14D-9 filed with the SEC June 20, 2003.
(5)	Previously filed as an exhibit to PeopleSoft’s Amendment No. 4 to Schedule 14D-9 filed with the SEC July 3, 2003.
(6)	Previously filed as an exhibit to PeopleSoft’s Amendment No. 5 to Schedule 14D-9 filed with the SEC July 25, 2003.
(7)	Previously filed as an exhibit to PeopleSoft’s Amendment No. 6 to Schedule 14D-9 filed with the SEC August 14, 2003.
(8)	Previously filed as an exhibit to PeopleSoft’s Amendment No. 7 to Schedule 14D-9 filed with the SEC August 22, 2003.
(9)	Previously filed as an exhibit to PeopleSoft’s Amendment No. 8 to Schedule 14D-9 filed with the SEC September 11, 2003.
(10)	Previously filed as an exhibit to PeopleSoft’s Amendment No. 9 to Schedule 14D-9 filed with the SEC October 27, 2003.
+	This exhibit replaces and supersedes exhibit (e)(6), which previously was filed in error.
(11)	Previously filed as an exhibit to PeopleSoft’s Amendment No. 11 to Schedule 14D-9 filed with the SEC November 17, 2003.
(12)	Previously filed as an exhibit to PeopleSoft’s Amendment No. 12 to Schedule 14D-9 filed with the SEC November 19, 2003.
(13)	Previously filed as an exhibit to PeopleSoft’s Amendment No. 13 to Schedule 14D-9 filed with the SEC December 5, 2003.
(14)	Previously filed as an exhibit to PeopleSoft’s Amendment No. 14 to Schedule 14D-9 filed with the SEC December 15, 2003.
(15)	Previously filed as an exhibit to PeopleSoft’s Amendment No. 15 to Schedule 14D-9 filed with the SEC December 19, 2003.
(16)	Previously filed as an exhibit to PeopleSoft’s Amendment No. 16 to Schedule 14D-9 filed with the SEC January 13, 2004.
(17)	Previously filed as an exhibit to PeopleSoft’s Amendment No. 17 to Schedule 14D-9 filed with the SEC February 3, 2004.
(18)	Previously filed as an exhibit to PeopleSoft’s Amendment No. 18 to Schedule 14D-9 filed with the SEC February 4, 2004.
(19)	Previously filed as an exhibit to PeopleSoft’s Amendment No. 19 to Schedule 14D-9 filed with the SEC February 9, 2004.
(20)	Previously filed as an exhibit to PeopleSoft’s Amendment No. 20 to Schedule 14D-9 filed with the SEC March 4, 2004.
(21)	Previously filed as an exhibit to PeopleSoft’s Amendment No. 21 to Schedule 14D-9 filed with the SEC March 15, 2004.
(22)	Previously filed as an exhibit to PeopleSoft’s Amendment No. 22 to Schedule 14D-9 filed with the SEC March 29, 2004.
(23)	Previously filed as an exhibit to PeopleSoft’s Amendment No. 23 to Schedule 14D-9 filed with the SEC April 13, 2004.
(24)	Previously filed as an exhibit to PeopleSoft’s Amendment No. 24 to Schedule 14D-9 filed with the SEC April 28, 2004.
(25)	Previously filed as an exhibit to PeopleSoft’s Amendment No. 25 to Schedule 14D-9 filed with the SEC May 20, 2004.
(26)	Previously filed as an exhibit to PeopleSoft’s Amendment No. 26 to Schedule 14D-9 filed with the SEC May 27, 2004.
(27)	Previously filed as an exhibit to PeopleSoft’s Amendment No. 27 to Schedule 14D-9 filed with the SEC June 14, 2004.
(28)	Previously filed as an exhibit to PeopleSoft’s Amendment No. 28 to Schedule 14D-9 filed with the SEC June 16, 2004.
(29)	Previously filed as an exhibit to PeopleSoft’s Amendment No. 29 to Schedule 14D-9 filed with the SEC June 18, 2004.
(30)	Previously filed as an exhibit to PeopleSoft’s Amendment No. 30 to Schedule 14D-9 filed with the SEC June 28, 2004.

6

SIGNATURE

     After reasonable inquiry and to the best of my knowledge and belief, I certify that the information set forth in this statement is true, complete, and correct.

		PEOPLESOFT, INC.
	By:	/s/ KEVIN T. PARKER
		Kevin T. Parker Executive Vice President Finance and Administration, Chief Financial Officer
Date: July 2, 2004

7

Exhibit No.	Document
(a)(1)	Press release issued by PeopleSoft on June 12, 2003 (1)
(a)(2)	Press release issued by PeopleSoft on June 6, 2003 (incorporated by reference to PeopleSoft’s Schedule 14D-9C filed with the SEC on June 7, 2003) (1)
(a)(3)	Letter, dated June 13, 2003, to PeopleSoft’s stockholders (2)
(a)(4)	Letter to customers issued June 16, 2003 (incorporated by reference to PeopleSoft’s June 16, 2003 425 filing) (3)
(a)(5)	Investor presentation materials (incorporated by reference to PeopleSoft’s June 17, 2003 425 filing) (3)
(a)(6)	Press release issued by CRN (incorporated by reference to PeopleSoft’s June 19, 2003 425 filing) (4)
(a)(7)	Press release issued by CNET News.com (incorporated by reference to PeopleSoft’s June 19, 2003 425 filing) (4)
(a)(8)	Transcript of conference call held by PeopleSoft (incorporated by reference to PeopleSoft’s June 13, 2003 425 filing) (4)
(a)(9)	Complaint filed by PeopleSoft in the Superior Court of the State of California, County of Alameda (4)
(a)(10)	Press release issued by PeopleSoft (incorporated by reference to PeopleSoft’s June 16, 2003 425 filing) (4)
(a)(11)	Press release issued by ComputerWeekly.com (incorporated by reference to PeopleSoft’s June 19, 2003 425 filing) (4)
(a)(12)	Press release issued by The Motley Fool (incorporated by reference to PeopleSoft’s June 19, 2003 425 filing) (4)
(a)(13)	Press release issued by the Higher Education User Group (incorporated by reference to PeopleSoft’s June 19, 2003 425 filing) (4)
(a)(14)	Text of information posted on PeopleSoft’s website (incorporated by reference to PeopleSoft’s June 18, 2003 425 filing) (4)
(a)(15)	Press release issued by PeopleSoft (incorporated by reference to PeopleSoft’s June 18, 2003 425 filing) (4)
(a)(16)	Press release issued by the Distributors & Manufacturers’ User Group (incorporated by reference to PeopleSoft’s June 19, 2003 425 filing) (4)
(a)(17)	Press release issued by the Connecticut Attorney General’s Office (4)
(a)(18)	Press release issued by PeopleSoft on June 20, 2003 (4)
(a)(19)	Investor presentation materials (incorporated by reference to PeopleSoft’s June 23, 2003 425 filing) (5)
(a)(20)	Letter to PeopleSoft employees (incorporated by reference to PeopleSoft’s June 23, 2003 425 filing) (5)
(a)(21)	Press release issued by eWeek (incorporated by reference to PeopleSoft’s June 23, 2003 425 filing) (5)
(a)(22)	Press release issued by the Healthcare Industry User Group (incorporated by reference to PeopleSoft’s June 24, 2003 425 filing) (5)
(a)(23)	Advertisement placed by PeopleSoft (incorporated by reference to PeopleSoft’s June 23, 2003 425 filing) (5)
(a)(24)	Advertisement placed by PeopleSoft (incorporated by reference to PeopleSoft’s June 24, 2003 425 filing) (5)
(a)(25)	Text of International Customer Advisory Board’s e-mail posted on PeopleSoft’s website (incorporated by reference to PeopleSoft’s June 24, 2003 425 filing) (5)
(a)(26)	Press release issued by Quest (incorporated by reference to PeopleSoft’s June 24, 2003 425 filing) (5)
(a)(27)	Advertisement placed by PeopleSoft (incorporated by reference to PeopleSoft’s June 25,

8

	2003 425 filing) (5)
(a)(28)	Transcript of TriNet webcast posted on PeopleSoft’s website (incorporated by reference to PeopleSoft’s June 25, 2003 425 filing) (5)
(a)(29)	Transcript of CNBC webcast posted on PeopleSoft’s website (incorporated by reference to PeopleSoft’s June 25, 2003 425 filing) (5)
(a)(30)	Press release issued by the International Customer Advisory Board (incorporated by reference to PeopleSoft’s June 23, 2003 425 filing) (5)
(a)(31)	Press release issued by PeopleSoft (incorporated by reference to PeopleSoft’s July 1, 2003 425 filing) (5)
(a)(32)	Press release issued by PeopleSoft (incorporated by reference to PeopleSoft’s July 1, 2003 425 filing) (5)
(a)(33)	Press release issued by PeopleSoft (incorporated by reference to PeopleSoft’s July 2, 2003 425 filing) (5)
(a)(34)	Transcript of conference call held by PeopleSoft (incorporated by reference to PeopleSoft’s July 2, 2003 425 filing) (5)
(a)(35)	Advertisement placed by PeopleSoft on July 2, 2003 (incorporated by reference to PeopleSoft’s July 2, 2003 425 filing) (5)
(a)(36)	Press release issued by PeopleSoft (incorporated by reference to PeopleSoft’s July 7, 2003 425 filing) (6)
(a)(37)	Press release issued by InformationWeek (incorporated by reference to PeopleSoft’s July 8, 2003 425 filing) (6)
(a)(38)	Press release issued by PeopleSoft (incorporated by reference to PeopleSoft’s July 14, 2003 425 filing) (6)
(a)(39)	Press release issued by PeopleSoft (incorporated by reference to PeopleSoft’s July 16, 2003 425 filing) (6)
(a)(40)	Press release issued by CRMDaily.com (incorporated by reference to PeopleSoft’s July 16, 2003 425 filing) (6)
(a)(41)	Press release issued by PeopleSoft (incorporated by reference to PeopleSoft’s July 18, 2003 425 filing) (6)
(a)(42)	Advertisement placed by PeopleSoft (incorporated by reference to PeopleSoft’s July 22, 2003 425 filing) (6)
(a)(43)	Advertisement placed by PeopleSoft (incorporated by reference to PeopleSoft’s July 24, 2003 425 filing) (6)
(a)(44)	Press release issued by PeopleSoft (incorporated by reference to PeopleSoft’s July 24, 2003 425 filing) (6)
(a)(45)	Press release issued by PeopleSoft (incorporated by reference to PeopleSoft’s July 28, 2003 425 filing) (7)
(a)(46)	Press release issued by International Customer Advisory Board and Quest (incorporated by reference to PeopleSoft’s July 29, 2003 425 filing) (7)
(a)(47)	Press release issued by PeopleSoft (incorporated by reference to PeopleSoft’s August 13, 2003 425 filing) (7)
(a)(48)	Redacted version of First Amended Complaint filed by PeopleSoft in the Superior Court of the State of California, County of Alameda (7)
(a)(49)	Press release issued by PeopleSoft on August 29, 2003 (9)
(a)(50)	Press release issued by PeopleSoft on September 4, 2003 (9)
(a)(51)	Unredacted version of First Amended Complaint filed by PeopleSoft in the Superior Court of the State of California, County of Alameda (9)
(a)(52)	Transcript of PeopleSoft Analyst Day conference held by PeopleSoft on September 4, 2003 (9)
(a)(53)	PeopleSoft Analyst Day Power Point presentation materials (9)
(a)(54)	PeopleSoft Analyst Day reconciliation of Non-GAAP to GAAP financial measures (9)
(a)(55)	Excerpts from transcript of conference call held by PeopleSoft on October 23, 2003 (10)

9

(a)(56)	Order entered by the Superior Court of California, County of Alameda (11)
(a)(57)	Press release issued by PeopleSoft on November 17, 2003 (12)
(a)(58)	Press release issued by PeopleSoft on December 2, 2003 (13)
(a)(59)	Second Amended Complaint filed by PeopleSoft in the Superior Court of the State of California, County of Alameda (14)
(a)(60)	Text of background information provided to PeopleSoft sales force to respond to customer inquiries following Oracle statements of November 24, 2003 (15)
(a)(61)	Press release issued by PeopleSoft on December 19, 2003 (16)
(a)(62)	Press release issued by PeopleSoft on January 12, 2004 (16)
(a)(63)	E-mail sent to PeopleSoft employees (17)
(a)(64)	Letter to the Wall Street Journal dated January 22, 2004 (17)
(a)(65)	Press release issued by PeopleSoft on January 24, 2004 (17)
(a)(66)	Press release issued by PeopleSoft on January 30, 2004 (17)
(a)(67)	Excerpts from transcript of conference call held by PeopleSoft on January 29, 2004 (17)
(a)(68)	Order entered by the Superior Court of California, County of Alameda (17)
(a)(69)	Oracle and Pepper Acquisition Corp. Notice of Demurrer and Demurrer (17)
(a)(70)	Oracle and Pepper Acquisition Corp. Notice of Motion to Strike and Motion to Strike (17)
(a)(71)	Advertisement placed by PeopleSoft on February 3, 2004 (17)
(a)(72)	Press release issued by PeopleSoft on February 4, 2004 (18)
(a)(73)	Message sent to PeopleSoft employees (18)
(a)(74)	Press release issued by PeopleSoft on February 9, 2004 (19)
(a)(75)	Letter to employees dated February 9, 2004 (19)
(a)(76)	Letter to customers dated February 9, 2004 (19)
(a)(77)	E-mail sent to PeopleSoft employees on February 25, 2004 (20)
(a)(78)	E-mail sent to PeopleSoft employees on February 27, 2004 (20)
(a)(79)	Investor presentation materials (20)
(a)(80)	Press release issued by PeopleSoft on February 10, 2004 (20)
(a)(81)	Press release issued by U.S. Department of Justice on February 26, 2004 (20)
(a)(82)	Press release issued by PeopleSoft on February 26, 2004 (20)
(a)(83)	Order entered by the Superior Court of California, County of Alameda, overruling Defendants’ Demurrer to Plaintiffs’ Second Amended Complaint (20)
(a)(84)	Order entered by the Superior Court of California, County of Alameda, denying Defendants’ Motion to Strike Portions of Plaintiffs’ Second Amended Complaint (20)
(a)(85)	Letter to stockholders dated March 3, 2004 (20)
(a)(86)	OneVoice e-mail newsletter sent to PeopleSoft employees on March 4, 2004(21)
(a)(87)	OneVoice e-mail newsletter sent to PeopleSoft employees on March 9, 2004(21)
(a)(88)	OneVoice e-mail newsletter sent to PeopleSoft employees on March 12, 2004(21)
(a)(89)	Press release issued by PeopleSoft on March 12, 2004(21)
(a)(90)	Press release issued by PeopleSoft on March 12, 2004(21)
(a)(91)	Press release issued by PeopleSoft on March 19, 2004(22)
(a)(92)	Presentation given at PeopleSoft’s 2004 Annual Meeting of Stockholders(22)
(a)(93)	Transcript of PeopleSoft’s 2004 Annual Meeting of Stockholders(22)
(a)(94)	Press release issued by PeopleSoft on March 25, 2004(22)
(a)(95)	Transcript of conference call held by PeopleSoft on January 29,2004 (incorporated by reference to Exhibit 99.2 to PeopleSoft’s February 4, 2004 Form 8-K)
(a)(96)	Redacted version of the Cross-complaint filed by Oracle in the Superior Court of the State of California, County of Alameda(23)
(a)(97)	Press release issued by the Michigan Attorney General’s Office on April 7, 2004(23)
(a)(98)	Press release issued by the Ohio Attorney General’s Office on April 9, 2004(23)
(a)(99)	Excerpts from transcript of conference call held by PeopleSoft on April 22, 2004(24)
(a)(100)	Press release issued by PeopleSoft on May 14, 2004(25)
(a)(101)	Excerpts from transcript of PeopleSoft’s 2004 Leadership Summit News Conference held on May 18, 2004(25)
(a)(102)	Press release issued by PeopleSoft on May 26, 2004(26)
(a)(103)	Press release issued by PeopleSoft on May 26, 2004(26)
(a)(104)	Redacted version of the Amended Complaint for Declaratory and Injunctive Relief filed by Oracle and Pepper Acquisition Corp. in the Delaware Court of Chancery
(e)(1)	Excerpts from PeopleSoft’s Definitive Proxy Statement dated April 28, 2003 relating to the 2003 Annual Meeting of Stockholders (1)
(e)(2)	Employment Agreement, dated May 10, 1999, by and between Craig Conway and PeopleSoft, Inc., (incorporated by reference to Exhibit 10.47 filed with PeopleSoft’s Annual Report on Form 10-K for the year ended December 31, 1999) (1)
(e)(3)	Employment Contract, dated as of January 1, 2000, with addendums thereto dated as of January 1, 2000, and January 1, 2001, by and between Guy Dubois and PeopleSoft France S.A. (incorporated by reference to Exhibit 10.45 filed with PeopleSoft’s Annual Report on Form 10-K for the year ended December 31, 2001) (1)
(e)(4)	Executive Severance Policy – Executive Vice Presidents, effective as of January 1, 2003 (8)
(e)(5)	Executive Severance Policy – Senior Vice Presidents, effective as of January 1, 2003 (8)
(e)(6)	Terms of Customer Assurance Program (revised) (10)
(e)(6)(i)	Terms of Customer Assurance Program (replacement version) + (11)
(e)(7)	Form of letter sent to customers (11)
(e)(8)	Terms of Customer Assurance Program (11)
(e)(9)	Amendment No. 1 to the Bylaws of PeopleSoft (11)
(e)(10)	Terms of Customer Assurance Program (extension term) (12)
(e)(11)	Employment Agreement, dated January 30, 2004, by and between Craig Conway and PeopleSoft, Inc. (17)
(e)(12)	Excerpts from PeopleSoft’s Definitive Proxy Statement dated February 20, 2004 relating to the 2004 Annual Meeting of Stockholders (22)
(e)(13)	Memorandum of understanding regarding settlement of stockholder class actions(26)
(e)(14)	Amendment to memorandum of understanding regarding settlement of stockholder class actions(27)
(e)(15)	White paper dated February 1, 2004(28)
(e)(16)	June 15, 2004 weblog postings(28)
(e)(17)	Second amendment to memorandum of understanding regarding settlement of stockholder class actions(28)
(e)(18)	Amended Executive Severance Policy—Executive Vice Presidents, amended as of June 14, 2004(29)
(e)(19)	Amended Executive Severance Policy—Senior Vice Presidents, amended as of June 14, 2004(29)
(e)(20)	June 16 and 17, 2004 weblog postings(29)
(e)(21)	Stipulation and Agreement of Compromise, Settlement and Release dated June 17, 2004(29)
(e)(22)	June 20, 21 and 23, 2004 weblog postings(30)
(e)(23)	June 26, 28 and 30, and July 1, 2004 weblog postings
(e)(23)(a)	Chart referred to in the “The Government Rests” weblog posting

(1)	Previously filed as an exhibit to PeopleSoft’s Schedule 14D-9 filed with the SEC June 12, 2003.
(2)	Previously filed as an exhibit to PeopleSoft’s Amendment No. 1 to Schedule 14D-9 filed with the SEC June 13, 2003.
(3)	Previously filed as an exhibit to PeopleSoft’s Amendment No. 2 to Schedule 14D-9 filed with the SEC June 17, 2003.
(4)	Previously filed as an exhibit to PeopleSoft’s Amendment No. 3 to Schedule 14D-9 filed with the SEC June 20, 2003.
(5)	Previously filed as an exhibit to PeopleSoft’s Amendment No. 4 to Schedule 14D-9 filed with the SEC July 3, 2003.
(6)	Previously filed as an exhibit to PeopleSoft’s Amendment No. 5 to Schedule 14D-9 filed with the SEC July 25, 2003.
(7)	Previously filed as an exhibit to PeopleSoft’s Amendment No. 6 to Schedule 14D-9 filed with the SEC August 14, 2003.
(8)	Previously filed as an exhibit to PeopleSoft’s Amendment No. 7 to Schedule 14D-9 filed with the SEC August 22, 2003.
(9)	Previously filed as an exhibit to PeopleSoft’s Amendment No. 8 to Schedule 14D-9 filed with the SEC September 11, 2003.
(10)	Previously filed as an exhibit to PeopleSoft’s Amendment No. 9 to Schedule 14D-9 filed with the SEC October 27, 2003.
+	This exhibit replaces and supersedes exhibit (e)(6), which previously was filed in error.
(11)	Previously filed as an exhibit to PeopleSoft’s Amendment No. 11 to Schedule 14D-9 filed with the SEC November 17, 2003.
(12)	Previously filed as an exhibit to PeopleSoft’s Amendment No. 12 to Schedule 14D-9 filed with the SEC November 19, 2003.
(13)	Previously filed as an exhibit to PeopleSoft’s Amendment No. 13 to Schedule 14D-9 filed with the SEC December 5, 2003.
(14)	Previously filed as an exhibit to PeopleSoft’s Amendment No. 14 to Schedule 14D-9 filed with the SEC December 15, 2003.
(15)	Previously filed as an exhibit to PeopleSoft’s Amendment No. 15 to Schedule 14D-9 filed with the SEC December 19, 2003.
(16)	Previously filed as an exhibit to PeopleSoft’s Amendment No. 16 to Schedule 14D-9 filed with the SEC January 13, 2004.
(17)	Previously filed as an exhibit to PeopleSoft’s Amendment No. 17 to Schedule 14D-9 filed with the SEC February 3, 2004.
(18)	Previously filed as an exhibit to PeopleSoft’s Amendment No. 18 to Schedule 14D-9 filed with the SEC February 4, 2004.
(19)	Previously filed as an exhibit to PeopleSoft’s Amendment No. 19 to Schedule 14D-9 filed with the SEC February 9, 2004.
(20)	Previously filed as an exhibit to PeopleSoft’s Amendment No. 20 to Schedule 14D-9 filed with the SEC March 4, 2004.
(21)	Previously filed as an exhibit to PeopleSoft’s Amendment No. 21 to Schedule 14D-9 filed with the SEC March 15, 2004.
(22)	Previously filed as an exhibit to PeopleSoft’s Amendment No. 22 to Schedule 14D-9 filed with the SEC March 29, 2004.
(23)	Previously filed as an exhibit to PeopleSoft’s Amendment No. 23 to Schedule 14D-9 filed with the SEC April 13, 2004.
(24)	Previously filed as an exhibit to PeopleSoft’s Amendment No. 24 to Schedule 14D-9 filed with the SEC April 28, 2004.
(25)	Previously filed as an exhibit to PeopleSoft’s Amendment No. 25 to Schedule 14D-9 filed with the SEC May 20, 2004.
(26)	Previously filed as an exhibit to PeopleSoft’s Amendment No. 26 to Schedule 14D-9 filed with the SEC May 27, 2004.
(27)	Previously filed as an exhibit to PeopleSoft’s Amendment No. 27 to Schedule 14D-9 filed with the SEC June 14, 2004.
(28)	Previously filed as an exhibit to PeopleSoft’s Amendment No. 28 to Schedule 14D-9 filed with the SEC June 16, 2004.
(29)	Previously filed as an exhibit to PeopleSoft’s Amendment No. 29 to Schedule 14D-9 filed with the SEC June 18, 2004.
(30)	Previously filed as an exhibit to PeopleSoft’s Amendment No. 30 to Schedule 14D-9 filed with the SEC June 28, 2004.

10 EX-99.(A)(104) 2 f97751a9exv99wxayx104y.txt EXHIBIT (A)(104) Exhibit (a)(104) [STATE OF DELAWARE COURT OF CHANCERY SEAL] EFILED: JUN 24 2004 4:31 PM EDT FILING ID 3789626 EXHIBIT A IN THE COURT OF CHANCERY OF THE STATE OF DELAWARE IN AND FOR NEW CASTLE COUNTY ORACLE CORPORATION, a Delaware ) corporation, and PEPPER ACQUISITION ) CORP, a Delaware corporation, ) ) Plaintiffs, ) ) Civil Action No 20377 NC V ) ) PUBLIC VERSION PEOPLESOFT, INC, a Delaware corporation, ) REDACTED CRAIG A. CONWAY, DAVID A. ) DUFFIELD, ANEEL BHUSRI, STEVEN D. ) GOLDBY, A. GEORGE BATTLE, FRANK ) J. FANZILLI and CYRIL J. YANSOUNI, ) ) Defendants. ) AMENDED COMPLAINT FOR DECLARATORY AND INJUNCTIVE RELIEF Plaintiffs Oracle Corporation ("Oracle") and Pepper Acquisition Corp ("Purchaser"), a wholly-owned subsidiary of Oracle, for their complaint against defendants PeopleSoft, Inc. ("PeopleSoft"), Craig A. Conway, David A. Duffield, Aneel Bhusri, Steven D. Goldby, A. George Battle, Frank J. Fanzilli, Cyril J. Yansouni and Michael J. Maples ("Director Defendants") allege, upon knowledge as to themselves and their own acts and upon information and belief as to all other matters, as follows: INTRODUCTION 1. Oracle launched an all-cash, all-shares tender offer for shares of PeopleSoft stock (the "Tender Offer") more than a year ago, on June 9, 2003. The Tender Offer is non-coercive and represents a healthy premium: Oracle's most recent $21 per share bid for PeopleSoft stock, announced on May 14, 2004, represented a 21% premium over that day's close of $17.30 per share. On information and belief, the small increases in PeopleSoft's stock price (though still well below the offer price) since the beginning of the trial in the related case of United States v. Oracle have been in whole or in part based on speculation that Oracle's tender offer will succeed. 2. In the past year, PeopleSoft's board of directors (the "Board" or "PeopleSoft's Board") has had ample time to pursue alternatives to Oracle's Tender Offer, and PeopleSoft stockholders have been inundated with information and analysis of the Tender Offer from both PeopleSoft and Oracle, as well as from industry and financial analysts, the media and other interested parties. In light of these facts, there is no legitimate justification for the Board's continued refusal to redeem PeopleSoft's poison pill, exempt the Tender Offer from Section 203 of the Delaware General Corporation Law ("Section 203") and permit PeopleSoft's stockholders to consider Oracle's Tender Offer. Nor can the Board justify its decision to continue the other defensive measures it has erected, including most notably the customer give-away program PeopleSoft termed the "Customer Assurance Program" (the "CAP"), which likewise are intended to block Oracle from completing the Tender Offer. 3. The Board's refusal to redeem its poison pill or discontinue the other defensive measures is part of a scheme of entrenchment, intended to defeat Oracle's bid at any price, that began just hours after Oracle announced the Tender Offer and continues to the present. On June 7, 2003, less than a day after Oracle announced the Tender Offer, PeopleSoft's President, CEO and director Craig Conway was quoted as having stated that he "could imagine no price or combination of price and other conditions to recommend accepting the offer to our stockholders" Motivated by their desire to retain their jobs in the face of an unsolicited offer, -2- Conway and other PeopleSoft senior executives committed themselves to a course of action that has as its goal defeating Oracle's Tender Offer at whatever cost. 4. PeopleSoft's Board nonetheless abdicated to PeopleSoft's management the Board's responsibility to oversee the business and affairs of PeopleSoft during this critical time, and gave management free reign to use whatever means they considered necessary to thwart the Tender Offer. Taking advantage of the Board's lack of independence and oversight, management implemented, without proper Board approval or supervision, the CAP, a novel and improper defensive measure designed to block Oracle and the Tender Offer. To date, more than $2 billion in potential cash payments has been promised to customers under the CAP. 5. None of the first three versions of the CAP - which collectively account for approximately $900 million of potential payments to customers - was approved by the Board prior to their use by PeopleSoft's management. The Board also has failed to evaluate adequately the unsupported assertions of PeopleSoft's management that "customer concern" justifies continuation of the CAP. 6. In addition to its other faults, the CAP is disproportionate to any perceived legitimate threat to PeopleSoft. The Board also failed to consider alternatives before permitting PeopleSoft's management to implement the CAP, including, for example, using the source code escrow provisions already being used to "protect" the customer, offering the CAP term on a negotiated (as opposed to across-the-board) basis or creating a multiplier scale that would approximate anticipated customer harm. -3- 7. In addition, the Board has approved and implemented an array of defensive measures over the past year that, taken together, are disproportionate to any threat that Peoplesoft may claim is posed by Oracle's non-coercive, all-cash, all-shares Tender Offer. The Board's illegal defensive measures include the CAP described above, which was not justified by the level or nature of customer concerns then being expressed and is a disproportionate reaction to them; amending the PeopleSoft / JD Edwards merger agreement to deny PeopleSoft stockholders a vote on that transaction; extending the director nomination notice provision in the company's bylaws, and then manipulating the meeting date so as to remove the purported benefits of that extended notice period, all with the purpose of thwarting a proxy contest; and launching a $550 million stock repurchase program under which, the company's financial advisors exhorted, [REDACTED] The Board's illegal defensive measures were facilitated by a practice of too-little-too-late disclosures that prevented PeopleSoft stockholders from challenging any of these measures until they were a fait-accompli, and/or denied stockholders relevant and timely information to evaluate management's acts. 8. The Board's refusal to redeem its poison pill and exempt the deal from Section 203, and its use of the CAP and other defensive measures, are only the latest moves in a series of actions the Board has taken over the past year to entrench itself and defeat Oracle's Tender Offer at any cost. Because the Board continues to breach its fiduciary duties by refusing to allow PeopleSoft stockholders to decide for themselves whether Oracle's Tender Offer is in their best interests, the Court should grant Oracle the relief sought in this complaint. -4- THE PARTIES 9. Plaintiff Oracle is a Delaware corporation with its principal executive offices in Redwood City, California. Oracle develops, manufactures, markets and distributes computer software that helps its customers manage and grow their businesses and manage their operations. Most of Oracle's software products can be categorized into two broad areas: database technology software and applications software. Oracle's database software products are highly sophisticated information management programs that allow users to store, retrieve, manage and use their business data. Oracle's applications software can be accessed with standard web browsers and can be used to automate business processes and to provide business intelligence for a company's marketing, sales, order management, procurement, supply chain, manufacturing, service and human resources functions. In addition to computer software products, Oracle offers a range of consulting, education, support and other services. For customers who choose not to install its applications on their own hardware, Oracle also offers outsourcing services that permit web browser access to Oracle's database technology and software applications hosted at sites that Oracle operates or manages. 10. Oracle is both a holder of record and a beneficial owner of PeopleSoft common stock. 11. Plaintiff Purchaser is a newly incorporated Delaware corporation and a wholly owned subsidiary of Oracle with its principal executive offices in Redwood City, California. To date, Purchaser has engaged in no activities other than those incident to its formation and the commencement of the Tender Offer. -5- 12. Defendant PeopleSoft is a Delaware corporation with its principal executive offices in Pleasanton, California. According to its most recent Form 10-K, PeopleSoft "designs, develops, markets and supports enterprise application software products for use throughout large and medium sized organizations worldwide. [PeopleSoft] provide[s] enterprise application software for customer relationship management, human capital management, financial management and supply chain management, each with a range of industry-specific features and functions. Within each of [PeopleSoft's] application suites, [PeopleSoft] offers] embedded analytics and portal applications. In addition, [PeopleSoft] offer[s] a suite of products for application integration and analytic capability, including portal applications, an integration broker and enterprise warehouse products". 13. Defendant Conway is the President, Chief Executive Officer ("CEO") and a director of PeopleSoft. According to PeopleSoft's Schedule 14D-9 filed June 12, 2003, prior to June 6, 2003, the date Oracle announced its intent to commence a Tender Offer, the Compensation Committee of the Board (the "Compensation Committee") approved amendments to Conway's employment agreement to increase the severance payments to two years' base salary plus target bonus and to provide for the following accelerated vesting of options and restricted stock: (i) immediate vesting of all unvested options and restricted stock upon a change of control of PeopleSoft and (ii) two years' service credit upon termination of employment without cause or voluntary termination with good reason. 14. On August 5, 2003, the Compensation Committee granted Defendant Conway the option to purchase 375,000 shares at an exercise price of $16.56, which fully vest upon a change -6- of control. On August 20, 2003, PeopleSoft again revisited the compensation arrangements for the company's senior executives, including Defendant Conway, and amended their severance arrangements. As explained in PeopleSoft's February 3, 2004 14D-9 with respect to Defendant Conway, if "the vesting of Mr Conway's restricted stock cannot be accelerated, PeopleSoft will make a cash payment equal to the fair market value of the unvested restricted stock owned by Mr. Conway on the date of termination of employment or on the date of the change of control, as applicable." Thus, the Compensation Committee approved lucrative increases in compensation arrangements for Defendant Conway and other PeopleSoft senior executives, including potentially substantial cash payments as a windfall in the event Oracle's Tender Offer is successful, all at a cost to be borne by PeopleSoft stockholders.' 15. Defendants David A. Duffield, Aneel Bhusti, Steven D. Goldby, A. George Battle, Frank J. Fanzilli, Cyril J. Yansouni and Michael J. Maples are members of PeopleSoft's Board. The Director Defendants owe the highest fiduciary duties of loyalty, care and good faith to PeopleSoft's stockholders. FACTUAL BACKGROUND 16. Since the Tender Offer was launched, PeopleSoft has implemented various defensive measures and taken many other actions intended to deter Oracle's bid. On June 9, 2003, three days after the Tender Offer was announced, PeopleSoft launched the CAP, promising customers between two and five times their money back if PeopleSoft were acquired and the acquirer took certain actions thereafter. Later that same week, Defendant Conway and PeopleSoft's other managers began an aggressive campaign to pressure customers into -7- contacting state and federal antitrust regulators, despite the fact that PeopleSoft itself had proposed a combination of the two companies - to be managed by PeopleSoft executives - just a year earlier. Customers were asked to send regulators messages pre-scripted by PeopleSoft, creating the artificial impression of unified customer competitive concerns. 17. On June 12, 2003, over a week before any response was required under SEC rules, PeopleSoft rejected Oracle's bid on grounds that it was likely to run into the antitrust problems they were actively fomenting, and that it undervalued the company. On June 16, 2003, PeopleSoft amended the PeopleSoft / JD Edwards merger agreement to eliminate the need for PeopleSoft stockholders to vote on the transaction, at a cost to PeopleSoft of an additional $875 million in cash. Two days later, Oracle increased its bid to $19.50 per share, again requesting a meeting with PeopleSoft's Board to discuss the offer. PeopleSoft's Board rejected the $19.50 bid the very next day. PeopleSoft closed its merger with J D Edwards a month later, on July 18, 2003. 18. In August 2003, PeopleSoft expanded its defensive efforts to its executive compensation arrangements. On August 5, 2003, Defendant Conway was granted 375,000 stock options that would vest immediately upon a change of control, at an exercise price of $16.56 per share - $3.00 below the Oracle bid they had rejected six weeks earlier. On August 20, 2003 the Board amended severance arrangements for Defendant Conway and other PeopleSoft senior executives to provide that they would be entitled to cash payments equal to the value of restricted stock options vesting upon a change of control. -8- 19. Also in August 2003, PeopleSoft announced in its 10-Q for the period ending June 30, 2003 that liabilities under the CAP totaled $391 million for the first month of the program alone, and that the CAP had been extended "indefinitely." Meanwhile, PeopleSoft's management was secretly developing a new and enhanced CAP (the "Version 2 CAP") designed to be more onerous for an acquirer, and that could be triggered by a successful proxy contest, not just a change of control. PeopleSoft's management rolled out the Version 2 CAP to customers on September 12, 2003, although it did not publicly announce it until October 27, 2003 - ten days after the program had expired. Realizing that inclusion of the proxy contest trigger jeopardized their ability to recognize revenues under the Version 2 CAP contracts, on September 30, 2003 PeopleSoft rolled out a third version of the CAP (the "Version 3 CAP"), which specifies that a successful proxy must be followed by a change of control, but is otherwise identical to the Version 2 CAP. PeopleSoft's Board was not given the terms of either the Version 2 or Version 3 CAP until after they expired. 20. On September 4, 2003, the Board initiated a $350 million stock repurchase program, having been told by their financial advisors that they could "repurchase shares first from those most likely to tender." On November 5, 2003, the Board amended PeopleSoft's bylaws to extend the minimum notice period for stockholder nominations to the Board from 20 days to at least 120 days before the anniversary of the prior year's stockholder meeting, on grounds that stockholders needed the maximum possible time to consider any insurgent slate. On January 30, 2004, two weeks after Oracle met the accelerated deadline, PeopleSoft -9- announced that the 2004 annual stockholders meeting would be held on March 25, 2003 - two full months earlier than in any of the previous seven years. 21. On November 17, 2003, the Board responded to preliminary injunction motions filed by Oracle and the Shareholder Plaintiffs challenging the Version 2 and 3 CAPs by implementing the Version 4 CAP. By March 31, 2003, total liabilities under the Version 4 CAP would reach $1.084 billion. By that date, $1.9 billion in potential customer payments had been promised under the four CAP terms. This represents approximately one third of PeopleSoft's market capitalization, and four times more than the company's current cash on hand as of March 31, 2004. 22. On January 12, 2004, PeopleSoft announced that the Board had extended the stock repurchase program, authorizing PeopleSoft to purchase an additional $200 million worth of PeopleSoft stock, even though the Board's financial advisors had raised a concern, when the first Stock Repurchase Program was initiated in September, that a "[l]arger repurchase would reduce cash balance and flexibility going forward". 23. On February 4, 2004, Oracle announced that it had increased its bid to $26 per share, representing a 19% premium over the previous day's closing price of $21.89. PeopleSoft rejected this bid on February 7, 2004. Over the course of the next several weeks, PeopleSoft's stock dropped to $16.53 per share, responding to the company's poor financial results. Still, PeopleSoft's Board refused to meet with Oracle to discuss the bid, even though the bid at one point represented a 57% premium to PeopleSoft stockholders. On May 14, 2004, in response to changes in market conditions and in PeopleSoft's market valuation, Oracle revised its bid to $21, -10- reflecting a 21% premium over that day's closing price of $17.30. On May 26, 2004, the Board rejected this bid as well and persisted in its refusal to negotiate with Oracle, or to redeem its structural defensive measures. PeopleSoft's poison pill and the application of DGCL Section 203 to the Tender Offer. I. PEOPLESOFT'S RIGHTS PLAN NO LONGER SERVES A LEGITIMATE PURPOSE AND SHOULD BE REDEEMED. A. THE BOARD FAILED TO MAKE A REASONABLE AND GOOD FAITH INVESTIGATION OF ORACLE'S OFFER. 24. On June 6, 2003, the day that Oracle announced its intention to launch the Tender Offer, Defendant Conway categorically rejected it, sight unseen. Two days later, the Board established a Transaction Committee to review the Tender Offer. 25. The Transaction Committee, however, has authority only to review and make non-binding recommendations to the Board concerning the Tender Offer. More importantly, the Transaction Committee has only received advice of legal counsel. and investment bankers retained by and beholden to PeopleSoft's management. The Transaction Committee has no truly independent legal or financial advisors. 26. In addition, the Transaction Committee has never met independently to consider Oracle's Tender Offer. Rather, minutes of PeopleSoft's Board meetings indicate [REDACTED] This includes Defendant Duffield, PeopleSoft's founder and a current director, who is not independent from PeopleSoft, Conway or the rest of -11- PeopleSoft's management. Furthermore, upon information and belief, Defendant Duffield's son, daughter-in-law and brother are or were employed by PeopleSoft. 27. The Transaction Committee itself also is critically flawed because one or more of the directors serving on the Committee is interested in the outcome of the Tender Offer. For example, Transaction Committee member Michael J. Maples is a former Executive Vice President of Microsoft Corporation, and upon information and belief, he continues to serve as a paid Microsoft consultant. Microsoft is a fierce competitor, and has an active interest in the outcome of Oracle's Tender Offer. Indeed, Microsoft announced on June 7, 2004 that until just a few months ago, it was engaged in merger talks with SAP, another major player in this highly competitive business. 28. Consistent with Conway's categorical rejection of the offer, and the Board's and the Transaction Committee's lack of independence and refusal to consider fairly the Tender Offer, at no time has PeopleSoft's Board agreed to meet with Oracle to discuss its offer, even when Oracle's bid was as high as $9.50 (or a 57% premium) above PeopleSoft's trading price. Over the past year, PeopleSoft's Board has rejected four separate bids from Oracle, including Oracle's current bid of $21, each time claiming the offer was too low. 29. Yet, Oracle's $21 per share bid represented a 21% premium above PeopleSoft's stock price. Presentations made to PeopleSoft's Board by its own financial advisors in connection with prior bids from Oracle reveal that [REDACTED] PeopleSoft's bid for J.D. Edwards involved a 19.4% premium. -12- 30. The Board's continued refusal either to negotiate with Oracle, or to allow PeopleSoft stockholders to consider for themselves a non-coercive, all-cash, all-shares offer at a substantial premium, demonstrates the Board's failure to make a reasonable and good faith investigation of the Tender Offer. The Board has not shown and can not show that the Tender Offer poses a threat to PeopleSoft's corporate policy or effectiveness sufficient to deny stockholders the opportunity to tender their shares and receive the benefits of the Tender Offer. B. THE PASSAGE OF TIME HAS REMOVED ANY LEGITIMATE PURPOSE FOR PEOPLESOFT'S POISON PILL. 31. In response to Oracle's offer, PeopleSoft's Board has refused to take down its preexisting structural defenses, including its Rights Plan and the application of Delaware's antitakeover statute ("Section 203"). Considering that a full year has passed since Oracle launched the Tender Offer, the premium the Tender Offer represents and other facts, the Board's refusal to redeem the poison pill and exempt the Tender Offer from Section 203 represents a breach of the Board's fiduciary duties to PeopleSoft's stockholders. 32. PeopleSoft's Rights Plan was adopted on or about February 15, 1995 and amended in 1997. The Rights are distributed and become exercisable for one one-thousandth share of PeopleSoft's Series A Participating Preferred Stock (the "Series A Preferred") at a price of $190 on the close of business ten days after the earlier of (i) it is announced that any person has acquired beneficial ownership of 20% or more of PeopleSoft's common stock (an "Acquiring Person"), or (ii) the publication of a tender or exchange offer which, if successful, would result in the beneficial acquisition by any person of 20% or more of PeopleSoft's common stock (the earlier of (i) and (ii) being referred to as the "Distribution Date"). As the Rights Plan -13- permits, on June 11, 2003 the Board postponed the Distribution Date, which otherwise would have been triggered by the Tender Offer, until just before the consummation of a transaction through which a person would become an Acquiring Person. The Rights expire on February 15, 2005, unless earlier redeemed or exchanged by PeopleSoft. 33. Under the Rights Plan, if and when Oracle becomes an Acquiring Person, all Rights other than those held by Oracle will "flip-in" and each right becomes exercisable for shares of PeopleSoft common stock equivalent in value to twice the exercise price of the Right. Thus, for the exercise price of $190, the holder of a Right may purchase PeopleSoft common stock having a market value of $380. If and when PeopleSoft engages in a merger or a sale of 50% or more of its assets, as is contemplated under the Tender Offer, the Rights "flip-over" and become exercisable for shares of the acquirer's common stock at the same deep discount price of two for the price of one. 34. The PeopleSoft Board may redeem the Rights, at a redemption price of $.01 per Right, any time prior to the close of business on the earlier of (i) the tenth day following the date of public announcement of the fact that an Acquiring Person has become such, or (ii) February 15, 2005. 35. Oracle's acceptance (via Purchaser) of shares tendered pursuant to the Tender Offer will result in it becoming an Acquiring Person, which, in turn, will make the Rights exercisable for shares of PeopleSoft's common stock at a discount of 50% of their market value (rendering the Tender Offer economically infeasible for Oracle to accomplish) and will deprive -14- PeopleSoft's stockholders of the ability to tender their shares unless PeopleSoft's Board redeems the Rights or otherwise amends the Rights Plan to render it inapplicable to the Tender Offer. 36. The Board also has refused to exempt the Tender Offer from the application of Section 203. Section 203 would delay the Proposed Merger for at least three years. During this period, substantial benefits of the Proposed Acquisition would be forever lost, and any number of events could occur that would prevent the Proposed Merger altogether. 37. PeopleSoft's poison pill no longer serves a valid purpose. In the year since Oracle launched its Tender Offer, PeopleSoft, Oracle, industry and financial analysts, the media and other commentators have provided PeopleSoft's stockholders with considerable information and analysis concerning both the Tender Offer and PeopleSoft's current financial condition and future prospects. As a result, PeopleSoft's stockholders already have the information (much of which was reviewed by the SEC) they need to evaluate the Tender Offer, and have had more than a year to make that evaluation. 38. Likewise, the intervening twelve months since the Tender Offer was announced have given PeopleSoft's Board ample "breathing room" to explore alternatives, examine the merits of the Tender Offer and communicate their views to PeopleSoft's stockholders. To the extent PeopleSoft's stockholders still do not have enough information about PeopleSoft's prospects as a stand-alone company, it is because PeopleSoft's Board and management have failed to fulfill their obligation to provide a coherent statement of its expectations about the future value of the company, despite having had a full year to do so. -15- 39. The PeopleSoft Board's continued refusal to redeem PeopleSoft's poison pill or to exempt the Tender Offer from Section 203, both individually and in concert, are disproportionate reactions to Oracle's Tender Offer and therefore constitute improper defensive measures. Accordingly, the Board's continued refusal to remove PeopleSoft's structural defensive measures is a breach of its fiduciary duties to PeopleSoft's stockholders, and furthermore threatens to deny Oracle the benefits of the proposed combination. II. PEOPLESOFT'S CAP IS AN IMPROPER DEFENSIVE MEASURE. 40. On June 9, 2003, three days after Oracle announced its Tender Offer, PeopleSoft's management began sending form letters to customers announcing the first in what would become a series of customer rebate programs that PeopleSoft called the "customer assurance program" (the "CAP"). The CAP purports to provide customers refunds of between two and five times any licensing and maintenance fees paid up front under software licensing agreements signed while Oracle's offer is pending if (i) PeopleSoft is acquired within a certain amount of time after the contract date; and (ii) the acquirer takes (or fails to take) certain actions with respect to the products licensed by the customer pursuant to that contract. Because the higher CAP multipliers are reserved for larger deals, CAP liabilities accrue at a rate that is independent of increases in licensing revenues. Potential CAP liability for single contracts executed thus far by PeopleSoft range as high as nearly $80 million, although there is no limit to the total possible payment that can be given under a single contract. Unlike a traditional poison pill, payment promises under the CAP cannot be redeemed by the Board, and would affect all acquirers over a two year period, even in a "friendly" transaction. -16- 41. PeopleSoft waited to publicly announce the CAP until July 2, 2003, two days after it had initially expired, denying stockholders any opportunity to object until after a $354 million potential liability had accrued under the Version 1 CAP. When PeopleSoft filed its 10-Q for the period ending June 30, 2003 (on August 14, 2003), it revised that potential liability figure upwards to $391 million for the same period, and announced that the CAP had been extended indefinitely. The Version 1 CAP payments could only be triggered if PeopleSoft is acquired within one year of the contract date and then the acquirer takes one of the actions described above within two years of the contract's date. 42. On October 27, 2003, PeopleSoft announced that it had introduced a second CAP offer (the "Version 2 CAP"), this time 10 days after it had expired. According to PeopleSoft's Supplemental Response and Objections to the Joint Interrogatories, a total of $170 million has been promised under the Version 2 CAP. Launched by PeopleSoft's management on September 12, 2003, the Version 2 CAP contains several "enhancements" to the Version 1 CAP language. First, the time horizons were expanded from one and two years to two and four years, respectively. Second, the business events covered were revised in ways that substantially increased the probability that they would be triggered, even if only unintentionally, by an acquirer. For example, a customer could argue that these terms were triggered if an acquirer were to immaterially delay the timing of product updates. After this offer had officially expired but months before its actual use was finally discontinued (at the end of December 2003), one or more of PeopleSoft's directors acknowledged [REDACTED] -17- 43. Third, the definition of an "acquisition" was also changed in the Version 2 CAP to include the election of a majority of new directors, not approved by the incumbent PeopleSoft Board (the "Board Composition Trigger"). This expansion was problematic in two respects. First, the Board Composition Trigger raised the specter that the Version 2 CAP could affect stockholders' voting decisions in a contested proxy contest, as the future value of the company under a new slate of directors would be diminished by the risk-weighted value of the Version 2 CAP payments promised to customers. PeopleSoft did not publicly acknowledge the potential effect of the CAP on the stockholder franchise until a March 4, 2004 amendment to its Schedule 14D-9 filings. 44. Second, the Board Composition Trigger affects the revenue recognition treatment for revenues associated with the Version 2 CAP contracts. According to the company's own analysis, [REDACTED] However, under the Version 2 CAP definition of an "acquisition," CAP liabilities could be triggered upon a successful proxy contest for the company, regardless of whether there was a subsequent change of control. Thus, the revenues attributable to these contracts should have been deferred. Because PeopleSoft chose to recognize - rather than defer - this revenue, PeopleSoft's third and fourth quarter 2003 revenues have been inflated by the amount of the Version 2 CAP contract revenues. PeopleSoft added a proviso to the Board Composition Trigger for Version 2 CAP contracts dated after September 30, 2003, providing that the company must -18- be acquired following a successful proxy contest for the term to take effect, calling this a new, Version 3 CAP. 45. On October 10, 2003 - 10 days after the close of the quarter in which the licensing revenues under those contracts was booked - the company began sending letters to customers whose contracts contained the Version 2 CAP, attempting to unilaterally amend the contracts after the fact to include the new provision [REDACTED] At a minimum, PeopleSoft should have deferred these revenues unless and until each customer accepted the unilateral amendment. 46. Despite PeopleSoft's repeated assertions to the contrary, the company did not seek SEC pre-clearance for their accounting treatment of the Version 2 CAP. The SEC's June review of the CAP was expressly limited to the particular language of the Version 1 CAP as provided to the SEC (which did not include substantial variations included in over 250 contracts). PeopleSoft first discussed the accounting treatment for their Version 2 CAP months after those revenues bad been booked by PeopleSoft and only after the SEC inquired about their accounting treatment. This process eventually resulted in PeopleSoft revising its disclosures about the Version 2 CAP. 47. In total, management signed contracts providing for a total of $170 million of potential liability under the problematic Version 2 CAP, with an additional $33 million worth of potential liability under contracts containing the Version 3 CAP term -19- 48. On November 5 and 10, 2003, respectively, the stockholder plaintiffs and Oracle filed motions for a preliminary injunction against continued use of the Version 2 CAP One week later, PeopleSoft implemented a new, Version 4 CAP. The Version 4 CAP retains the extended time horizons and many of the enhanced business provisions of the Version 2 CAP. 49. PeopleSoft announced the Version 4 CAP just hours before the November 19, 2003 Delaware Court hearing on Oracle's motion challenging the CAP; as of March 31, 2003 it had promised customers potential payments totaling $1.084 billion under this term A. THE BOARD DID NOT MAKE A GOOD FAITH AND REASONABLE INVESTIGATION OF THE TENDER OFFER'S PURPORTED THREAT TO CUSTOMERS. 50. PeopleSoft's Board did not make a reasonable and good faith investigation of the claims made by PeopleSoft's management that the Tender Offer posed a threat to PeopleSoft's customer base [REDACTED] 51. PeopleSoft's documents from that period reflect a campaign to generate rather than mitigate fear among its customer base, in an effort to mobilize customers against Oracle's offer [REDACTED] PeopleSoft's management spread the myth that Oracle would not support PeopleSoft products, when in fact Oracle had stated, in its initial June 6, 2003 conference call announcing the offer and subsequently, that it would enhance and continue support for the PeopleSoft products. -20- 52. PeopleSoft has not identified a single customer who proposed or demanded that any particular language, whether linked to a change of control or otherwise, be added to their contracts prior to PeopleSoft extending the CAP offer. PeopleSoft CFO Kevin Parker, who claims responsibility for the original idea to add a CAP clause to customer contracts, has testified that [REDACTED] On June 9, 2003, of its own accord, PeopleSoft made the first of several mass mailings of offer letters to customers describing the Version 1 CAP. 53. By the time the Board first made a serious review of the CAP, in the end of October and early November 2003, PeopleSoft had already reported two quarters of purportedly strong financial performance, meeting or exceeding analyst expectations, while the Tender Offer was pending. On September 4, 2003 Conway announced at the company's Analyst Day conference, "I don't think the Oracle bid is a current issue. It's a movie that's been playing a long time. I think people have lost interest in it" He added, "[t]he last remaining customers whose business decisions were being delayed have actually completed their sales and completed their orders." Recognizing that this statement undermined PeopleSoft's arguments before this Court (as well as in a pending action brought by PeopleSoft in California state court), PeopleSoft replaced the above statement in the version of this transcript that PeopleSoft filed with the SEC with the following. "Oracle's tactics have created concern among many users, and that's a problem for us. Fortunately, we've been able to overcome much of it and we expect that we will continue to be able to do so." Remarkably, PeopleSoft's public filing notes that this language -21- "ha[s] been corrected to reflect the intended meaning of the speaker " (emphasis added). Two weeks later, Conway said that "[t]he actual impact [of the Tender Offer] has been remarkably beneficial for PeopleSoft." Yet, despite these indications, on November 17, 2003, the Board approved a whole new CAP offer, under which $1.084 billion in customer payments had been awarded by March 31, 2004. B. THE CAP WAS IMPLEMENTED TO HELP REPEL ORACLE'S OFFER. 54. [REDACTED] 55. The CAP serves this purpose in two ways. First, as PeopleSoft asserted to the SEC in its June 16, 2003 letter seeking pre-clearance for their intended accounting treatment of the CAP terms, the CAP "acts as an economic disincentive in the event of a hostile takeover" Others of PeopleSoft's internal documents are more blunt. [REDACTED] -22- 56. Other evidence of the CAP'S true purpose includes the fact that the CAP, as a matter of policy, has been given to all PeopleSoft customers signing new licensing deals since the Tender Offer's initiation, whether or not such extraordinary terms were commercially reasonable and necessary (or even requested by customers). Twenty percent of the total CAP liability as of March 31, 2004 is attributable to just eight customers, whose individual payments under the CAP range from $37 million to almost $80 million. Notably, not one of these customers was identified in PeopleSoft's Supplemental Response and Objections to the Joint Interrogatories as having advised PeopleSoft "that inclusion of the [CAP] was essential in order for them to purchase PeopleSoft products." 57. Second, PeopleSoft's management has used the CAP to aid in their efforts to defeat the Tender Offer by artificially increasing the company's reported revenues in the critical quarters after the Tender Offer was first announced, making Oracle's bid look artificially less attractive to stockholders. 58. PeopleSoft's internal documents and communications reveal that [REDACTED] Customers were exhorted to make additional, unplanned or accelerated software purchases, as a way to "protect" their previous software investments. The CAP facilitated this effort on behalf of PeopleSoft's management, as it created an incentive for customers to sign up while Oracle's offer was pending (which was initially anticipated to be for only a limited period of time). Conway stated both publicly and privately -23- that some customers did in fact "accelerate" their purchases, to "help" PeopleSoft in its efforts to repel Oracle's offer. 59. All the while, PeopleSoft's management assured industry analysts that the CAP was having no effect on the company's pipeline of future deals, despite these indications that the CAP was used as an incentive for customers to move forward planned purchases. In addition, PeopleSoft has inflated its revenues in the third and fourth quarters of 2003 by recognizing up front the licensing fees attributable to contracts containing the Version 2 CAP term, as described above. C. THE CAP IS DISPROPORTIONATE TO ANY PURPORTED CUSTOMER THREAT. 60. PeopleSoft's Board also failed to consider whether customers' purported concerns regarding the Tender Offer could be satisfied through means other than the particular CAP term proposed by PeopleSoft's management, or whether the CAP would be proportional to any legitimate threat they believed Oracle's Tender Offer posed. 61. First, the Board did not investigate whether existing contractual provisions already provided adequate protection to customers. For example, many of PeopleSoft's contracts include source code escrow provisions, under which the customer is entitled to access the source code so that it can make other arrangements in the event that PeopleSoft (or its successor) announces its inability or unwillingness to continue to support the products. 62. Second, the Board did not consider options for protecting customers against changes to the support, development or marketing of the licensed products regardless of whether there was a change of control. PeopleSoft claims in its Response and Objections to the Joint -24- Interrogatories that the CAP was limited to take effect only after a change of control because of requirements under the revenue recognition rules applicable to software. In fact, on one or more occasions PeopleSoft gave a customer protections that were not dependent on a change of control and yet, upon information and belief, still accounted for the revenue in the quarter that the contract was signed. 63. Third, having decided to allow PeopleSoft's management to proceed with the CAP, at no time did the Board provide any guidance to them regarding, or place any ceilings on, the extent of promises that it would be reasonable to make under the CAP. It was not until November 17, 2003 - once over $800 million in liability had accrued under two separate CAP offers and after Oracle filed its motion challenging the CAPs - that the Board first received an analysis from its financial advisors of the potential impact of the CAP on an acquirer, or of the per share cost of the program. Nor was such an analysis prepared for the Board's subsequent consideration of whether to extend the CAP through to the present. 64. Similarly, the Board did not fully consider whether the CAP could be used sparingly, in response to specific customer requests, rather than on an across-the-board basis. Software licensing agreements for enterprise software are heavily negotiated contracts, containing dozens of carefully crafted provisions that are modified to meet the demands of the customer. Upon information and belief, terms with a financial impact are subject to sales approval processes, including at the highest level CFO Kevin Parker, to ensure that the company's long term financial health is protected. Yet, as early as June 11, 2003, the Board decided that the CAP term - under which penalty payments in one or more individual contracts -25- could exceed $80 million - would be given to all customers, whether or not they requested it and with no executive approval required. 65. Fourth, the Board failed to consider whether a CAP-like provision could be implemented in a way that would have a less severe potential impact on an acquirer and therefore on PeopleSoft stockholders. For example, Parker testified that, [REDACTED] At no time did the Board request or receive an analysis of the correlation between the actual harm a customer might suffer should one of the acts covered by the CAP occur and the size of the customer's licensing deal. [REDACTED] 66. The CAP is a disproportionate response to the threat purportedly posed by Oracle's offer because the CAP gives customers a remedy that far exceeds any actual harm they may suffer, and creates obligations for an acquirer that PeopleSoft - under current management - does not have. 67. The CAP terms, or "penalty provisions" as they were referred to in an early draft of the accounting pre-clearance letter PeopleSoft sent to the SEC on June 16, 2003, were not -26- tailored to correlate to any actual harm that customers would suffer. For example, the size of the CAP payments depends only on the size of the licensing deal of which it was a part, not on the severity of the triggering event's purported impact on the customer's business. Thus, a customer's CAP payment would be the same in the event of an immaterial delay in the timing of a product update's release as in the event of a permanent end to all product support services. 68. Through the CAP, PeopleSoft also gave customers a remedy that was substantially more generous than the limitations of liability terms the company had previously negotiated with these very same customers and thus reflect the customer's actual commercial demands. These terms govern in the event of any damages arising under these same contracts. Even among the eight customers whose CAP promises collectively represent twenty percent of the total potential CAP liability - all of whom received multipliers of "5x" - PeopleSoft's liability (in the absence of a change of control) is limited under the associated master Software License and Services Agreement to which the CAP-enhanced licensing schedule is an attachment, to one or two (or, in one case, three) times the total fees attributable to the particular product or service with respect to which the claim arises. 69. The CAP terms also create substantial obligations for an acquirer that PeopleSoft currently does not have and has specifically refused to take on [REDACTED] -27- [REDACTED] 70. Moreover, the CAP terms purport to penalize an acquirer for making business decisions that PeopleSoft, under current management, is free to make. For example, Parker stated in a June 2, 2004 deposition that [REDACTED] 71. When it decided, on November 17, 2003, to implement the current, Version 4 CAP, the Board relied on assertions by PeopleSoft Executive Vice President for Products & Technology Ram Gupta that the CAP terms place no more obligations on an acquirer than those PeopleSoft has itself and therefore are, by implication, reasonable. The Board was not entitled to rely on these assertions. PeopleSoft's Board consists of savvy individuals with substantial experience in the software industry and, in some cases, company insiders with substantial knowledge of PeopleSoft itself. Minutes of PeopleSoft Board meetings cite this experience as a -28- basis for key decisions made by the Board with respect to the Tender Offer. The Board therefore cannot reasonably rely on unsupportable and inaccurate assertions by PeopleSoft's management to justify the CAP. D. THE BOARD NEVER ESTABLISHED APPROPRIATE MONITORING AND REPORTING CONTROLS FOR THE CAP. 72. PeopleSoft's Board failed to provide appropriate approvals or supervision over management's activities with respect to the CAP. The Board's inaction amounted to an abdication of its responsibility to manage the business and affairs of the company particularly in light of the Tender Offer, to management they knew to be adamantly opposed to Oracle's offer. This calls into serious question whether the Board was in fact acting in good faith with respect to the Tender Offer more generally. 73. Parker has testified in deposition that [REDACTED] However, none of the Board meeting minutes produced to date include any Board approval for the Version 1 CAP, which had been first extended to customers two days previously. Moreover, the Board never reviewed a copy of the actual language of the Version 1 CAP term. The only indication that the Board was aware of the language being provided to customers is an incomplete summary of the offer's terms that was sent to the Board on June 18, 2003, in the form of what Parker testified in deposition, as PeopleSoft's corporate representative, [REDACTED] There is no record of PeopleSoft's Board having met to consider extending or reinstating the program at any time during the summer of 2003. By the time management abandoned the Version 1 CAP on September 30, 2003, the total payments under this program had mounted to $670 million. -29- 74. Likewise, there is no indication that either the Audit Committee or the company's external auditors reviewed the accounting treatment for the Version 2 CAP, including management's proposal to send "clarifying" letters to customers, at any time, despite the ramifications of the Board Composition Trigger - and management's attempted after-the-fact "fix" of the Version 2 CAP terms - for the company's financial statements and critical accounting policies. 75. Despite the substantial and problematic changes introduced with the Version 2 and 3 CAPs, neither term was ever approved by PeopleSoft's Board. PeopleSoft contends that these terms were approved at an August 5, 2003 Board meeting. However, the minutes of that meeting state only that [REDACTED] 76. The Board appears to have been unaware that a new CAP term had been implemented. Following the August 5, 2003 Board meeting at which "enhancements" to the CAP were being considered, the next reference to the CAP in any Board minutes is not until the October 21, 2003 Audit Committee meeting. At that meeting, Parker informed the Board that -30- [REDACTED] What followed was a flurry of management activity, including efforts to prepare a presentation on the CAP for the first time. On October 29, 2003, two days after the Version 2 CAP was publicly announced for the first time in a 14D-9 filing, PeopleSoft General Counsel Anne Jordan emailed the Board members to request a special meeting to review and approve the CAP so that pending deals could be closed. For the first time, the language of the Version 2 CAP term was attached, along with a comparison with the language of the Version 1 CAP. 77. Minutes of the October 31, 2003 Board meeting indicate that the Board approved [REDACTED] This is the first reference in the Board minutes to actual Board approval of a CAP term. Documents concerning this meeting indicate that [REDACTED] Thus, of the $880 million of potential payments promised to customers prior to Oracle and the stockholder plaintiffs filing their motions for preliminary injunction in the Delaware Chancery Court, only $7 million was promised pursuant to terms that had actually been approved by PeopleSoft's Board. -31- 78. On November 5, 2003, the Board again met to consider the CAP and for the first time received draft language for revising the CAP. That day, the stockholder plaintiffs in an action pending against PeopleSoft in this Court filed a motion for a preliminary injunction against the Version 2 CAP, which was followed by a similar motion from Oracle on November 10, 2003. In response, PeopleSoft's Board met November 17, 2003 and developed a new, Version 4 CAP term. For the first time, the Board received analyses by the company's financial advisors of the potential impact the offer would have on an acquirer. Also for the first time, the November 17 Board minutes indicate that [REDACTED] This record stands in sharp contrast to the elusive references PeopleSoft relies upon for its assertion that the versions of the CAP which predated Oracle's preliminary injunction motion were likewise "approved" by the Board prior to their usage. 79. As with the development of the Version 2 CAP, however, the company's independent auditors were not included in deliberations regarding the Version 4 CAP term. Rather, the auditors were only given a few hours to review the new term, after the Board had finished its deliberations. [REDACTED] -32- [REDACTED] 80. The Board likewise failed to establish appropriate monitoring and controls over the CAP's implementation. By June 18, 2003, the Board should at least have been aware of the broad brush strokes of the CAP, including that the multiplier scale would cause liability under the CAP to mount rapidly, outstripping growth in the associated licensing fees. Accordingly, the Board should have established strict monitoring, reporting and controls on the CAP's implementation and extension. 81. However, in the year that the CAP has run, upon information and belief, the Board never set a limit on the total liability that would be acceptable under one or all CAP contracts, or even requested regular reporting from PeopleSoft's management on the potential liability accruing under the CAP. In fact, PeopleSoft CFO Kevin Parker has testified that [REDACTED] -33- 82. PeopleSoft's management responded by introducing new language to the CAP terms as they were provided to hundreds of customers, without Board review or approval. The most obvious example of this was the introduction of the Version 2 CAP, discussed above. Before that, however, PeopleSoft's management had added language to the Version 1 CAP term as it appears in literally hundreds of contracts that purports to bind a successor not only to continue licensing PeopleSoft's current line of products to new customers, but also "the successors to [the licensed products that] would have been developed by PeopleSoft prior to an acquisition or merger." This new language was not included in the summary of the Version 1 CAP term, in the form of a "draft transmittal letter to customers" that was sent to PeopleSoft's Board on June 18, 2003. By June 18, PeopleSoft had been giving customers the new language for at least a week. 83. PeopleSoft's management also liberally gave individual customers exceptions under the CAP, without regard to the potential impact of those exceptions on an acquirer or stockholders, and without Board oversight. [REDACTED] -34- [REDACTED] 84. On one or more other occasions, PeopleSoft's management even agreed to back date CAP contracts, in some cases to before the date Oracle's Tender Offer was announced. One or more customers received a CAP in a contract that was back-dated across the quarter line, which, if recognized as revenue in the quarter prior to the actual contract date, would be impermissible under the applicable accounting standards. 85. Meanwhile, PeopleSoft's management misled regulators and the investing public regarding the nature and effect of the CAP on one or more occasions with the Board's acquiescence. [REDACTED] Just two weeks earlier, on June 16, 2003, in a letter to the SEC office of the chief accountant seeking pre-clearance for the revenue recognition treatment of CAP contracts listed as one of two business purposes for the CAP that it "acts as an economic disincentive in the event of a hostile takeover" [REDACTED] 86. The Board's abdication to PeopleSoft's management of its responsibility to oversee the CAP reveals its lack of good faith in its response to the Tender Offer. -35- COUNT I (BREACH OF FIDUCIARY DUTY: DISLOYALTY AND ENTRENCHMENT) 87. Plaintiffs repeat and reallege each and every allegation set forth in paragraphs 1 through 86 as if fully set forth herein. 88. The Director Defendants owe PeopleSoft's stockholders, including Oracle, the highest duties of care, loyalty and good faith, and have failed from the outset to meet those duties. 89. The Tender Offer is non-coercive and non-discriminatory, it is fair to PeopleSoft stockholders, it poses no threat to PeopleSoft's corporate policy and effectiveness, and it represents a substantial premium over the market price of PeopleSoft's stock. 90. The Director Defendants have and are continuing to breach the fiduciary duties owed PeopleSoft's stockholders by refusing to accept the Tender Offer, or otherwise negotiate with respect to it, by adopting or allowing PeopleSoft's management to adopt the defensive measures detailed above, and by failing to remove the barriers of its Rights Plan and Section 203. Such efforts to frustrate the completion of the Tender Offer serve no legitimate purpose, and are not reasonable responses to the Proposed Acquisition, which poses no threat to the interests of PeopleSoft's stockholders or to PeopleSoft's corporate policy and effectiveness. As such, the actions of the Director Defendants are in breach of the fiduciary duties the Director Defendants owe to PeopleSoft's stockholders under applicable Delaware law. 91. The unlawful actions of the Director Defendants are preventing PeopleSoft's stockholders from receiving the benefits of the Proposed Acquisition and are thereby causing and will cause PeopleSoft's stockholders irreparable harm. -36- 92. The actions complained of herein further threaten to deprive Oracle of the opportunity to proceed with its Tender Offer and the Proposed Acquisition. The loss of these opportunities, which are unique, constitutes irreparable harm which can not adequately be remedied later. Unless PeopleSoft's Board is restrained by this Court, the substantial benefits of the Proposed Acquisition may be forever lost. 93. Plaintiffs have no adequate remedy at law. COUNT II (BREACH OF FIDUCIARY DUTY: FAILURE TO ACT WITH DUE CARE AND ABDICATION OF RESPONSIBILITIES) 94. Plaintiffs repeat and reallege each and every allegation set forth in paragraphs 1 through 93 as if fully set forth herein. 95. The Director Defendants prejudged and summarily rejected the Tender Offer without conducting a reasonable and good faith investigation of the merits of the Proposed Acquisition and any threat purportedly posed by it to corporate policy and effectiveness. The Director Defendants' failure to adequately consider the Tender Offer constituted gross negligence and an abandonment of their fiduciary obligations. 96. The Director Defendants adopted defensive measures that individually and collectively are neither reasonable nor proportionate to any threat posed by the Tender Offer and constitute an illegal scheme of entrenchment. 97. Furthermore, the Director Defendants allowed PeopleSoft's management to implement one or more improper defensive measures without proper Board approvals, monitoring or supervision. -37- 98. The Director Defendants' actions are in breach of the fiduciary duties owed to PeopleSoft's stockholders, including Oracle, under applicable Delaware law as Defendants' conduct is causing irreparable harm to Oracle. 99. Plaintiffs have no adequate remedy at law, as Defendants' conduct is causing irreparable harm to Oracle. COUNT III (VIOLATION OF SECTION 141 AND ULTRA VIRES) 100. Plaintiffs repeat and reallege each and every allegation set forth in paragraphs 1 through 99 as if fully set forth herein. 101. Under 8 Del C Section 141 ("Section 141"), the board of directors has a continuing duty to oversee, manage and direct the business and affairs of the corporation as circumstances evolve. By implementing the CAP Offers, the Director Defendants have impermissibly restricted their own and any future board's authority to manage PeopleSoft under Section 141 and the board's concomitant fiduciary duties. 102. Plaintiffs have no adequate remedy at law, as Defendants' conduct is causing irreparable harm to Oracle. PRAYER FOR RELIEF WHEREFORE, Plaintiffs respectfully request that this Court: (a) compel PeopleSoft and its Director Defendants to redeem the Rights or to render the Rights Plan inapplicable to the Proposed Acquisition; (b) compel the Director Defendants to approve the Proposed Acquisition for -38- purposes of Section 203 and enjoin them from taking any action to enforce or apply Section 203 that would impede, thwart, frustrate or interfere with the Proposed Acquisition; (c) enjoin PeopleSoft from entering into any further CAP contracts; extending, continuing or expanding the CAP; or making changes that are outside the ordinary course of business to the update or support practices that constitute the standards an acquirer would need to maintain in order to avoid triggering potential liability under the Versions 2, 3 and 4 CAPs; (d) enjoin PeopleSoft and the Director Defendants from adopting any measure that has the effect of impeding, thwarting, frustrating or interfering with the Tender Offer, or the Proposed Merger, including, but not limited to, amendments to the Rights Plan, amendments to PeopleSoft's bylaws, the initiation or extension of unreasonable rebate programs (such as the CAP Offers), stock repurchase programs, pursuit of alternative transactions with substantial break-up fees and/or lock-ups, "White Knight" stock issuances, changes to licensing agreements, or executive compensation arrangements with substantial payments triggered by a change in control; (e) declare that the CAP contracts entered into by PeopleSoft are unlawful and a breach of the fiduciary duties owed by the Director Defendants; (f) declare that the CAP contracts are coercive and impermissibly interfere with future directors' exercise of their fiduciary duties, (g) award Plaintiffs their costs and disbursements in this action, including reasonable attorneys' and experts' fees, and -39- (h) grant Plaintiffs such other and further relief as this Court may deem just and proper. /s/ Brock E. Czeschin -------------------------------- Allen M. Terrell (#709) Daniel A. Dreisbach (#2583) Brock E. Czeschin (#3938) Dawn N. Zubrick (#4327) Richards, Layton & Finger One Rodney Square P.O. Box 551 Wilmington, Delaware 19899 (302) 651-7700 Of Counsel: Attorneys for Plaintiffs Michael P. Carroll Joel M. Cohen Davis Polk & Wardwell 450 Lexington Avenue New York, New York 10017 Dated: June 18, 2004 -40- EX-99.(E)(23) 3 f97751a9exv99wxeyx23y.txt EXHIBIT (E)(23) Exhibit (e)(23) The following weblog installments have been posted to PeopleSoft's website at www.peoplesoft.com. These are the most recent in a series of weblog commentaries written by Gary Reback regarding the ongoing Oracle antitrust trial. These and the other weblog installments may be accessed without charge by clicking on "View from the Court" at PeopleSoft's website at www.peoplesoft.com. - -------------------------------------------------------------------------------- TRIAL DAY 11: JUNE 22, 2004 Today, Oracle began its case with the expert testimony of Tom Campbell, Dean of the Haas School of Business at U.C. Berkeley. In a trial, you usually start with your best and strongest, and everybody assumed Oracle would do the same. After all, Tom must have the world's best resume: Ph.D. in economics, law degree from Harvard, impressive government appointments and elective positions, Supreme Court clerkship, countless accolades and honors. But once the testimony started, it seemed to me that everything seemed to turn around. First, Dr. Campbell asserted that he was assuming the Government's market definition placed each procurement in a separate antitrust market. He was primed to explain how such an approach was inappropriate. The problem was, the Government specifically disclaimed such an approach during Professor Elzinga's testimony. Rather, Dr. Elzinga explained, and the Government's complaint clearly said, that all procurements of high function core HR software were in a single market, and all procurements of high function core financial software were in a second market. Two markets, not many markets. TRIAL DAY 11 The Judge agreed to hear the testimony of Dr. Campbell over the Government's objection of relevance (i.e., Campbell was arguing against something the Government was not even asserting), but the Judge came back to ask Campbell later in the testimony whether any of what he had said made any difference, given the Government's market definition. (Campbell maintained that there was still some relevance.) Then it turned out that Professor Campbell had not even done a study. Rather, he took a set of assumptions given him by Oracle's attorneys, and from those assumptions, he concluded that the Government's conclusions were incorrect. But some of Campbell's assumptions seemed dubious, and contradicted by the evidence in the case. For example, Campbell assumes that HR and financial management software are "durable goods" -- like the washers or dryers you buy at Sears -- that last 10 years. Software code doesn't wear out, Campbell reasoned. But on cross, the witness was confronted by all the testimony of customer witnesses saying they needed frequent upgrades to keep their software current. Campbell then maintained that customers might not need upgrades, particularly if they are in a "mature industry," in which case, to quote the witness, "the software may do just fine without upgrades." TRIAL DAY 11 Few in the courtroom could believe their ears. The Judge had to make sure he heard it right: The Court: "Well, Dean Campbell, we are talking here about financial management and human relation software, and even if the company's business does not change, the tax laws change. I don't have to tell you as a former member of Congress that Congress -- (laughter) -- is always tinkering with the tax laws, they're tinkering with the employment laws, and all of these changes have to be built in and incorporated in these systems. So an enterprise really has no choice but to maintain and upgrade the system going forward." The witness agreed, but then asserted that the upgrade could be done by someone other than the vendor. "Who?" you might ask. A best-of-breed provider, said the witness. Now the Judge was really puzzled. "Where is that in the record before the court?" the Judge asked. The witness claimed to have read that information in Forrester and Gartner reports. The Judge next asked: TRIAL DAY 11 "Well, doesn't this undermine your assumption that the product is durable?" "No" said the witness. Actually, he said it was not a question that could be answered "yes" or "no." Professor Campbell also assumes that customers had the option to "do nothing" -- that is not to buy new software if it was too expensive. In fact, Campbell claimed that "doing nothing" deserves its own market share. So that if you were representing the market shares of U.S. automakers, the results might read: "GM, 40 percent; Ford, 30 percent; Chrysler, 20 percent; doing nothing, 10 percent." Once again, the witness was confronted with customer testimony about how customer after customer was required to purchase a new system because it had outgrown its existing system or because the vendor of the existing system had gone out of business. As the witness from Verizon observed: "[T] the point is that you can't delay it [the purchase] much. You can't let these things simply quiesce and run your company the way they run it today. That's not an option." Confronted with this testimony, Professor Campbell readily agreed that the fact a company can make one delay "doesn't mean it can delay for much." TRIAL DAY 11 Professor Campbell actually made four additional assumptions, but constraints of time preclude me from describing the testimony on all these assumptions in similar detail. Suffice it to say that from these assumptions, Campbell concluded that "economic theory" was unable to predict whether prices would be higher or lower if there were three vendors, and two of them merged. The witness appeared to get the Judge's attention with that testimony: The Court: "[A]re there other situations in economics where less is better than more, which is what you seem to be saying?" The witness gave a long-winded answer to the effect that a bilateral negotiation might produce a better combination of functionalities than a situation in which multiple vendors compete. "Well, why is that?" the Judge wanted to know. Because negotiation is different than bidding replied the witness. More questions from the Judge, who wanted to know, if you are building a house, why you wouldn't get a better price negotiating with several contractors instead of just one. The witness answered this question, but even after rereading his answer in the transcript, I'm not sure what he meant. TRIAL DAY 11 But the Government attorney couldn't resist piling on. He confirmed the witness's testimony that "economic theory" cannot tell us whether having three vendor choices would produce better prices than just having two. Would that also be true, if there were only two vendor choices, going to one vendor choice post-acquisition? asked the lawyer. "Economic theory does not predict," replied the witness. All right, said the lawyer, suppose PeopleSoft, Oracle, and SAP were in a cartel. Would that result in higher prices? "No," said the witness. "It's indeterminate." Suppose, that Oracle, SAP and PeopleSoft were all going to merge into a single company, the Government lawyer asked. Would that result in higher prices? "Economic theory does not predict the outcome of bilateral monopoly," said the witness. Okay, the DOJ lawyer continued, you say that "economic theory" can't predict. But, he asked, do you have an opinion as to whether a merger of all three vendors in this market would result in higher prices? Professor Campbell said, "I'm saying I was not asked to, nor did I form such an opinion." The Judge noted the contrast between Campbell's testimony and that of the customer witnesses: TRIAL DAY 11 "The Court has heard testimony about a number of customers of high-function software that are very knowledgeable, very sophisticated individuals, and they have expressed concern about this merger. Now, given their obvious knowledge about the business, their business, their knowledge about the product involved, what should the court make of this testimony?" The witness replied that customers always complain about mergers, at least that was his experience when he held a high position at the Federal Trade Commission. So, in the end, the witness continued to refuse to agree with customers that a "three to two" merger will likely produce price increases. Here is what the leading FTC economists now say on the FTC website about "three to two" mergers: "It is probably in the mainstream of economics and antitrust that a merger in an industry with only three competitors in a well-defined market protected by barriers to entry is also likely to be problematic, absent convincing efficiencies or unusual facts. [A] merger of the two of the three significant competitors in TRIAL DAY 11 a well-defined market appears to have an almost non-rebutable presumption of being anticompetitive." As bad as it was for Oracle in the courtroom it was far worse out in the hall with the press. "Wasn't the Judge skeptical?" a reporter asked Dan Wall. "Why can't the witness predict higher prices if all suppliers merge?" another journalist wanted to know. Dan Wall's reply was to cite an editorial in the Wall Street Journal, arguing that the Government should not have brought this case. Then, a few words into a second or third answer, Wall just stopped, mid-sentence, and walked away from the press. Tomorrow the Government's Microsoft witness testifies. Stay tuned. [END] THE GOVERNMENT RESTS After submitting hundreds of pages of depositions, testimony and documents, the Government rested its case in chief following the testimony of its Microsoft witnesses on Wednesday and Thursday. So it's time to ask how the Government is doing in terms of proving its case. As I explained at the beginning of this journal, it is fair to use the Government's Pretrial Brief in which the DOJ laid out what it would try to prove, as a scorecard in this case. Here is my perspective. First, the Government said it would try to prove that there are a group of customers who have only Oracle, SAP and PeopleSoft as potential vendors for core HR and core Financial software. This point has been proven in spades. Customer after customer has taken the stand to say just that. And, underscoring the point, Government expert witnesses have analyzed and displayed in court internal Oracle discount records that reveal in many cases the key competition for the account is between PeopleSoft and Oracle. The Government has demonstrated through the testimony of customers and experts, and through Oracle's own documents, that this competition is especially vigorous and has saved customers millions of dollars. The best Oracle has been able to do is "pick around the edges," as a DOJ lawyer explained. Oracle has shown that large companies sometimes use outsourcing for a particular function, like payroll for example. And large companies sometimes use software from Lawson or other vendors to run a discrete division or function, but not for the enterprise platform. And customers sometimes use software of best-of-breed providers for particular functions. And, indeed, if the customer's IT department cannot get adequate funding, sometimes the deployments of enterprise HR or financial management platforms are delayed. The press has sometimes sensationalized testimony about these subjects, characterizing such testimony as "admissions." This is not the case, in my opinion. In fact, it is important to understand what Oracle has not been able to prove. No Government customer witness has agreed under cross-examination that it could just as well use software from some company other than the Big 3 for its HR or financial platform. They have all rejected this key Oracle assertion and they have all maintained that they would be harmed by the proposed transaction. And, in a display of documentary and testimonial support, the Government has been able to show that Oracle's internal documents (the very documents Oracle lawyers tried to avoid producing), as well as the testimony of Oracle executives, support the Government's case. Who could forget Professor Elzinga's bar chart summarizing Oracle's own internal discount request forms, showing far more requests to discount against PeopleSoft than any other vendor. SAP was next. But, then, in the words of Dr. Elzinga, it looked like the numbers "fell off a ledge." Here is Elzinga's chart: [SEE CHART IN EXHIBIT (e)(23)(a) ATTACHED HERETO] If those other sources of software -- outsourcers, Microsoft, Lawson, and the like -- actually constrained Oracle's ability to raise prices, Elzinga asked, why weren't they listed as justifications for discounts more frequently? In short, the Government made its point by testimony, by expert case study, and by a review of Oracle's own documents. One such document said it all: "Remember, SAP is not the enemy.... PeopleSoft is the enemy." The Government's pretrial plan next argued that there are other customers, beyond those in the upmarket, that have more than 3 choices -- Oracle, SAP, and PeopleSoft to be sure, but sometimes another vendor, such as AMS or Lawson. In my opinion, the Government has proven this contention. Oracle lawyers denied even the existence of a mid-market. But the testimony of Oracle executives like Chairman Jeff Henley, and the very organization of the Oracle salesforce into smaller customers and larger customer groupings, admitted the point away. The Government showed that the competition from AMS is limited to financial software in some Government installations -- indeed, limited to Government installations where AMS is already running. And the Government showed through customer testimony and through the expert testimony of Harvard Professor Marco Iansiti (who actually analyzed all vendors' software) that Lawson's products cannot do the job in the upmarket, both for functional reasons and for lack of sufficient worldwide support. Third, the Government has stated in its Pretrial Brief that the Big 3 vendors, through the knowledge they obtain in the sales cycle, know the needs of potential customers and therefore know which of the other vendors constitute actual competition for the account. The Big 3 vendors can therefore price discriminate -- charge more to some customers than others, depending on the customer's needs and the level of competition. This has been proven through the testimony of PeopleSoft, the deposition testimony of SAP executives and through the admissions of Oracle's own executives. This is a key point. Government experts have demonstrated that Oracle's own discount documents show key evidence of price discrimination. So even if Lawson competes for an occasional upmarket account, that competition will not constrain Oracle post acquisition in accounts for which Lawson is not a good substitute. The proof goes even further, demonstrating that Oracle and PeopleSoft are head-to-head competitors for many accounts. The ability of the post-acquisition firm (Oracle) to price discriminate means that even some competition from SAP for those accounts will not prevent Oracle post acquisition from gouging those customers. This proves the barriers to entry are substantial. If it would take even Microsoft years to become a viable competitor, Oracle would be able to price gouge post-acquisition. So, all in all, the Government has presented a compelling case. A word about the Government's presentation. Some analysts and reporters have criticized the Government's cross-examinations, in particular, as not always being as polished as those of Oracle's lawyers. But, in my opinion, the Government has conceptualized and put on a complicated case in a way in which all important parts come together at the end. The Government began with Oracle customers, then PeopleSoft customers, assessing their procurements and the need for competition from PeopleSoft to guarantee the benefits of free market competition. The Government put on PeopleSoft executives to explain what the company makes and how it competes. The Government showed that Oracle executives under oath in deposition see the market in much the same way as PeopleSoft does. Ditto for SAP, whose testimony came in through deposition. And, at the end, the Government experts, one traditional economist and the other a famous name in auction-based modeling, pulled it all together. From the perspective of traditional economics (represented by Professor Elzinga) the markets are highly concentrated, the barriers to entry are high, and the merger will produce anti-competitive results. Oracle's own discount forms demonstrate that. And from the perspective of economic effects, the result is equally clear. Dr. McAfee's analysis showed that PeopleSoft's competition has resulted in lower prices to customers and that without PeopleSoft's competition, prices would go up significant amounts. Three different lines of Dr. McAfee's analyses support the same conclusion. I expect that Oracle will continue to pick at the studies by the Government experts. Perhaps they included an erroneous Oracle form or two. Perhaps they should have drawn the upmarket line as $600,000 as opposed to $500,000. Perhaps, perhaps, perhaps. The body of the Government's case has come into the record unscathed. No Government customer witness under cross-examination ended up agreeing with Oracle's lawyers. They all held their ground and rejected Oracle's contentions. There were no gaffes in the testimony of the three Government experts. Perhaps Oracle can continue to "nibble around the edges" but the experts themselves stood by their powerful conclusions. [END] TRIAL DAY 12: JUNE 23, 2004 The judge released an order today, denying the media's request to unseal all highly confidential documents previously submitted into the record under seal and to unseal, as well, the report of Judge Legge, appointed as a special master in the case. The order pointed out that Judge Legge's report includes many confidential documents that did not end up becoming part of the record. The order agreed that many companies designated as "highly confidential" documents that really did not deserve that characterization, but the order said that the court had remedied that practice by permitting only appropriate documents to be kept confidential, and if some documents have been improperly designated, the court will deal with them on a case-by-case basis. The first witness of the day was Richard Knowles, the vice president of Operations -- North America, for SAP America. Knowles was put on by Oracle. One of the SAP lawyers told me at the beginning of the trial that if Oracle called SAP as a witness, SAP's testimony would help the Government just as much as Oracle. That is precisely what happened. TRIAL DAY 12 On direct examination, Oracle elicited testimony that SAP was a large global competitor, sufficient at least in the witness's view to compete for any customer in any vertical. The witness explained that SAP separated customers into categories based on revenue size, rather than functional needs. The witness claimed that an Oracle acquisition of PeopleSoft would result in more rather than less competition, and the witness claimed that SAP was "neutral" with respect to the Oracle proposal to acquire PeopleSoft, not taking sides on the issue. Finally, the witness identified internal documents that evidenced concern about competition from Microsoft. In one puzzling bit of direct testimony, Oracle elicited testimony that SAP's internal discount request forms showed relatively little final round bid competition with either Oracle or PeopleSoft. Since Professor Elzinga has already shown that Oracle's forms show a great deal of head-to-head competition between Oracle and PeopleSoft, the testimony elicited by Oracle from Knowles appeared to confirm a key component of the Government's case -- that Oracle and PeopleSoft compete directly against each other far more often than either company competes with SAP. On cross-examination, the SAP witness said a number of things that support the Government's position. First, with respect to TRIAL DAY 12 Microsoft, the witness admitted that with all of Microsoft's efforts, that company could compete for only those customers up to $300 million in revenue -- the lower part of the mid-market. Next, the witness admitted that SAP's marketshares for the U.S. were different than its shares worldwide, bolstering the Government's view of a U.S.-only market. And the witness admitted that SAP had failed to capture significant U.S. business in a number of verticals, until including utilities and banking. The witness testified that it might take three years for SAP's alliance with Accenture to generate a fully functional product, bolstering the Government's position that the product markets at issue in this case are hard to enter. The witness acknowledged under cross-examination that the view that an Oracle acquisition of PeopleSoft would produce more competition was simply the witness's personal view, and conceded that internal SAP documents recognize that a combined PeopleSoft/J.D. Edwards will be SAP's key competition for application software. The witness also acknowledged that SAP was not, in fact, neutral -- SAP had testified in the E.U. on Oracle's behalf, and the CEO of SAP had publicly criticized the DOJ for bringing this case. All in all, it is difficult for me to see how the SAP appearance helped Oracle's case all that much. TRIAL DAY 12 In the afternoon, Doug Burgum, senior vice president of Microsoft Business Solutions testified, out of order, as the final live Government witness. Burgum had been the CEO of Great Plains. In fact, he testified how he had mortgaged his family farm in North Dakota to seed the company and how he had taken the company public. When the company was sold to Microsoft, he became a key executive of that company, running all four Microsoft back-office product lines. Burgum produced internal Microsoft documentation showing precisely the market segments by size and description that Microsoft products are aimed at -- all within the small enterprise and mid-market space. Burgum explained that not all customers fit neatly into a category; nevertheless, it is clear what markets Microsoft's products are intended to address. Microsoft mid-market products may also be targeted at subsidiaries and divisions of large corporations, provided the subsidiaries or divisions need a low level of functionality. The witness went on to explain (and showed internal Microsoft documents to support his claim) that Microsoft has no intention to ever sell to large complex customers in the up-market. That market, according to the witness, requires a direct sales force TRIAL DAY 12 and large amounts of support, and includes relatively few potential customers, as compared with the mass distribution markets that Microsoft focuses on. Burgum described, and showed documents about, Microsoft's acquisition talks with SAP. Microsoft started those talks in response to Oracle's hostile tender offer, according to the witness. Of significant interest was the fact that Microsoft's internal analysis, made at the beginning of the SAP talks, concluded that Oracle's unsolicited tender offer, even if unsuccessful, would weaken PeopleSoft and distract its executives. The witness testified that the talks with SAP were abandoned when Microsoft focused on how difficult integration of such a large acquisition would be. Most significant was Burgum's testimony about the heretofore secret details of Microsoft's "Project Green." According to the witness, Microsoft is writing a common code base to integrate all four of its product lines (each product line came to Microsoft through acquisition of a company). Release of the first version of this new code base is dependent upon the prior release of a new Microsoft operating system, and a new Microsoft server product, both code named "Longhorn." According to Burgum, the new application product, Project Green, will not likely be TRIAL DAY 12 initially released until 2008 or 2009, at best, and the new product will not get even the functionality of the existing Microsoft mid-market products until about 2011. Oracle's cross-examination did not seem well focused. Oracle first attempted to show Microsoft's bias, by arguing that Oracle's tender offer would endanger Microsoft's database business. Then, Oracle attempted to prove that Microsoft was already selling into the up-market--something Oracle would have to prove in order to demonstrate the potential of timely entry by Microsoft, given the fact that Microsoft's new product won't be ready until 2011. But Dan Wall, Oracle's attorney, could show, at best, only one or two large Microsoft application customers, and some internal prospect lists, claiming a handful of opportunities. In any case, the witness explained that he had set up an internal clearance procedure to make sure that Microsoft salesmen did not sell its product into large accounts, beyond the capabilities of those products. And the witness stuck to what the internal document showed -- that Microsoft does not target anything larger than the upper mid-market. Wall also attempted to get the witness to agree that the up-market is saturated, with few new procurement opportunities. But out in the hall, after the court session, the press wondered how TRIAL DAY 12 Oracle could argue on the one hand that Microsoft wanted to enter the up-market, but at the same time claim that the up-market was saturated. Why would Microsoft invest hundreds of millions of dollars to enter a saturated market? Oracle originally planned to call its own Microsoft witness, but Wall suggested that Oracle would forego that opportunity. Given the state of the record, it is difficult to see how Oracle can prevail on the issue of Microsoft's potential entry. It is one thing to suggest to the press that Microsoft is really lying and actually intends to go into the up-market, however implausible that seems. But proof in court is something altogether different. Oracle must prove that entry by Microsoft is "likely," "timely," and of sufficient competitive scale before the law will recognize Oracle's claims. Right now, I believe Oracle is nowhere near to satisfying these requirements. [End] TRIAL DAY 13: JUNE 24, 2004 Oracle lawyer Dan Wall began the day by announcing that Oracle will eliminate a number of the witnesses it previously intended to call, and that the trial might conclude the following Wednesday or Thursday. The first Oracle witness of the day was Michael Sternklar, of FESCo, a division of Fidelity Investments. Oracle has argued that outsourcing is a major source of competition for the three (Oracle, PeopleSoft, and SAP) application software vendors, and FESCo, a large outsourcer, was expected to support Oracle's claim. The witness did assert that business processing outsourcers ("BPO") have competitive advantages over traditional software vendors in terms of service offerings, but between the direct and cross, the witness admitted: 1. FESCo only has 10 large customers, including its own parent company, and has plans to add only one new customer next year and five the following year; 2. FESCo cannot provide services to employees located outside the U.S. even if its potential customers might be headquartered inside the U.S.; and TRIAL DAY 13 3. Most BPO companies run software provided by Oracle, SAP, or PeopleSoft. In fact, most run PeopleSoft software; FESCo itself runs Oracle software. The witness testified that FESCo made an acquisition of some new software from IBM and that FESCo is in the process of rewriting and upgrading its software. The witness said that FESCo will offer this new product within 18-24 months and will migrate existing customers (all of which are running on Oracle) to that new platform within five years. However, he conceded that these goals are "aspirational." Under questioning from the judge, the witness admitted that there have already been significant delays in the "major rewrite" of its new product, currently in process. Oracle's second witness of the day was Ken Harris, a consultant who markets his services under the name "Retail.In.Genius." Harris testified he had prior CIO experience at Taco Bell, Nike, and Gap. He expounded at great length about all the techniques he used to get vendors to lower their prices to him, including the alternative of "doing nothing." He emphasized with great TRIAL DAY 13 gusto all the alternatives he claimed enterprise customers have for back-office software needs. The direct testimony was punctuated by a chess match between Dan Wall and the Government attorney, Steven Kramer. Whenever Wall would ask a conclusary question about what enterprise customers do, Kramer would object on the ground that the witness could not testify beyond his experience at the three customers where he had been employed. Only witnesses qualified by the court as "experts" can testify more broadly, and this witness, Mr. Harris, was not qualified as an expert, argued Kramer. Wall kept suggesting that he might offer Mr. Harris as an expert, but he never did. So, the record of this witness's testimony is what is called "lay opinion," the weakest and least credible form of testimony. Far worse for Oracle was the judge's response to Kramer's various objections. When Dan Wall argued that Mr. Harris should be able to give the same kind of testimony that the Government's customer witnesses had, the judge himself pointed out that, unlike those customers, Mr. Harris was a "paid witness." Touche. Equally telling to me was the fact that the judge had no questions of his own for this witness. TRIAL DAY 13 I would have been tempted to pass on the opportunity to cross-examination someone who even the judge understood was a "paid witness." But Kramer had even more for the record. Mr. Harris admitted under cross that only one of the three procurements he talked about was a major procurement, and that occurred in 1989. More importantly, he conceded that during his entire job tenure for each procurement, he selected either PeopleSoft, Oracle, or SAP as a final bidder and winner. He also admitted that he never evaluated Lawson in any detail. I have no idea what Oracle's lawyers were thinking when they decided to call the witness. I think it would have been better for them to have no customer testimony than to have, in opposition to all of the Government's customer witnesses, this paid witness. The third Oracle witness of the day was the director of Personnel Service Delivery at Bank of America. He testified that prior to its merger with Fleet, the witness used PeopleSoft core HR products, along with many point solutions. Thereafter, as a result of Fleet's prior use of outsourcing, Bank of America has used FESCo, running Oracle software. Indeed, on cross-examination, the witness admitted that despite all this testimony about outsourcing, what basically happened with adoption of the Fleet system is that Oracle replaced PeopleSoft. TRIAL DAY 13 Finally, DOJ officially concluded its case by playing a videotape of the deposition testimony of Orlando Ayala, a senior VP at Microsoft. Dan Wall had previously observed that the witness, like Madonna, goes by only a single name, in this case "Orlando." This piqued the judge's interest -- he subsequently commented about the interesting appellation. Orlando's testimony hammered home the points the Government has made about Microsoft's effect on the high-function software market for HR and financial services. The witness testified: 1. Non-established vendors have no sales opportunities with the largest customers, as it takes many years and substantial investment to enter the up-market, even if the vendor has a technically capable product; 2. On the other hand, Oracle, SAP, and PeopleSoft, which are established in the up-market, have substantial opportunities; 3. Microsoft's Axapta product is not capable of meeting the needs of multinational customers; 4. Microsoft has scaled back its marketing messages in order to avoid overselling its product and producing failed implementations; TRIAL DAY 13 5. While Microsoft may occasionally get a large customer, Microsoft's offerings are not suitable for all such customers; 6. Microsoft has over-emphasized its business software, and its real life experience in trying to break into even the mid-market as an application vendor has been "a humbling experience." Following the Orlando videotape, the Government officially rested its case. Dan Wall moved for a "directed verdict" -- a routine motion at this stage of the case. The motion argued that the Government hasn't proven the basic elements of its case, so there is no point in making Oracle put on a defense. Claude Scott responded by going down chapter and verse many of the proof points of the Government's case. The judge asked a couple of questions and took the motion under submission, to be ruled upon at a later time, perhaps not until a final decision on the merits. [End] TRIAL DAY 14: JUNE 25, 2004 The judge began the day by telling both sides that he is going to order them to provide definitions to certain terms. They are to work together to provide agreed-upon definitions, if possible, and otherwise, they can each submit a set of definitions. The terms include "high-function software," "function," "ERP," and "EAS." I believe the judge asked for these definitions because he has begun to think about writing his opinion in this matter. Typically, a judge's opinion in a complicated technology case will start with definitions of key terms used in the decision, so the reader does not have to be versed in industry jargon to get the gist of what the judge is saying. The judge gave no indication as to how he will rule on this matter. But it is hard to view the fact that he is asking for a definition of "high-function software" -- something Oracle claims does not exist -- as a happy development for Oracle. Nevertheless, out in the hall at the break, Oracle's lawyers told the press, and some journalists dutifully reported, that the judge's request "appeared to be questioning the basis of the Government's case." TRIAL DAY 14 Oracle called its own Executive Vice President of Application Development, Ron Wohl, as the first witness of the day. Wohl made three basic points in his direct testimony: 1. Scalability: Wohl said scalability of application code is "not a primary constraint" on the ability of enterprise software to serve large numbers of users and hence, a company like Microsoft can scale its mid-market product to serve much larger concerns; 2. Integration: Wohl said that EAS suites are best suited to smaller companies, while larger companies operate a wide variety of software; 3. Innovation: Wohl claimed that competition is but one driver for innovation and that Oracle adds new features not so much to compete with PeopleSoft, but to satisfy customer demand. Before the Government lawyer could even begin his examination, the judge began to ask Wohl questions. Under questioning from the judge, Wohl conceded that an application can scale easily to larger numbers of employees only if other needs remain constant. The Government lawyer followed this up, getting Wohl to concede that complex scalability issues cannot be solved through TRIAL DAY 14 architecture alone (as he had testified on direct) but also must be addressed in the software code. The Government lawyer, Phil Malone, next showed Wohl internal Oracle documents and got him to admit that PeopleSoft's innovations (and superior HR products) forced Oracle in turn to innovate. This is an important concession. The Government has already shown that head-to-head competition with PeopleSoft produces lower Oracle prices to customers. Now the Head of Application Development of Oracle conceded (and his own documents confirmed) that head-to-head competition with PeopleSoft produces greater innovation by Oracle. Showing that competition between the merging companies has produced lower prices and greater innovation is the key to the Government's case. Anticipating arguments that I expect Oracle will make in the coming days, Malone also forced Wohl to concede that Oracle's products, such as application servers and integration products, are not substitutes for enterprise applications. Finally, under cross, Wohl admitted that he was not consulted regarding plans for R & D head count post-acquisition and that head count numbers he provided in that regard are "speculation." TRIAL DAY 14 Oracle's final witness of the day was Safra Catz, its Co-President and a board member. Oracle's other Co-President, Chuck Phillips, was publicly leading the effort to acquire PeopleSoft until about the time his rather unfortunate "oligopoly" language came to light, at which point Catz seemed to become the primary spokesperson on the issue. Catz tended to give long-winded answers to nearly every question, so the significance of what she said, particularly on cross-examination, seemed to be lost in media reports. Catz's first major point was that Oracle sales for what the Government calls "high-function software" (more precisely, Oracle's sales of HR and financial platforms to large corporations) are "chump change." Such sales totaled about $18.6 million for the first three quarters of 2003. Her point was that any anti-competitive effects of the proposed acquisition on such sales would be more than offset by the large efficiencies the deal will net for Oracle. But on cross, the Government lawyer, Claude Scott, showed data making it clear that the Oracle-PeopleSoft rivalry is hardly over "chump change." Scott showed the witness internal Oracle win-loss data revealing that for a single quarter during the relevant period, Oracle won only three deals against PeopleSoft. But PeopleSoft won 11 deals against Oracle, and the two companies were continuing to compete over no fewer than 59 deals in the pipeline. TRIAL DAY 14 Second, on direct examination, Catz contrasted her view of the trivially small affected product market with the vast savings she claims Oracle will achieve in the transaction. She presented a "model" -- a set of formulas and relationships - -- to demonstrate what would happen to the two companies post-acquisition. Her point was that the post-acquisition company will be able to function at a higher scale. She said she got this idea after PeopleSoft proposed that it acquire Oracle's application business in 2002. Catz said the idea for a more efficient, combined company was first proposed to her at a meeting with PeopleSoft CFO Kevin Parker. Catz's testimony on the post acquisition company goes to the antitrust issue of "efficiencies." If a proposed deal will cause untoward concentration and likely decreases competition, the defendant can still prevail if it can show that the deal will make the post-acquisition company so much more efficient -- in certain specified ways -- and these efficiencies will be passed on to consumers in the form of lower prices. TRIAL DAY 14 There are several unusual aspects of Catz's model. Of course, it is normal in an antitrust case for a defendant to argue that the proposed deal will make the merged company more efficient. But this type of testimony usually comes into the record as a learned study prepared by a distinguished expert. Here, Oracle is presenting its efficiencies study through a lay witness. Indeed, Claude Scott established on cross that none of Oracle's distinguished economic experts, nor even Ron Wohl, the Oracle executive in charge of application R & D, were consulted with regard to the numbers in the model. From the testimony, it appears that Catz and Ellison just took a pre-existing model developed to evaluate potential acquisitions of J.D. Edwards and Lawson, and made up numbers for a proposed PeopleSoft acquisition. Scott established on cross that there are no extant work papers that show how Catz and Ellison calculated the various numbers put into the model. Catz's model showed that Oracle can save over $1 billion a year post-acquisition by firing about 6,000 PeopleSoft/JDE employees. Scott established that under the model, the post-acquisition firm would keep virtually no PeopleSoft sales and marketing personnel (because it won't be selling and marketing PeopleSoft products) and only about half of PeopleSoft's development personnel (who would work on the next generation Oracle product). The large numbers of fired employees got most of the headlines. TRIAL DAY 14 But to me, the most important part of the testimony was Catz's admission, under cross, that there are no features that PeopleSoft products have that Oracle could not develop on its own -- without acquiring PeopleSoft. Indeed, Catz admitted again and again that Oracle and PeopleSoft HR and financial management offerings are virtually identical. She also admitted that Oracle wants PeopleSoft for its customers, and perhaps for some of its developers, but not for any of its technology. The key foundational claim of the Government's case -- found in the Government's trial brief among many other places -- is that by this tender offer, Oracle is trying to acquire PeopleSoft's customers without competing for them in the free market. Catz admitted the point. And further, the Government argues that Oracle will inflict upon customers enormous switching costs, and upon thousands of people the burdens of job loss, not to get some greater functionality that will be used to create a better product, but to be able to sell PeopleSoft customers all the various products Oracle makes. Catz admitted that post-acquisition, new Oracle application products, like the current products, will only run on the Oracle database. She said PeopleSoft customers who choose to migrate to TRIAL DAY 14 the new Oracle applications will be given an Oracle database free at first (this is a wholly new claim, not found on the Oracle website, as far as I know) and that Oracle will do whatever possible to facilitate the migration to the new platform. Of course, Catz admitted that Oracle will not pay the costs of re-implementation and integration for both applications and a new database -- costs that are at least seven to eight times the cost of the software itself, according to prior witnesses. Catz also admitted that had Oracle acquired J.D. Edwards or Lawson, Oracle would have increased maintenance prices to the customers of those companies. [End] TRIAL DAY 15: JUNE 28, 2004 Oracle lawyer Dan Wall began today by announcing that Oracle will not be calling PeopleSoft CEO Craig Conway as a witness after all. Oracle has had Conway on its witness list since Oracle's initial witness designation in April. Oracle claimed that it would question Craig for six hours, more than any other witness. When I asked the Court to cut the amount of time Oracle proposed to interrogate Conway, Oracle insisted that it needed the full six hours. Now it turns out that Oracle has decided that it doesn't need Craig's testimony at all. Perhaps Oracle never did intend to call Craig -- and only went through the process of subpoenaing him for PR reasons or to attempt to distract him from focusing on business. Or perhaps Oracle's lawyers actually intended to call Craig, but finally came to their senses. We'll never know. But it seems clear to me that Craig's testimony would have severely damaged Oracle's case. Craig would have explained that PeopleSoft has gone through the past year, retaining the trust of customers and employees, in the face of Oracle's unsolicited tender offer. It would have been powerful testimony, but we won't see it, because Oracle apparently decided it was afraid to examine Craig in open court. TRIAL DAY 15 After the announcement abandoning Craig's interrogation, Oracle put on its first witness of the day, John Coughlan, CEO of Lawson Software. Throughout the trial, Oracle has touted Lawson as a rising upmarket competitor, sufficient in its own right to prevent Oracle from raising prices following an acquisition of PeopleSoft. On direct, Coughlan took issue with statements made by government attorneys to the effect that Lawson is "not a serious player" in the upmarket and is "not likely to be" in the future. Coughlan spoke of his successes among large, complex customers, especially in health care, retail, and the public sector, listing several marquee customers in each vertical. Coughlan claimed that Lawson has competed with Oracle and PeopleSoft on "almost every account." But the cross-examination turned into a disaster for both Lawson and Oracle. Here is the lead from INFORMATIONWEEK: CEO John Coughlan, testifying at the Oracle antitrust trial, tried to paint his company as more than a mid-market business software vendor, but was confronted TRIAL DAY 15 with examples of enterprise customers that the Justice Department used as examples of failure. Lawson Software, Inc. and CEO and president John Coughlan had a bad day Monday. If anything, I think that INFORMATIONWEEK understates the extent of the carnage. On direct, Coughlan hyped the fact that he had taken his company public. On cross, he admitted that he had taken the company out at $14 a share, and it was now trading at $7 a share. In 2003, it was revealed, Lawson had a net operating loss. Government attorney Claude Scott then showed example after example of Lawson problems that left customers dissatisfied, and having to turn to Oracle and PeopleSoft. He backed up each example with internal Lawson documents and customer memoranda. MasterCard complained to Lawson about inadequate payroll technology, for example, and concluded that it was not interested in Lawson's software because of lack of functionality. Scott next showed that the City of Dallas, after installing Lawson's product, complained of not being able to pay its police and firemen. The witness acknowledged that the City of Dallas stopped paying Lawson. TRIAL DAY 15 Scott showed that both Johnson & Johnson and McGraw-Hill, used as examples of Lawson trophy customers in the direct examination, rejected further Lawson purchases, because of Lawson's inadequate functionality. And so it was, account after account. Problems at Hyatt Hotels, problems at Northrop Grumman, etc. And more product deficiencies. Lawson can't support Asian languages, Lawson has little revenue from international customers, etc. Coughlan had claimed in direct testimony that his company had sales to many customers in excess of $500,000 in licensing revenues, but on cross it was revealed that his list included services, not just licensing revenue. Lawson, it turns out, has an R&D budget about 7 percent of that of PeopleSoft. Finally, Scott turned to full-page ads in the WALL STREET JOURNAL which Lawson took out to publicly thank Oracle for putting Lawson in the same market as PeopleSoft and SAP, based on Oracle's rhetoric in this trial. The witness admitted that his company had spent more than $300,000 on these "Thank you Oracle" ads. In the afternoon, Oracle put on one of its principal economic experts, MIT Professor Jerry Hausman. Hausman testified for several hours, attacking the testimony of both of the government's principal experts, Ken Elzinga and Preston McAfee. TRIAL DAY 15 Hausman first criticized Elzinga's work, finding that the relevant market should be worldwide (or at least Europe and the United States), rather than just the U.S. He next claimed that the product market was poorly and vaguely defined, and claimed that the market should include competition from outsourcers, and from Microsoft, AMS, and Lawson. Actually, the government experts included AMS and Lawson in their analysis, but Hausman claimed those companies were insufficiently represented in the government's data set. Hausman said his own studies showed that between AMS and Lawson, the two companies covered half of the relevant market, and would keep Oracle post-acquisition from raising prices. He went on to explain that he thought Lawson would constrain Oracle post-acquisition, even though Lawson is present in just a couple of verticals, because Lawson would move to additional verticals if Oracle tried to raise prices. How Lawson would be able to do this was never explained. Hausman also claimed that because few financial or HR products are sold without the inclusion of products from other pillars, resulting in blended pricing, these sales "in bundles" would in some way constrain Oracle from raising prices on financial and HR software. And Hausman testified that the largest customers get TRIAL DAY 15 the largest discounts, so there really isn't a class of customers to price-discriminate against. Most of Hausman's testimony was directed against Professor McAfee's work. McAfee had testified extensively as to the anticompetitive effects of the proposed acquisition -- likely substantial price increases and a diminution in innovation, in McAfee's opinion. Hausman, on the other hand, argued that this case was not proper for a "unilateral effects" approach because the elimination of PeopleSoft would not affect price, post-acquisition. According to Hausman, customers involve only 1-2 vendors in pricing negotiations. Furthermore, he ran his own regression analyses, getting results different from McAfee; according to Hausman, the presence of additional vendors does not result in lower prices. As the day went on, Hausman's testimony seemed to get more and more extreme. He claimed that his study showed that SAP constrains Oracle's prices just as much as PeopleSoft -- notwithstanding the fact that Oracle's own discount forms show much greater competition from PeopleSoft. He claimed that just two vendors, SAP and Oracle, would provide sufficient competition to keep Oracle from raising prices post-acquisition, despite the testimony of customers, conspicuously including Oracle's TRIAL DAY 15 customers, to the contrary. Hausman concluded that prices would not rise at all after the acquisition, and innovation would not be affected by the absence of PeopleSoft. Hausman's direct testimony ended at the conclusion of the court day. He will be cross-examined tomorrow. [END] TRIAL DAY 16: JUNE 29, 2004 There was only a half-day of court time available today, and the cross-examination of one of Oracle's principal economic experts, Jerry Hausman, consumed all of it. Hausman was cross-examined by Government attorney Sandy Adler. Sandy came into town from Washington just for the cross-examination; he has no other role in the case. In fact, whenever Jerry Hausman testifies as an expert against the Government, Sandy Adler comes into town to cross-examine him. This is like having a "defense specialist" on the bench of your basketball team that you bring into the game just to shut down the other team's shooting guard. Before the cross-examination started, Hausman stopped by to talk to someone in the row behind me. In a stage whisper everyone could hear, Hausman said, "[the] last time Adler cross-examined me, I made minced meat out of him." Well, not today, Jerry. Adler's manner is scholarly rather than accusatory. When he burrows in on the wording of the expert's written report, it sounds like he is exploring the subtleties of a scholarly tract. He sounds well versed in economics and statistics -- and he looks like he could be an expert witness himself. Today, aided by the witness's own demeanor, Sandy started with a few real zingers. Sandy's first job, I believe, was to attack the witness's credibility - -- to make the witness sound like he is TRIAL DAY 16 given to overstatement and exaggeration, and therefore should not be believed. Sandy showed on the screen a sentence from Hausman's written report in this case, in which Hausman claimed that Professor Elzinga's reliance on the testimony of PeopleSoft executives is: "Akin to using the testimony of a drug dealer to decide that crack cocaine is good for you." "Did I read that correctly?" Sandy asked. "Yes," Hausman replied. Next, Sandy raised the issue of how Oracle's "efficiency story" was prepared. In an earlier journal installment, I explained that in all cases I have been involved in, efficiency justifications are made through learned economic studies, put into the record through expert testimony. Here, Safra Catz herself did the efficiency study. This is extremely unusual. Sandy asked Hausman if Hausman did any original research on Oracle's efficiency report. Hausman claimed that he had. So Sandy read into the record Hausman's deposition testimony in which Hausman said, "I did no original research. I took what Oracle had done and tried to look at it...." Next, Sandy read the witness a statement from Oracle's white paper submitted to the Department of Justice during the investigatory phase of this matter. During that phase, Oracle used a different economic expert, Fredrick Warren-Boulton, but TRIAL DAY 16 the earlier expert report was still submitted in Oracle's name. The Oracle economist at the earlier phase (Warren-Boulton) submitted a report saying that the "markets for EAS software functions similarly to auction or other bidding markets in important respects." This is precisely the point that Professor McAfee, a Government expert witness, made earlier in the trial. Everyone in the courtroom knew that Hausman would have to contradict the expert report Oracle itself submitted to the Department of Justice. That was embarrassing enough. But Hausman went further and said: "I want to say this in the nicest possible way, but Dr. Warren-Boulton, in my view, is an extremely poor economist. So there is no reason I would agree with him." Wasn't Warren-Boulton formerly the head of the Economics Department of the Antitrust Division, Adler asked. "That's true," replied Hausman. Asked whether he even bothered to read the earlier economist's report, Hausman said that he started to, "[b]ut I thought it was of extremely poor quality, so I stopped very early on." So you reject the view of Oracle's previous economic expert, Sandy wanted to know. "Not only rejected that view, I rejected all his views. ...I wouldn't take very seriously what he had to say." TRIAL DAY 16 Adler next turned to the issue of whether Oracle post-acquisition would be able to raise prices. On an earlier trial day, Oracle expert witness Tom Campbell said he could not say whether, if Oracle, PeopleSoft, and SAP all merged into a single firm, that firm could raise prices. Hausman did Campbell one better. Hausman started by claiming that if all three vendors merged into a single firm, that firm COULD NOT raise prices. Actually, when Hausman equivocated on the stand, Adler played his deposition testimony on the court monitors. Then Hausman claimed that even if all vendors of packaged software joined together, they COULD NOT raise prices. Again, the witness equivocated, so Adler played the witness's deposition testimony in which Hausman claimed that even if all vendors of packaged software joined together, they would not satisfy the "hypothetical monopolist test" of the Merger Guidelines. On the witness stand, Hausman retracted his earlier position (also given under oath), and claimed that the companies could raise prices for financial software, but he was not sure about HR software. Adler pressed the point. Did the witness ever correct his prior erroneous testimony that he was now retracting? After all, the witness made many other corrections to his deposition transcript. TRIAL DAY 16 The witness said he just corrected the typos in his transcript. "But you couldn't tell the other side that you've changed your entire view of the market?" asked Adler. The witness said, "No," and provided explanation. The next subject in the testimony was a key to the Government's case. Hausman said in his direct testimony that he had conducted his own regression studies and those studies showed that when additional bidders were added to the mix, prices did not go down -- that is, increasing the number of competitors does not result in greater competition in the form of lower prices. This is Oracle's fundamental position. But through elaborate and difficult interrogation of a hostile witness, Adler got Hausman to admit that whatever the general truths of Hausman's statements on adding bidders, Hausman's own studies showed that adding PeopleSoft as a competitive bidder induces Oracle to give over a 10 percent greater discount. Then the witness went on to contradict the next Oracle expert, Dale Kutnick, who hasn't even testified yet. Hausman said he doesn't expect "technological breakthroughs in the industry." But Kutnick is expected to argue that there will be changes in technology so dramatic that the Government should stop worrying about whether Oracle can increase prices post-acquisition. TRIAL DAY 16 There was an extensive discussion about whether large customers actually pay lower prices for the same software. On direct, Hausman said they do. But on cross, his testimony was equivocal. Even the judge repeatedly asked questions about this. Hausman agreed that large corporate customers pay more per seat than smaller customers, but he also argued that they are not buying the same product. The Judge seemed baffled by that explanation. Finally, Adler asked Hausman when certain data he presented in his direct testimony had been prepared. Adler made it clear that the data had not been provided to the Government, as required by the court's procedures. Hausman admitted he had made the calculations days ago. Adler made it clear that Hausman had done a lot of work since his deposition was taken and his expert report was filed, but none of that had ever been given to the Government. Adler then revealed that Hausman had previously been sanctioned by a federal court for not providing information to the other side in an appropriate manner. The court in that case struck Hausman's entire testimony and the court of appeals affirmed that decision. [End] Important Additional Information PeopleSoft has filed a Solicitation/Recommendation Statement on Schedule 14D-9 regarding Oracle's tender offer. PeopleSoft stockholders should read the Schedule 14D-9 (including any amendments or supplements) because these documents contain important information. The Schedule 14D-9 and other public filings made by PeopleSoft with the SEC are available without charge from the SEC's website at www.sec.gov and from PeopleSoft at www.peoplesoft.com. EX-99.(E)(23)(A) 4 f97751a9exv99wxeyx23yxay.htm EXHIBIT (E)(23)(A)

Exhibit(e)(23)(a)

Chart produced by Dr. Elzinga in testimony in the antitrust trial — see “The Government Rests” in exhibit (e)(23).

EX-99.(A)(104)

                                                                Exhibit (a)(104)


                                      [STATE OF DELAWARE COURT OF CHANCERY SEAL]

                                                EFILED: JUN 24 2004 4:31 PM EDT
                                                               FILING ID 3789626

                                    EXHIBIT A



                IN THE COURT OF CHANCERY OF THE STATE OF DELAWARE
                          IN AND FOR NEW CASTLE COUNTY

ORACLE CORPORATION, a Delaware              )
corporation, and PEPPER ACQUISITION         )
CORP, a Delaware corporation,               )
                                            )
                       Plaintiffs,          )
                                            )        Civil Action No 20377 NC
              V                             )
                                            )             PUBLIC VERSION
PEOPLESOFT, INC, a Delaware corporation,    )                REDACTED
CRAIG A. CONWAY, DAVID A.                   )
DUFFIELD, ANEEL BHUSRI, STEVEN D.           )
GOLDBY, A. GEORGE BATTLE, FRANK             )
J. FANZILLI and CYRIL J. YANSOUNI,          )
                                            )
                     Defendants.            )

                              AMENDED COMPLAINT FOR
                        DECLARATORY AND INJUNCTIVE RELIEF

      Plaintiffs Oracle Corporation ("Oracle") and Pepper Acquisition Corp
("Purchaser"), a wholly-owned subsidiary of Oracle, for their complaint against
defendants PeopleSoft, Inc. ("PeopleSoft"), Craig A. Conway, David A. Duffield,
Aneel Bhusri, Steven D. Goldby, A. George Battle, Frank J. Fanzilli, Cyril J.
Yansouni and Michael J. Maples ("Director Defendants") allege, upon knowledge as
to themselves and their own acts and upon information and belief as to all other
matters, as follows:

                                  INTRODUCTION

      1. Oracle launched an all-cash, all-shares tender offer for shares of
PeopleSoft stock (the "Tender Offer") more than a year ago, on June 9, 2003. The
Tender Offer is non-coercive and represents a healthy premium: Oracle's most
recent $21 per share bid for PeopleSoft stock, announced on May 14, 2004,
represented a 21% premium over that day's close of $17.30 per



share. On information and belief, the small increases in PeopleSoft's stock
price (though still well below the offer price) since the beginning of the trial
in the related case of United States v. Oracle have been in whole or in part
based on speculation that Oracle's tender offer will succeed.

      2. In the past year, PeopleSoft's board of directors (the "Board" or
"PeopleSoft's Board") has had ample time to pursue alternatives to Oracle's
Tender Offer, and PeopleSoft stockholders have been inundated with information
and analysis of the Tender Offer from both PeopleSoft and Oracle, as well as
from industry and financial analysts, the media and other interested parties. In
light of these facts, there is no legitimate justification for the Board's
continued refusal to redeem PeopleSoft's poison pill, exempt the Tender Offer
from Section 203 of the Delaware General Corporation Law ("Section 203") and
permit PeopleSoft's stockholders to consider Oracle's Tender Offer. Nor can the
Board justify its decision to continue the other defensive measures it has
erected, including most notably the customer give-away program PeopleSoft termed
the "Customer Assurance Program" (the "CAP"), which likewise are intended to
block Oracle from completing the Tender Offer.

      3. The Board's refusal to redeem its poison pill or discontinue the other
defensive measures is part of a scheme of entrenchment, intended to defeat
Oracle's bid at any price, that began just hours after Oracle announced the
Tender Offer and continues to the present. On June 7, 2003, less than a day
after Oracle announced the Tender Offer, PeopleSoft's President, CEO and
director Craig Conway was quoted as having stated that he "could imagine no
price or combination of price and other conditions to recommend accepting the
offer to our stockholders" Motivated by their desire to retain their jobs in the
face of an unsolicited offer,

                                       -2-



Conway and other PeopleSoft senior executives committed themselves to a course
of action that has as its goal defeating Oracle's Tender Offer at whatever cost.

      4. PeopleSoft's Board nonetheless abdicated to PeopleSoft's management the
Board's responsibility to oversee the business and affairs of PeopleSoft during
this critical time, and gave management free reign to use whatever means they
considered necessary to thwart the Tender Offer. Taking advantage of the Board's
lack of independence and oversight, management implemented, without proper Board
approval or supervision, the CAP, a novel and improper defensive measure
designed to block Oracle and the Tender Offer. To date, more than $2 billion in
potential cash payments has been promised to customers under the CAP.

      5. None of the first three versions of the CAP - which collectively
account for approximately $900 million of potential payments to customers - was
approved by the Board prior to their use by PeopleSoft's management. The Board
also has failed to evaluate adequately the unsupported assertions of
PeopleSoft's management that "customer concern" justifies continuation of the
CAP.

      6. In addition to its other faults, the CAP is disproportionate to any
perceived legitimate threat to PeopleSoft. The Board also failed to consider
alternatives before permitting PeopleSoft's management to implement the CAP,
including, for example, using the source code escrow provisions already being
used to "protect" the customer, offering the CAP term on a negotiated (as
opposed to across-the-board) basis or creating a multiplier scale that would
approximate anticipated customer harm.

                                       -3-



      7. In addition, the Board has approved and implemented an array of
defensive measures over the past year that, taken together, are disproportionate
to any threat that Peoplesoft may claim is posed by Oracle's non-coercive,
all-cash, all-shares Tender Offer. The Board's illegal defensive measures
include the CAP described above, which was not justified by the level or nature
of customer concerns then being expressed and is a disproportionate reaction to
them; amending the PeopleSoft / JD Edwards merger agreement to deny PeopleSoft
stockholders a vote on that transaction; extending the director nomination
notice provision in the company's bylaws, and then manipulating the meeting date
so as to remove the purported benefits of that extended notice period, all with
the purpose of thwarting a proxy contest; and launching a $550 million stock
repurchase program under which, the company's financial advisors exhorted,

                                   [REDACTED]

The Board's illegal defensive measures were facilitated by a practice of
too-little-too-late disclosures that prevented PeopleSoft stockholders from
challenging any of these measures until they were a fait-accompli, and/or denied
stockholders relevant and timely information to evaluate management's acts.

      8. The Board's refusal to redeem its poison pill and exempt the deal from
Section 203, and its use of the CAP and other defensive measures, are only the
latest moves in a series of actions the Board has taken over the past year to
entrench itself and defeat Oracle's Tender Offer at any cost. Because the Board
continues to breach its fiduciary duties by refusing to allow PeopleSoft
stockholders to decide for themselves whether Oracle's Tender Offer is in their
best interests, the Court should grant Oracle the relief sought in this
complaint.

                                       -4-



                                   THE PARTIES

      9. Plaintiff Oracle is a Delaware corporation with its principal executive
offices in Redwood City, California. Oracle develops, manufactures, markets and
distributes computer software that helps its customers manage and grow their
businesses and manage their operations. Most of Oracle's software products can
be categorized into two broad areas: database technology software and
applications software. Oracle's database software products are highly
sophisticated information management programs that allow users to store,
retrieve, manage and use their business data. Oracle's applications software can
be accessed with standard web browsers and can be used to automate business
processes and to provide business intelligence for a company's marketing, sales,
order management, procurement, supply chain, manufacturing, service and human
resources functions. In addition to computer software products, Oracle offers a
range of consulting, education, support and other services. For customers who
choose not to install its applications on their own hardware, Oracle also offers
outsourcing services that permit web browser access to Oracle's database
technology and software applications hosted at sites that Oracle operates or
manages.

      10. Oracle is both a holder of record and a beneficial owner of PeopleSoft
common stock.

      11. Plaintiff Purchaser is a newly incorporated Delaware corporation and a
wholly owned subsidiary of Oracle with its principal executive offices in
Redwood City, California. To date, Purchaser has engaged in no activities other
than those incident to its formation and the commencement of the Tender Offer.

                                       -5-



      12. Defendant PeopleSoft is a Delaware corporation with its principal
executive offices in Pleasanton, California. According to its most recent Form
10-K, PeopleSoft "designs, develops, markets and supports enterprise application
software products for use throughout large and medium sized organizations
worldwide. [PeopleSoft] provide[s] enterprise application software for customer
relationship management, human capital management, financial management and
supply chain management, each with a range of industry-specific features and
functions. Within each of [PeopleSoft's] application suites, [PeopleSoft]
offers] embedded analytics and portal applications. In addition, [PeopleSoft]
offer[s] a suite of products for application integration and analytic
capability, including portal applications, an integration broker and enterprise
warehouse products".

      13. Defendant Conway is the President, Chief Executive Officer ("CEO") and
a director of PeopleSoft. According to PeopleSoft's Schedule 14D-9 filed June
12, 2003, prior to June 6, 2003, the date Oracle announced its intent to
commence a Tender Offer, the Compensation Committee of the Board (the
"Compensation Committee") approved amendments to Conway's employment agreement
to increase the severance payments to two years' base salary plus target bonus
and to provide for the following accelerated vesting of options and restricted
stock: (i) immediate vesting of all unvested options and restricted stock upon a
change of control of PeopleSoft and (ii) two years' service credit upon
termination of employment without cause or voluntary termination with good
reason.

      14. On August 5, 2003, the Compensation Committee granted Defendant Conway
the option to purchase 375,000 shares at an exercise price of $16.56, which
fully vest upon a change

                                       -6-



of control. On August 20, 2003, PeopleSoft again revisited the compensation
arrangements for the company's senior executives, including Defendant Conway,
and amended their severance arrangements. As explained in PeopleSoft's February
3, 2004 14D-9 with respect to Defendant Conway, if "the vesting of Mr Conway's
restricted stock cannot be accelerated, PeopleSoft will make a cash payment
equal to the fair market value of the unvested restricted stock owned by Mr.
Conway on the date of termination of employment or on the date of the change of
control, as applicable." Thus, the Compensation Committee approved lucrative
increases in compensation arrangements for Defendant Conway and other PeopleSoft
senior executives, including potentially substantial cash payments as a windfall
in the event Oracle's Tender Offer is successful, all at a cost to be borne by
PeopleSoft stockholders.'

      15. Defendants David A. Duffield, Aneel Bhusti, Steven D. Goldby, A.
George Battle, Frank J. Fanzilli, Cyril J. Yansouni and Michael J. Maples are
members of PeopleSoft's Board. The Director Defendants owe the highest fiduciary
duties of loyalty, care and good faith to PeopleSoft's stockholders.

                               FACTUAL BACKGROUND

      16. Since the Tender Offer was launched, PeopleSoft has implemented
various defensive measures and taken many other actions intended to deter
Oracle's bid. On June 9, 2003, three days after the Tender Offer was announced,
PeopleSoft launched the CAP, promising customers between two and five times
their money back if PeopleSoft were acquired and the acquirer took certain
actions thereafter. Later that same week, Defendant Conway and PeopleSoft's
other managers began an aggressive campaign to pressure customers into

                                       -7-



contacting state and federal antitrust regulators, despite the fact that
PeopleSoft itself had proposed a combination of the two companies - to be
managed by PeopleSoft executives - just a year earlier. Customers were asked to
send regulators messages pre-scripted by PeopleSoft, creating the artificial
impression of unified customer competitive concerns.

      17. On June 12, 2003, over a week before any response was required under
SEC rules, PeopleSoft rejected Oracle's bid on grounds that it was likely to run
into the antitrust problems they were actively fomenting, and that it
undervalued the company. On June 16, 2003, PeopleSoft amended the PeopleSoft /
JD Edwards merger agreement to eliminate the need for PeopleSoft stockholders
to vote on the transaction, at a cost to PeopleSoft of an additional $875
million in cash. Two days later, Oracle increased its bid to $19.50 per share,
again requesting a meeting with PeopleSoft's Board to discuss the offer.
PeopleSoft's Board rejected the $19.50 bid the very next day. PeopleSoft closed
its merger with J D Edwards a month later, on July 18, 2003.

      18. In August 2003, PeopleSoft expanded its defensive efforts to its
executive compensation arrangements. On August 5, 2003, Defendant Conway was
granted 375,000 stock options that would vest immediately upon a change of
control, at an exercise price of $16.56 per share - $3.00 below the Oracle bid
they had rejected six weeks earlier. On August 20, 2003 the Board amended
severance arrangements for Defendant Conway and other PeopleSoft senior
executives to provide that they would be entitled to cash payments equal to the
value of restricted stock options vesting upon a change of control.

                                       -8-



      19. Also in August 2003, PeopleSoft announced in its 10-Q for the period
ending June 30, 2003 that liabilities under the CAP totaled $391 million for the
first month of the program alone, and that the CAP had been extended
"indefinitely." Meanwhile, PeopleSoft's management was secretly developing a new
and enhanced CAP (the "Version 2 CAP") designed to be more onerous for an
acquirer, and that could be triggered by a successful proxy contest, not just a
change of control. PeopleSoft's management rolled out the Version 2 CAP to
customers on September 12, 2003, although it did not publicly announce it until
October 27, 2003 - ten days after the program had expired. Realizing that
inclusion of the proxy contest trigger jeopardized their ability to recognize
revenues under the Version 2 CAP contracts, on September 30, 2003 PeopleSoft
rolled out a third version of the CAP (the "Version 3 CAP"), which specifies
that a successful proxy must be followed by a change of control, but is
otherwise identical to the Version 2 CAP. PeopleSoft's Board was not given the
terms of either the Version 2 or Version 3 CAP until after they expired.

      20. On September 4, 2003, the Board initiated a $350 million stock
repurchase program, having been told by their financial advisors that they could
"repurchase shares first from those most likely to tender." On November 5, 2003,
the Board amended PeopleSoft's bylaws to extend the minimum notice period for
stockholder nominations to the Board from 20 days to at least 120 days before
the anniversary of the prior year's stockholder meeting, on grounds that
stockholders needed the maximum possible time to consider any insurgent slate.
On January 30, 2004, two weeks after Oracle met the accelerated deadline,
PeopleSoft

                                       -9-



announced that the 2004 annual stockholders meeting would be held on March 25,
2003 - two full months earlier than in any of the previous seven years.

      21. On November 17, 2003, the Board responded to preliminary injunction
motions filed by Oracle and the Shareholder Plaintiffs challenging the Version 2
and 3 CAPs by implementing the Version 4 CAP. By March 31, 2003, total
liabilities under the Version 4 CAP would reach $1.084 billion. By that date,
$1.9 billion in potential customer payments had been promised under the four CAP
terms. This represents approximately one third of PeopleSoft's market
capitalization, and four times more than the company's current cash on hand as
of March 31, 2004.

      22. On January 12, 2004, PeopleSoft announced that the Board had extended
the stock repurchase program, authorizing PeopleSoft to purchase an additional
$200 million worth of PeopleSoft stock, even though the Board's financial
advisors had raised a concern, when the first Stock Repurchase Program was
initiated in September, that a "[l]arger repurchase would reduce cash balance
and flexibility going forward".

      23. On February 4, 2004, Oracle announced that it had increased its bid to
$26 per share, representing a 19% premium over the previous day's closing price
of $21.89. PeopleSoft rejected this bid on February 7, 2004. Over the course of
the next several weeks, PeopleSoft's stock dropped to $16.53 per share,
responding to the company's poor financial results. Still, PeopleSoft's Board
refused to meet with Oracle to discuss the bid, even though the bid at one point
represented a 57% premium to PeopleSoft stockholders. On May 14, 2004, in
response to changes in market conditions and in PeopleSoft's market valuation,
Oracle revised its bid to $21,

                                      -10-



reflecting a 21% premium over that day's closing price of $17.30. On May 26,
2004, the Board rejected this bid as well and persisted in its refusal to
negotiate with Oracle, or to redeem its structural defensive measures.
PeopleSoft's poison pill and the application of DGCL Section 203 to the Tender
Offer.

I.    PEOPLESOFT'S RIGHTS PLAN NO LONGER SERVES A LEGITIMATE PURPOSE AND SHOULD
      BE REDEEMED.

      A.    THE BOARD FAILED TO MAKE A REASONABLE AND GOOD FAITH INVESTIGATION
            OF ORACLE'S OFFER.

      24. On June 6, 2003, the day that Oracle announced its intention to launch
the Tender Offer, Defendant Conway categorically rejected it, sight unseen. Two
days later, the Board established a Transaction Committee to review the Tender
Offer.

      25. The Transaction Committee, however, has authority only to review and
make non-binding recommendations to the Board concerning the Tender Offer. More
importantly, the Transaction Committee has only received advice of legal
counsel. and investment bankers retained by and beholden to PeopleSoft's
management. The Transaction Committee has no truly independent legal or
financial advisors.

      26. In addition, the Transaction Committee has never met independently to
consider Oracle's Tender Offer. Rather, minutes of PeopleSoft's Board meetings
indicate

                                   [REDACTED]

                                    This includes Defendant Duffield,
PeopleSoft's founder and a current director, who is not independent from
PeopleSoft, Conway or the rest of

                                      -11-



PeopleSoft's management. Furthermore, upon information and belief, Defendant
Duffield's son, daughter-in-law and brother are or were employed by PeopleSoft.

      27. The Transaction Committee itself also is critically flawed because one
or more of the directors serving on the Committee is interested in the outcome
of the Tender Offer. For example, Transaction Committee member Michael J. Maples
is a former Executive Vice President of Microsoft Corporation, and upon
information and belief, he continues to serve as a paid Microsoft consultant.
Microsoft is a fierce competitor, and has an active interest in the outcome of
Oracle's Tender Offer. Indeed, Microsoft announced on June 7, 2004 that until
just a few months ago, it was engaged in merger talks with SAP, another major
player in this highly competitive business.

      28. Consistent with Conway's categorical rejection of the offer, and the
Board's and the Transaction Committee's lack of independence and refusal to
consider fairly the Tender Offer, at no time has PeopleSoft's Board agreed to
meet with Oracle to discuss its offer, even when Oracle's bid was as high as
$9.50 (or a 57% premium) above PeopleSoft's trading price. Over the past year,
PeopleSoft's Board has rejected four separate bids from Oracle, including
Oracle's current bid of $21, each time claiming the offer was too low.

      29. Yet, Oracle's $21 per share bid represented a 21% premium above
PeopleSoft's stock price. Presentations made to PeopleSoft's Board by its own
financial advisors in connection with prior bids from Oracle reveal that

                                   [REDACTED]

PeopleSoft's bid for J.D. Edwards involved a 19.4% premium.

                                      -12-



      30. The Board's continued refusal either to negotiate with Oracle, or to
allow PeopleSoft stockholders to consider for themselves a non-coercive,
all-cash, all-shares offer at a substantial premium, demonstrates the Board's
failure to make a reasonable and good faith investigation of the Tender Offer.
The Board has not shown and can not show that the Tender Offer poses a threat to
PeopleSoft's corporate policy or effectiveness sufficient to deny stockholders
the opportunity to tender their shares and receive the benefits of the Tender
Offer.

      B.    THE PASSAGE OF TIME HAS REMOVED ANY LEGITIMATE PURPOSE FOR
            PEOPLESOFT'S POISON PILL.

      31. In response to Oracle's offer, PeopleSoft's Board has refused to take
down its preexisting structural defenses, including its Rights Plan and the
application of Delaware's antitakeover statute ("Section 203"). Considering that
a full year has passed since Oracle launched the Tender Offer, the premium the
Tender Offer represents and other facts, the Board's refusal to redeem the
poison pill and exempt the Tender Offer from Section 203 represents a breach of
the Board's fiduciary duties to PeopleSoft's stockholders.

      32. PeopleSoft's Rights Plan was adopted on or about February 15, 1995 and
amended in 1997. The Rights are distributed and become exercisable for one
one-thousandth share of PeopleSoft's Series A Participating Preferred Stock (the
"Series A Preferred") at a price of $190 on the close of business ten days after
the earlier of (i) it is announced that any person has acquired beneficial
ownership of 20% or more of PeopleSoft's common stock (an "Acquiring Person"),
or (ii) the publication of a tender or exchange offer which, if successful,
would result in the beneficial acquisition by any person of 20% or more of
PeopleSoft's common stock (the earlier of (i) and (ii) being referred to as the
"Distribution Date"). As the Rights Plan

                                      -13-



permits, on June 11, 2003 the Board postponed the Distribution Date, which
otherwise would have been triggered by the Tender Offer, until just before the
consummation of a transaction through which a person would become an Acquiring
Person. The Rights expire on February 15, 2005, unless earlier redeemed or
exchanged by PeopleSoft.

      33. Under the Rights Plan, if and when Oracle becomes an Acquiring Person,
all Rights other than those held by Oracle will "flip-in" and each right becomes
exercisable for shares of PeopleSoft common stock equivalent in value to twice
the exercise price of the Right. Thus, for the exercise price of $190, the
holder of a Right may purchase PeopleSoft common stock having a market value of
$380. If and when PeopleSoft engages in a merger or a sale of 50% or more of its
assets, as is contemplated under the Tender Offer, the Rights "flip-over" and
become exercisable for shares of the acquirer's common stock at the same deep
discount price of two for the price of one.

      34. The PeopleSoft Board may redeem the Rights, at a redemption price of
$.01 per Right, any time prior to the close of business on the earlier of
(i) the tenth day following the date of public announcement of the fact that an
Acquiring Person has become such, or (ii) February 15, 2005.

      35. Oracle's acceptance (via Purchaser) of shares tendered pursuant to the
Tender Offer will result in it becoming an Acquiring Person, which, in turn,
will make the Rights exercisable for shares of PeopleSoft's common stock at a
discount of 50% of their market value (rendering the Tender Offer economically
infeasible for Oracle to accomplish) and will deprive

                                      -14-



PeopleSoft's stockholders of the ability to tender their shares unless
PeopleSoft's Board redeems the Rights or otherwise amends the Rights Plan to
render it inapplicable to the Tender Offer.

      36. The Board also has refused to exempt the Tender Offer from the
application of Section 203. Section 203 would delay the Proposed Merger for at
least three years. During this period, substantial benefits of the Proposed
Acquisition would be forever lost, and any number of events could occur that
would prevent the Proposed Merger altogether.

      37. PeopleSoft's poison pill no longer serves a valid purpose. In the year
since Oracle launched its Tender Offer, PeopleSoft, Oracle, industry and
financial analysts, the media and other commentators have provided PeopleSoft's
stockholders with considerable information and analysis concerning both the
Tender Offer and PeopleSoft's current financial condition and future prospects.
As a result, PeopleSoft's stockholders already have the information (much of
which was reviewed by the SEC) they need to evaluate the Tender Offer, and have
had more than a year to make that evaluation.

      38. Likewise, the intervening twelve months since the Tender Offer was
announced have given PeopleSoft's Board ample "breathing room" to explore
alternatives, examine the merits of the Tender Offer and communicate their views
to PeopleSoft's stockholders. To the extent PeopleSoft's stockholders still do
not have enough information about PeopleSoft's prospects as a stand-alone
company, it is because PeopleSoft's Board and management have failed to fulfill
their obligation to provide a coherent statement of its expectations about the
future value of the company, despite having had a full year to do so.

                                      -15-



      39. The PeopleSoft Board's continued refusal to redeem PeopleSoft's poison
pill or to exempt the Tender Offer from Section 203, both individually and in
concert, are disproportionate reactions to Oracle's Tender Offer and therefore
constitute improper defensive measures. Accordingly, the Board's continued
refusal to remove PeopleSoft's structural defensive measures is a breach of its
fiduciary duties to PeopleSoft's stockholders, and furthermore threatens to deny
Oracle the benefits of the proposed combination.

II.   PEOPLESOFT'S CAP IS AN IMPROPER DEFENSIVE MEASURE.

      40. On June 9, 2003, three days after Oracle announced its Tender Offer,
PeopleSoft's management began sending form letters to customers announcing the
first in what would become a series of customer rebate programs that PeopleSoft
called the "customer assurance program" (the "CAP"). The CAP purports to provide
customers refunds of between two and five times any licensing and maintenance
fees paid up front under software licensing agreements signed while Oracle's
offer is pending if (i) PeopleSoft is acquired within a certain amount of time
after the contract date; and (ii) the acquirer takes (or fails to take) certain
actions with respect to the products licensed by the customer pursuant to that
contract. Because the higher CAP multipliers are reserved for larger deals, CAP
liabilities accrue at a rate that is independent of increases in licensing
revenues. Potential CAP liability for single contracts executed thus far by
PeopleSoft range as high as nearly $80 million, although there is no limit to
the total possible payment that can be given under a single contract. Unlike a
traditional poison pill, payment promises under the CAP cannot be redeemed by
the Board, and would affect all acquirers over a two year period, even in a
"friendly" transaction.

                                      -16-



      41. PeopleSoft waited to publicly announce the CAP until July 2, 2003, two
days after it had initially expired, denying stockholders any opportunity to
object until after a $354 million potential liability had accrued under the
Version 1 CAP. When PeopleSoft filed its 10-Q for the period ending June 30,
2003 (on August 14, 2003), it revised that potential liability figure upwards to
$391 million for the same period, and announced that the CAP had been extended
indefinitely. The Version 1 CAP payments could only be triggered if PeopleSoft
is acquired within one year of the contract date and then the acquirer takes one
of the actions described above within two years of the contract's date.

      42. On October 27, 2003, PeopleSoft announced that it had introduced a
second CAP offer (the "Version 2 CAP"), this time 10 days after it had expired.
According to PeopleSoft's Supplemental Response and Objections to the Joint
Interrogatories, a total of $170 million has been promised under the Version 2
CAP. Launched by PeopleSoft's management on September 12, 2003, the Version 2
CAP contains several "enhancements" to the Version 1 CAP language. First, the
time horizons were expanded from one and two years to two and four years,
respectively. Second, the business events covered were revised in ways that
substantially increased the probability that they would be triggered, even if
only unintentionally, by an acquirer. For example, a customer could argue that
these terms were triggered if an acquirer were to immaterially delay the timing
of product updates. After this offer had officially expired but months before
its actual use was finally discontinued (at the end of December 2003), one or
more of PeopleSoft's directors acknowledged

                                   [REDACTED]

                                      -17-



      43. Third, the definition of an "acquisition" was also changed in the
Version 2 CAP to include the election of a majority of new directors, not
approved by the incumbent PeopleSoft Board (the "Board Composition Trigger").
This expansion was problematic in two respects. First, the Board Composition
Trigger raised the specter that the Version 2 CAP could affect stockholders'
voting decisions in a contested proxy contest, as the future value of the
company under a new slate of directors would be diminished by the risk-weighted
value of the Version 2 CAP payments promised to customers. PeopleSoft did not
publicly acknowledge the potential effect of the CAP on the stockholder
franchise until a March 4, 2004 amendment to its Schedule 14D-9 filings.

      44. Second, the Board Composition Trigger affects the revenue recognition
treatment for revenues associated with the Version 2 CAP contracts. According to
the company's own analysis,

                                   [REDACTED]

            However, under the Version 2 CAP definition of an "acquisition," CAP
liabilities could be triggered upon a successful proxy contest for the company,
regardless of whether there was a subsequent change of control. Thus, the
revenues attributable to these contracts should have been deferred. Because
PeopleSoft chose to recognize - rather than defer - this revenue, PeopleSoft's
third and fourth quarter 2003 revenues have been inflated by the amount of the
Version 2 CAP contract revenues. PeopleSoft added a proviso to the Board
Composition Trigger for Version 2 CAP contracts dated after September 30, 2003,
providing that the company must

                                      -18-



be acquired following a successful proxy contest for the term to take effect,
calling this a new, Version 3 CAP.

      45. On October 10, 2003 - 10 days after the close of the quarter in which
the licensing revenues under those contracts was booked - the company began
sending letters to customers whose contracts contained the Version 2 CAP,
attempting to unilaterally amend the contracts after the fact to include the new
provision

                                   [REDACTED]

At a minimum, PeopleSoft should have deferred these revenues unless and until
each customer accepted the unilateral amendment.

      46. Despite PeopleSoft's repeated assertions to the contrary, the company
did not seek SEC pre-clearance for their accounting treatment of the Version 2
CAP. The SEC's June review of the CAP was expressly limited to the particular
language of the Version 1 CAP as provided to the SEC (which did not include
substantial variations included in over 250 contracts). PeopleSoft first
discussed the accounting treatment for their Version 2 CAP months after those
revenues bad been booked by PeopleSoft and only after the SEC inquired about
their accounting treatment. This process eventually resulted in PeopleSoft
revising its disclosures about the Version 2 CAP.

      47. In total, management signed contracts providing for a total of $170
million of potential liability under the problematic Version 2 CAP, with an
additional $33 million worth of potential liability under contracts containing
the Version 3 CAP term

                                      -19-




      48. On November 5 and 10, 2003, respectively, the stockholder plaintiffs
and Oracle filed motions for a preliminary injunction against continued use of
the Version 2 CAP One week later, PeopleSoft implemented a new, Version 4 CAP.
The Version 4 CAP retains the extended time horizons and many of the enhanced
business provisions of the Version 2 CAP.

      49. PeopleSoft announced the Version 4 CAP just hours before the November
19, 2003 Delaware Court hearing on Oracle's motion challenging the CAP; as of
March 31, 2003 it had promised customers potential payments totaling $1.084
billion under this term

      A.  THE BOARD DID NOT MAKE A GOOD FAITH AND REASONABLE INVESTIGATION OF
          THE TENDER OFFER'S PURPORTED THREAT TO CUSTOMERS.

      50. PeopleSoft's Board did not make a reasonable and good faith
investigation of the claims made by PeopleSoft's management that the Tender
Offer posed a threat to PeopleSoft's customer base

                                   [REDACTED]

      51. PeopleSoft's documents from that period reflect a campaign to generate
rather than mitigate fear among its customer base, in an effort to mobilize
customers against Oracle's offer

                                   [REDACTED]

      PeopleSoft's management spread the myth that Oracle would not support
PeopleSoft products, when in fact Oracle had stated, in its initial June 6, 2003
conference call announcing the offer and subsequently, that it would enhance and
continue support for the PeopleSoft products.

                                      -20-



      52. PeopleSoft has not identified a single customer who proposed or
demanded that any particular language, whether linked to a change of control or
otherwise, be added to their contracts prior to PeopleSoft extending the CAP
offer. PeopleSoft CFO Kevin Parker, who claims responsibility for the original
idea to add a CAP clause to customer contracts, has testified that

                                   [REDACTED]

On June 9, 2003, of its own accord, PeopleSoft made the first of several mass
mailings of offer letters to customers describing the Version 1 CAP.

      53. By the time the Board first made a serious review of the CAP, in the
end of October and early November 2003, PeopleSoft had already reported two
quarters of purportedly strong financial performance, meeting or exceeding
analyst expectations, while the Tender Offer was pending. On September 4, 2003
Conway announced at the company's Analyst Day conference, "I don't think the
Oracle bid is a current issue. It's a movie that's been playing a long time. I
think people have lost interest in it" He added, "[t]he last remaining customers
whose business decisions were being delayed have actually completed their sales
and completed their orders." Recognizing that this statement undermined
PeopleSoft's arguments before this Court (as well as in a pending action brought
by PeopleSoft in California state court), PeopleSoft replaced the above
statement in the version of this transcript that PeopleSoft filed with the SEC
with the following. "Oracle's tactics have created concern among many users, and
that's a problem for us. Fortunately, we've been able to overcome much of it and
we expect that we will continue to be able to do so." Remarkably, PeopleSoft's
public filing notes that this language

                                      -21-



"ha[s] been corrected to reflect the intended meaning of the speaker " (emphasis
added). Two weeks later, Conway said that "[t]he actual impact [of the Tender
Offer] has been remarkably beneficial for PeopleSoft." Yet, despite these
indications, on November 17, 2003, the Board approved a whole new CAP offer,
under which $1.084 billion in customer payments had been awarded by March 31,
2004.

      B.    THE CAP WAS IMPLEMENTED TO HELP REPEL ORACLE'S OFFER.

      54.                          [REDACTED]

      55.   The CAP serves this purpose in two ways. First, as PeopleSoft
asserted to the SEC in its June 16, 2003 letter seeking pre-clearance for their
intended accounting treatment of the CAP terms, the CAP "acts as an economic
disincentive in the event of a hostile takeover" Others of PeopleSoft's internal
documents are more blunt.

                                   [REDACTED]

                                      -22-



      56. Other evidence of the CAP'S true purpose includes the fact that the
CAP, as a matter of policy, has been given to all PeopleSoft customers signing
new licensing deals since the Tender Offer's initiation, whether or not such
extraordinary terms were commercially reasonable and necessary (or even
requested by customers). Twenty percent of the total CAP liability as of March
31, 2004 is attributable to just eight customers, whose individual payments
under the CAP range from $37 million to almost $80 million. Notably, not one of
these customers was identified in PeopleSoft's Supplemental Response and
Objections to the Joint Interrogatories as having advised PeopleSoft "that
inclusion of the [CAP] was essential in order for them to purchase PeopleSoft
products."

      57. Second, PeopleSoft's management has used the CAP to aid in their
efforts to defeat the Tender Offer by artificially increasing the company's
reported revenues in the critical quarters after the Tender Offer was first
announced, making Oracle's bid look artificially less attractive to
stockholders.

      58. PeopleSoft's internal documents and communications reveal that

                                   [REDACTED]

Customers were exhorted to make additional, unplanned or accelerated software
purchases, as a way to "protect" their previous software investments. The CAP
facilitated this effort on behalf of PeopleSoft's management, as it created an
incentive for customers to sign up while Oracle's offer was pending (which was
initially anticipated to be for only a limited period of time). Conway stated
both publicly and privately

                                      -23-



that some customers did in fact "accelerate" their purchases, to "help"
PeopleSoft in its efforts to repel Oracle's offer.

      59. All the while, PeopleSoft's management assured industry analysts that
the CAP was having no effect on the company's pipeline of future deals, despite
these indications that the CAP was used as an incentive for customers to move
forward planned purchases. In addition, PeopleSoft has inflated its revenues in
the third and fourth quarters of 2003 by recognizing up front the licensing fees
attributable to contracts containing the Version 2 CAP term, as described above.

      C.  THE CAP IS DISPROPORTIONATE TO ANY PURPORTED CUSTOMER THREAT.

      60. PeopleSoft's Board also failed to consider whether customers'
purported concerns regarding the Tender Offer could be satisfied through means
other than the particular CAP term proposed by PeopleSoft's management, or
whether the CAP would be proportional to any legitimate threat they believed
Oracle's Tender Offer posed.

      61. First, the Board did not investigate whether existing contractual
provisions already provided adequate protection to customers. For example, many
of PeopleSoft's contracts include source code escrow provisions, under which the
customer is entitled to access the source code so that it can make other
arrangements in the event that PeopleSoft (or its successor) announces its
inability or unwillingness to continue to support the products.

      62. Second, the Board did not consider options for protecting customers
against changes to the support, development or marketing of the licensed
products regardless of whether there was a change of control. PeopleSoft claims
in its Response and Objections to the Joint

                                      -24-



Interrogatories that the CAP was limited to take effect only after a change of
control because of requirements under the revenue recognition rules applicable
to software. In fact, on one or more occasions PeopleSoft gave a customer
protections that were not dependent on a change of control and yet, upon
information and belief, still accounted for the revenue in the quarter that the
contract was signed.

      63. Third, having decided to allow PeopleSoft's management to proceed with
the CAP, at no time did the Board provide any guidance to them regarding, or
place any ceilings on, the extent of promises that it would be reasonable to
make under the CAP. It was not until November 17, 2003 - once over $800 million
in liability had accrued under two separate CAP offers and after Oracle filed
its motion challenging the CAPs - that the Board first received an analysis from
its financial advisors of the potential impact of the CAP on an acquirer, or of
the per share cost of the program. Nor was such an analysis prepared for the
Board's subsequent consideration of whether to extend the CAP through to the
present.

      64. Similarly, the Board did not fully consider whether the CAP could be
used sparingly, in response to specific customer requests, rather than on an
across-the-board basis. Software licensing agreements for enterprise software
are heavily negotiated contracts, containing dozens of carefully crafted
provisions that are modified to meet the demands of the customer. Upon
information and belief, terms with a financial impact are subject to sales
approval processes, including at the highest level CFO Kevin Parker, to ensure
that the company's long term financial health is protected. Yet, as early as
June 11, 2003, the Board decided that the CAP term - under which penalty
payments in one or more individual contracts

                                      -25-



could exceed $80 million - would be given to all customers, whether or not they
requested it and with no executive approval required.

      65. Fourth, the Board failed to consider whether a CAP-like provision
could be implemented in a way that would have a less severe potential impact on
an acquirer and therefore on PeopleSoft stockholders. For example, Parker
testified that,

                                   [REDACTED]

At no time did the Board request or receive an analysis of the correlation
between the actual harm a customer might suffer should one of the acts covered
by the CAP occur and the size of the customer's licensing deal.

                                   [REDACTED]

      66. The CAP is a disproportionate response to the threat purportedly posed
by Oracle's offer because the CAP gives customers a remedy that far exceeds any
actual harm they may suffer, and creates obligations for an acquirer that
PeopleSoft - under current management - does not have.

      67. The CAP terms, or "penalty provisions" as they were referred to in an
early draft of the accounting pre-clearance letter PeopleSoft sent to the SEC on
June 16, 2003, were not

                                      -26-



tailored to correlate to any actual harm that customers would suffer. For
example, the size of the CAP payments depends only on the size of the licensing
deal of which it was a part, not on the severity of the triggering event's
purported impact on the customer's business. Thus, a customer's CAP payment
would be the same in the event of an immaterial delay in the timing of a product
update's release as in the event of a permanent end to all product support
services.

      68. Through the CAP, PeopleSoft also gave customers a remedy that was
substantially more generous than the limitations of liability terms the company
had previously negotiated with these very same customers and thus reflect the
customer's actual commercial demands. These terms govern in the event of any
damages arising under these same contracts. Even among the eight customers whose
CAP promises collectively represent twenty percent of the total potential CAP
liability - all of whom received multipliers of "5x" - PeopleSoft's liability
(in the absence of a change of control) is limited under the associated master
Software License and Services Agreement to which the CAP-enhanced licensing
schedule is an attachment, to one or two (or, in one case, three) times the
total fees attributable to the particular product or service with respect to
which the claim arises.

      69. The CAP terms also create substantial obligations for an acquirer that
PeopleSoft currently does not have and has specifically refused to take on

                                   [REDACTED]

                                      -27-



                                   [REDACTED]

      70. Moreover, the CAP terms purport to penalize an acquirer for making
business decisions that PeopleSoft, under current management, is free to make.
For example, Parker stated in a June 2, 2004 deposition that

                                   [REDACTED]

      71. When it decided, on November 17, 2003, to implement the current,
Version 4 CAP, the Board relied on assertions by PeopleSoft Executive Vice
President for Products & Technology Ram Gupta that the CAP terms place no more
obligations on an acquirer than those PeopleSoft has itself and therefore are,
by implication, reasonable. The Board was not entitled to rely on these
assertions. PeopleSoft's Board consists of savvy individuals with substantial
experience in the software industry and, in some cases, company insiders with
substantial knowledge of PeopleSoft itself. Minutes of PeopleSoft Board meetings
cite this experience as a

                                      -28-



basis for key decisions made by the Board with respect to the Tender Offer. The
Board therefore cannot reasonably rely on unsupportable and inaccurate
assertions by PeopleSoft's management to justify the CAP.

      D.    THE BOARD NEVER ESTABLISHED APPROPRIATE MONITORING AND REPORTING
            CONTROLS FOR THE CAP.

      72. PeopleSoft's Board failed to provide appropriate approvals or
supervision over management's activities with respect to the CAP. The Board's
inaction amounted to an abdication of its responsibility to manage the business
and affairs of the company particularly in light of the Tender Offer, to
management they knew to be adamantly opposed to Oracle's offer. This calls into
serious question whether the Board was in fact acting in good faith with respect
to the Tender Offer more generally.

      73. Parker has testified in deposition that [REDACTED]

            However, none of the Board meeting minutes produced to date include
any Board approval for the Version 1 CAP, which had been first extended to
customers two days previously. Moreover, the Board never reviewed a copy of the
actual language of the Version 1 CAP term. The only indication that the Board
was aware of the language being provided to customers is an incomplete summary
of the offer's terms that was sent to the Board on June 18, 2003, in the form of
what Parker testified in deposition, as PeopleSoft's corporate representative,

                                   [REDACTED]

There is no record of PeopleSoft's Board having met to consider extending or
reinstating the program at any time during the summer of 2003. By the time
management abandoned the Version 1 CAP on September 30, 2003, the total payments
under this program had mounted to $670 million.

                                      -29-



      74. Likewise, there is no indication that either the Audit Committee or
the company's external auditors reviewed the accounting treatment for the
Version 2 CAP, including management's proposal to send "clarifying" letters to
customers, at any time, despite the ramifications of the Board Composition
Trigger - and management's attempted after-the-fact "fix" of the Version 2 CAP
terms - for the company's financial statements and critical accounting policies.

      75. Despite the substantial and problematic changes introduced with the
Version 2 and 3 CAPs, neither term was ever approved by PeopleSoft's Board.
PeopleSoft contends that these terms were approved at an August 5, 2003 Board
meeting. However, the minutes of that meeting state only that

                                   [REDACTED]

      76. The Board appears to have been unaware that a new CAP term had been
implemented. Following the August 5, 2003 Board meeting at which "enhancements"
to the CAP were being considered, the next reference to the CAP in any Board
minutes is not until the October 21, 2003 Audit Committee meeting. At that
meeting, Parker informed the Board that

                                      -30-



                                   [REDACTED]

      What followed was a flurry of management activity, including efforts to
prepare a presentation on the CAP for the first time. On October 29, 2003, two
days after the Version 2 CAP was publicly announced for the first time in a
14D-9 filing, PeopleSoft General Counsel Anne Jordan emailed the Board members
to request a special meeting to review and approve the CAP so that pending deals
could be closed. For the first time, the language of the Version 2 CAP term was
attached, along with a comparison with the language of the Version 1 CAP.

      77. Minutes of the October 31, 2003 Board meeting indicate that the Board
approved

                                   [REDACTED]

      This is the first reference in the Board minutes to actual Board approval
of a CAP term. Documents concerning this meeting indicate that

                                   [REDACTED]

      Thus, of the $880 million of potential payments promised to customers
prior to Oracle and the stockholder plaintiffs filing their motions for
preliminary injunction in the Delaware Chancery Court, only $7 million was
promised pursuant to terms that had actually been approved by PeopleSoft's
Board.

                                      -31-



      78. On November 5, 2003, the Board again met to consider the CAP and for
the first time received draft language for revising the CAP. That day, the
stockholder plaintiffs in an action pending against PeopleSoft in this Court
filed a motion for a preliminary injunction against the Version 2 CAP, which was
followed by a similar motion from Oracle on November 10, 2003. In response,
PeopleSoft's Board met November 17, 2003 and developed a new, Version 4 CAP
term. For the first time, the Board received analyses by the company's financial
advisors of the potential impact the offer would have on an acquirer. Also for
the first time, the November 17 Board minutes indicate that

                                   [REDACTED]

      This record stands in sharp contrast to the elusive references PeopleSoft
relies upon for its assertion that the versions of the CAP which predated
Oracle's preliminary injunction motion were likewise "approved" by the Board
prior to their usage.

      79. As with the development of the Version 2 CAP, however, the company's
independent auditors were not included in deliberations regarding the Version 4
CAP term. Rather, the auditors were only given a few hours to review the new
term, after the Board had finished its deliberations.

                                   [REDACTED]

                                      -32-



                                   [REDACTED]

      80. The Board likewise failed to establish appropriate monitoring and
controls over the CAP's implementation. By June 18, 2003, the Board should at
least have been aware of the broad brush strokes of the CAP, including that the
multiplier scale would cause liability under the CAP to mount rapidly,
outstripping growth in the associated licensing fees. Accordingly, the Board
should have established strict monitoring, reporting and controls on the CAP's
implementation and extension.

      81. However, in the year that the CAP has run, upon information and
belief, the Board never set a limit on the total liability that would be
acceptable under one or all CAP contracts, or even requested regular reporting
from PeopleSoft's management on the potential liability accruing under the CAP.
In fact, PeopleSoft CFO Kevin Parker has testified that

                                   [REDACTED]

                                      -33-


      82. PeopleSoft's management responded by introducing new language to the
CAP terms as they were provided to hundreds of customers, without Board review
or approval. The most obvious example of this was the introduction of the
Version 2 CAP, discussed above. Before that, however, PeopleSoft's management
had added language to the Version 1 CAP term as it appears in literally hundreds
of contracts that purports to bind a successor not only to continue licensing
PeopleSoft's current line of products to new customers, but also "the successors
to [the licensed products that] would have been developed by PeopleSoft prior to
an acquisition or merger." This new language was not included in the summary of
the Version 1 CAP term, in the form of a "draft transmittal letter to customers"
that was sent to PeopleSoft's Board on June 18, 2003. By June 18, PeopleSoft had
been giving customers the new language for at least a week.

      83. PeopleSoft's management also liberally gave individual customers
exceptions under the CAP, without regard to the potential impact of those
exceptions on an acquirer or stockholders, and without Board oversight.

                                   [REDACTED]

                                      -34-



                                   [REDACTED]

      84. On one or more other occasions, PeopleSoft's management even agreed to
back date CAP contracts, in some cases to before the date Oracle's Tender Offer
was announced. One or more customers received a CAP in a contract that was
back-dated across the quarter line, which, if recognized as revenue in the
quarter prior to the actual contract date, would be impermissible under the
applicable accounting standards.

      85. Meanwhile, PeopleSoft's management misled regulators and the investing
public regarding the nature and effect of the CAP on one or more occasions with
the Board's acquiescence.

                                   [REDACTED]

Just two weeks earlier, on June 16, 2003, in a letter to the SEC office of the
chief accountant seeking pre-clearance for the revenue recognition treatment of
CAP contracts listed as one of two business purposes for the CAP that it "acts
as an economic disincentive in the event of a hostile takeover"

                                   [REDACTED]

      86. The Board's abdication to PeopleSoft's management of its
responsibility to oversee the CAP reveals its lack of good faith in its response
to the Tender Offer.

                                      -35-



                                    COUNT I

            (BREACH OF FIDUCIARY DUTY: DISLOYALTY AND ENTRENCHMENT)

      87. Plaintiffs repeat and reallege each and every allegation set forth in
paragraphs 1 through 86 as if fully set forth herein.

      88. The Director Defendants owe PeopleSoft's stockholders, including
Oracle, the highest duties of care, loyalty and good faith, and have failed from
the outset to meet those duties.

      89. The Tender Offer is non-coercive and non-discriminatory, it is fair to
PeopleSoft stockholders, it poses no threat to PeopleSoft's corporate policy and
effectiveness, and it represents a substantial premium over the market price of
PeopleSoft's stock.

      90. The Director Defendants have and are continuing to breach the
fiduciary duties owed PeopleSoft's stockholders by refusing to accept the Tender
Offer, or otherwise negotiate with respect to it, by adopting or allowing
PeopleSoft's management to adopt the defensive measures detailed above, and by
failing to remove the barriers of its Rights Plan and Section 203. Such efforts
to frustrate the completion of the Tender Offer serve no legitimate purpose, and
are not reasonable responses to the Proposed Acquisition, which poses no threat
to the interests of PeopleSoft's stockholders or to PeopleSoft's corporate
policy and effectiveness. As such, the actions of the Director Defendants are in
breach of the fiduciary duties the Director Defendants owe to PeopleSoft's
stockholders under applicable Delaware law.

      91. The unlawful actions of the Director Defendants are preventing
PeopleSoft's stockholders from receiving the benefits of the Proposed
Acquisition and are thereby causing and will cause PeopleSoft's stockholders
irreparable harm.

                                      -36-



      92. The actions complained of herein further threaten to deprive Oracle of
the opportunity to proceed with its Tender Offer and the Proposed Acquisition.
The loss of these opportunities, which are unique, constitutes irreparable harm
which can not adequately be remedied later. Unless PeopleSoft's Board is
restrained by this Court, the substantial benefits of the Proposed Acquisition
may be forever lost.

      93. Plaintiffs have no adequate remedy at law.

                                    COUNT II

                    (BREACH OF FIDUCIARY DUTY: FAILURE TO ACT
                WITH DUE CARE AND ABDICATION OF RESPONSIBILITIES)

      94. Plaintiffs repeat and reallege each and every allegation set forth in
paragraphs 1 through 93 as if fully set forth herein.

      95. The Director Defendants prejudged and summarily rejected the Tender
Offer without conducting a reasonable and good faith investigation of the merits
of the Proposed Acquisition and any threat purportedly posed by it to corporate
policy and effectiveness. The Director Defendants' failure to adequately
consider the Tender Offer constituted gross negligence and an abandonment of
their fiduciary obligations.

      96. The Director Defendants adopted defensive measures that individually
and collectively are neither reasonable nor proportionate to any threat posed by
the Tender Offer and constitute an illegal scheme of entrenchment.

      97. Furthermore, the Director Defendants allowed PeopleSoft's management
to implement one or more improper defensive measures without proper Board
approvals, monitoring or supervision.

                                      -37-



      98. The Director Defendants' actions are in breach of the fiduciary duties
owed to PeopleSoft's stockholders, including Oracle, under applicable Delaware
law as Defendants' conduct is causing irreparable harm to Oracle.

      99. Plaintiffs have no adequate remedy at law, as Defendants' conduct is
causing irreparable harm to Oracle.

                                    COUNT III

                    (VIOLATION OF SECTION 141 AND ULTRA VIRES)

      100. Plaintiffs repeat and reallege each and every allegation set forth in
paragraphs 1 through 99 as if fully set forth herein.

      101. Under 8 Del C Section 141 ("Section 141"), the board of directors has
a continuing duty to oversee, manage and direct the business and affairs of the
corporation as circumstances evolve. By implementing the CAP Offers, the
Director Defendants have impermissibly restricted their own and any future
board's authority to manage PeopleSoft under Section 141 and the board's
concomitant fiduciary duties.

      102. Plaintiffs have no adequate remedy at law, as Defendants' conduct is
causing irreparable harm to Oracle.

                                PRAYER FOR RELIEF

      WHEREFORE, Plaintiffs respectfully request that this Court:

            (a) compel PeopleSoft and its Director Defendants to redeem the
Rights or to render the Rights Plan inapplicable to the Proposed Acquisition;

            (b) compel the Director Defendants to approve the Proposed
Acquisition for

                                      -38-



purposes of Section 203 and enjoin them from taking any action to enforce or
apply Section 203 that would impede, thwart, frustrate or interfere with the
Proposed Acquisition;

            (c) enjoin PeopleSoft from entering into any further CAP contracts;
extending, continuing or expanding the CAP; or making changes that are outside
the ordinary course of business to the update or support practices that
constitute the standards an acquirer would need to maintain in order to avoid
triggering potential liability under the Versions 2, 3 and 4 CAPs;

            (d) enjoin PeopleSoft and the Director Defendants from adopting any
measure that has the effect of impeding, thwarting, frustrating or interfering
with the Tender Offer, or the Proposed Merger, including, but not limited to,
amendments to the Rights Plan, amendments to PeopleSoft's bylaws, the initiation
or extension of unreasonable rebate programs (such as the CAP Offers), stock
repurchase programs, pursuit of alternative transactions with substantial
break-up fees and/or lock-ups, "White Knight" stock issuances, changes to
licensing agreements, or executive compensation arrangements with substantial
payments triggered by a change in control;

            (e) declare that the CAP contracts entered into by PeopleSoft are
unlawful and a breach of the fiduciary duties owed by the Director Defendants;

            (f) declare that the CAP contracts are coercive and impermissibly
interfere with future directors' exercise of their fiduciary duties,

            (g) award Plaintiffs their costs and disbursements in this action,
including reasonable attorneys' and experts' fees, and

                                      -39-




            (h) grant Plaintiffs such other and further relief as this Court may
deem just and proper.

                                     /s/ Brock E. Czeschin
                                     --------------------------------
                                     Allen M. Terrell (#709)
                                     Daniel A. Dreisbach (#2583)
                                     Brock E. Czeschin (#3938)
                                     Dawn N. Zubrick (#4327)
                                     Richards, Layton & Finger
                                     One Rodney Square
                                     P.O. Box 551
                                     Wilmington, Delaware 19899
                                     (302) 651-7700
Of Counsel:                          Attorneys for Plaintiffs
Michael P. Carroll
Joel M. Cohen
Davis Polk & Wardwell
450 Lexington Avenue
New York, New York 10017

Dated: June 18, 2004

                                      -40-

EX-99.(E)(23)

                                                                 Exhibit (e)(23)

The following weblog installments have been posted to PeopleSoft's website at
www.peoplesoft.com. These are the most recent in a series of weblog commentaries
written by Gary Reback regarding the ongoing Oracle antitrust trial. These and
the other weblog installments may be accessed without charge by clicking on
"View from the Court" at PeopleSoft's website at www.peoplesoft.com.

- --------------------------------------------------------------------------------


TRIAL DAY 11:  JUNE 22, 2004

Today, Oracle began its case with the expert testimony of Tom Campbell, Dean of
the Haas School of Business at U.C. Berkeley. In a trial, you usually start with
your best and strongest, and everybody assumed Oracle would do the same. After
all, Tom must have the world's best resume: Ph.D. in economics, law degree from
Harvard, impressive government appointments and elective positions, Supreme
Court clerkship, countless accolades and honors.

But once the testimony started, it seemed to me that everything seemed to turn
around. First, Dr. Campbell asserted that he was assuming the Government's
market definition placed each procurement in a separate antitrust market. He was
primed to explain how such an approach was inappropriate. The problem was, the
Government specifically disclaimed such an approach during Professor Elzinga's
testimony. Rather, Dr. Elzinga explained, and the Government's complaint clearly
said, that all procurements of high function core HR software were in a single
market, and all procurements of high function core financial software were in a
second market. Two markets, not many markets.


TRIAL DAY 11


The Judge agreed to hear the testimony of Dr. Campbell over the Government's
objection of relevance (i.e., Campbell was arguing against something the
Government was not even asserting), but the Judge came back to ask Campbell
later in the testimony whether any of what he had said made any difference,
given the Government's market definition. (Campbell maintained that there was
still some relevance.)

Then it turned out that Professor Campbell had not even done a study. Rather, he
took a set of assumptions given him by Oracle's attorneys, and from those
assumptions, he concluded that the Government's conclusions were incorrect.

But some of Campbell's assumptions seemed dubious, and contradicted by the
evidence in the case. For example, Campbell assumes that HR and financial
management software are "durable goods" -- like the washers or dryers you buy at
Sears -- that last 10 years. Software code doesn't wear out, Campbell reasoned.
But on cross, the witness was confronted by all the testimony of customer
witnesses saying they needed frequent upgrades to keep their software current.
Campbell then maintained that customers might not need upgrades, particularly if
they are in a "mature industry," in which case, to quote the witness, "the
software may do just fine without upgrades."

TRIAL DAY 11


Few in the courtroom could believe their ears. The Judge had to make sure he
heard it right:

     The Court:     "Well, Dean Campbell, we are talking here about financial
                    management and human relation software, and even if the
                    company's business does not change, the tax laws change. I
                    don't have to tell you as a former member of Congress that
                    Congress -- (laughter) -- is always tinkering with the tax
                    laws, they're tinkering with the employment laws, and all of
                    these changes have to be built in and incorporated in these
                    systems.

                    So an enterprise really has no choice but to maintain and
                    upgrade the system going forward."

The witness agreed, but then asserted that the upgrade could be done by someone
other than the vendor. "Who?" you might ask. A best-of-breed provider, said the
witness. Now the Judge was really puzzled. "Where is that in the record before
the court?" the Judge asked. The witness claimed to have read that information
in Forrester and Gartner reports. The Judge next asked:

TRIAL DAY 11


                    "Well, doesn't this undermine your assumption that the
                    product is durable?"

"No" said the witness. Actually, he said it was not a question that could be
answered "yes" or "no."

Professor Campbell also assumes that customers had the option to "do nothing" --
that is not to buy new software if it was too expensive. In fact, Campbell
claimed that "doing nothing" deserves its own market share. So that if you were
representing the market shares of U.S. automakers, the results might read: "GM,
40 percent; Ford, 30 percent; Chrysler, 20 percent; doing nothing, 10 percent."

Once again, the witness was confronted with customer testimony about how
customer after customer was required to purchase a new system because it had
outgrown its existing system or because the vendor of the existing system had
gone out of business. As the witness from Verizon observed:

                    "[T] the point is that you can't delay it [the purchase]
                    much. You can't let these things simply quiesce and run your
                    company the way they run it today. That's not an option."

Confronted with this testimony, Professor Campbell readily agreed that the fact
a company can make one delay "doesn't mean it can delay for much."


TRIAL DAY 11


Professor Campbell actually made four additional assumptions, but constraints of
time preclude me from describing the testimony on all these assumptions in
similar detail. Suffice it to say that from these assumptions, Campbell
concluded that "economic theory" was unable to predict whether prices would be
higher or lower if there were three vendors, and two of them merged. The witness
appeared to get the Judge's attention with that testimony:


     The Court:     "[A]re there other situations in economics where less is
                    better than more, which is what you seem to be saying?"

The witness gave a long-winded answer to the effect that a bilateral negotiation
might produce a better combination of functionalities than a situation in which
multiple vendors compete. "Well, why is that?" the Judge wanted to know. Because
negotiation is different than bidding replied the witness. More questions from
the Judge, who wanted to know, if you are building a house, why you wouldn't get
a better price negotiating with several contractors instead of just one. The
witness answered this question, but even after rereading his answer in the
transcript, I'm not sure what he meant.

TRIAL DAY 11


But the Government attorney couldn't resist piling on. He confirmed the
witness's testimony that "economic theory" cannot tell us whether having three
vendor choices would produce better prices than just having two. Would that also
be true, if there were only two vendor choices, going to one vendor choice
post-acquisition? asked the lawyer. "Economic theory does not predict," replied
the witness. All right, said the lawyer, suppose PeopleSoft, Oracle, and SAP
were in a cartel. Would that result in higher prices? "No," said the witness.
"It's indeterminate."

Suppose, that Oracle, SAP and PeopleSoft were all going to merge into a single
company, the Government lawyer asked. Would that result in higher prices?
"Economic theory does not predict the outcome of bilateral monopoly," said the
witness.

Okay, the DOJ lawyer continued, you say that "economic theory" can't predict.
But, he asked, do you have an opinion as to whether a merger of all three
vendors in this market would result in higher prices? Professor Campbell said,
"I'm saying I was not asked to, nor did I form such an opinion."

The Judge noted the contrast between Campbell's testimony and that of the
customer witnesses:

TRIAL DAY 11

                    "The Court has heard testimony about a number of customers
                    of high-function software that are very knowledgeable, very
                    sophisticated individuals, and they have expressed concern
                    about this merger.

                    Now, given their obvious knowledge about the business, their
                    business, their knowledge about the product involved, what
                    should the court make of this testimony?"

The witness replied that customers always complain about mergers, at least that
was his experience when he held a high position at the Federal Trade Commission.
So, in the end, the witness continued to refuse to agree with customers that a
"three to two" merger will likely produce price increases.

Here is what the leading FTC economists now say on the FTC website about "three
to two" mergers:

                    "It is probably in the mainstream of economics and antitrust
                    that a merger in an industry with only three competitors in
                    a well-defined market protected by barriers to entry is also
                    likely to be problematic, absent convincing efficiencies or
                    unusual facts. [A] merger of the two of the three
                    significant competitors in



TRIAL DAY 11


                    a well-defined market appears to have an almost
                    non-rebutable presumption of being anticompetitive."

As bad as it was for Oracle in the courtroom it was far worse out in the hall
with the press. "Wasn't the Judge skeptical?" a reporter asked Dan Wall. "Why
can't the witness predict higher prices if all suppliers merge?" another
journalist wanted to know. Dan Wall's reply was to cite an editorial in the Wall
Street Journal, arguing that the Government should not have brought this case.
Then, a few words into a second or third answer, Wall just stopped,
mid-sentence, and walked away from the press.

Tomorrow the Government's Microsoft witness testifies. Stay tuned.


[END]


THE GOVERNMENT RESTS

After submitting hundreds of pages of depositions, testimony and documents, the
Government rested its case in chief following the testimony of its Microsoft
witnesses on Wednesday and Thursday.  So it's time to ask how the Government is
doing in terms of proving its case.  As I explained at the beginning of this
journal, it is fair to use the Government's Pretrial Brief in which the DOJ laid
out what it would try to prove, as a scorecard in this case.  Here is my
perspective.

First, the Government said it would try to prove that there are a group of
customers who have only Oracle, SAP and PeopleSoft as potential vendors for core
HR and core Financial software.  This point has been proven in spades.  Customer
after customer has taken the stand to say just that.  And, underscoring the
point, Government expert witnesses have analyzed and displayed in court internal
Oracle discount records that reveal in many cases the key competition for the
account is between PeopleSoft and Oracle. The Government has demonstrated
through the testimony of customers and experts, and through Oracle's own
documents, that this competition is especially vigorous and has saved customers
millions of dollars.

The best Oracle has been able to do is "pick around the edges," as a DOJ lawyer
explained.  Oracle has shown that large companies sometimes use outsourcing for
a particular function, like payroll for example.  And large companies sometimes
use software from Lawson or other vendors to run a discrete division or
function, but not for the enterprise platform.  And customers sometimes use
software of best-of-breed providers for particular functions.  And, indeed, if
the customer's IT department cannot get adequate funding, sometimes the
deployments of enterprise HR or financial management platforms are delayed.

The press has sometimes sensationalized testimony about these subjects,
characterizing such testimony as "admissions."  This is not the case, in my
opinion.  In fact, it is important to understand what Oracle has not been able
to prove.  No Government customer witness has agreed under cross-examination
that it could just as well use software from some company other than the Big 3
for its HR or financial platform.  They have all rejected this key Oracle
assertion and they have all maintained that they would be harmed by the proposed
transaction.

And, in a display of documentary and testimonial support, the Government has
been able to show that Oracle's internal documents (the very documents Oracle
lawyers tried to avoid producing), as well as the testimony of Oracle
executives, support the Government's case.  Who could forget Professor Elzinga's
bar chart summarizing Oracle's own internal discount request forms, showing far
more requests to discount against PeopleSoft than any other vendor.  SAP was
next.

But, then, in the words of Dr. Elzinga, it looked like the numbers
"fell off a ledge."  Here is Elzinga's chart:

[SEE CHART IN EXHIBIT (e)(23)(a) ATTACHED HERETO]

If those other sources of software -- outsourcers, Microsoft, Lawson, and the
like -- actually constrained Oracle's ability to raise prices, Elzinga asked,
why weren't they listed as justifications for discounts more frequently?  In
short, the Government made its point by testimony, by expert case study, and by
a review of Oracle's own documents.  One such document said it all:

     "Remember, SAP is not the enemy....  PeopleSoft is the enemy."

The Government's pretrial plan next argued that there are other customers,
beyond those in the upmarket, that have more than 3 choices -- Oracle, SAP,
and PeopleSoft to be sure, but sometimes another vendor, such as AMS or Lawson.
In my opinion, the Government has proven this contention.

Oracle lawyers denied even the existence of a mid-market.  But the testimony of
Oracle executives like Chairman Jeff Henley, and the very organization of the
Oracle salesforce into smaller customers and larger customer groupings, admitted
the point away.

The Government showed that the competition from AMS is limited to financial
software in some Government installations -- indeed, limited to Government
installations where AMS is already running. And the Government showed through
customer testimony and through the expert testimony of Harvard Professor Marco
Iansiti (who actually analyzed all vendors' software) that Lawson's products
cannot do the job in the upmarket, both for functional reasons and for lack of
sufficient worldwide support.

Third, the Government has stated in its Pretrial Brief that the Big 3 vendors,
through the knowledge they obtain in the sales cycle, know the needs of
potential customers and therefore know which of the other vendors constitute
actual competition for the account.  The Big 3 vendors can therefore price
discriminate -- charge more to some customers than others, depending on the
customer's needs and the level of competition.

This has been proven through the testimony of PeopleSoft, the deposition
testimony of SAP executives and through the admissions of Oracle's own
executives.  This is a key point.  Government experts have demonstrated that
Oracle's own discount documents show key evidence of price discrimination.  So
even if Lawson competes for an occasional upmarket account, that competition
will not constrain Oracle post acquisition in accounts for which Lawson is not a
good substitute.

The proof goes even further, demonstrating that Oracle and PeopleSoft are
head-to-head competitors for many accounts.  The ability of the post-acquisition
firm (Oracle) to price discriminate means that even some competition from SAP
for those accounts will not prevent Oracle post acquisition from gouging those
customers.

This proves the barriers to entry are substantial.  If it would take even
Microsoft years to become a viable competitor, Oracle would be able to price
gouge post-acquisition.

So, all in all, the Government has presented a compelling case.

A word about the Government's presentation.  Some analysts and reporters have
criticized the Government's cross-examinations, in particular, as not always
being as polished as those of Oracle's lawyers.  But, in my opinion, the
Government has conceptualized and put on a complicated case in a way in which
all important parts come together at the end.

The Government began with Oracle customers, then PeopleSoft customers, assessing
their procurements and the need for competition from PeopleSoft to guarantee the
benefits of free market competition.  The Government put on PeopleSoft
executives to explain what the company makes and how it competes.  The
Government showed that Oracle executives under oath in deposition see the market
in much the same way as PeopleSoft does.  Ditto for SAP, whose testimony came in
through deposition.

And, at the end, the Government experts, one traditional economist and the other
a famous name in auction-based modeling, pulled it all together.  From the
perspective of traditional economics (represented by Professor Elzinga) the
markets are highly concentrated, the barriers to entry are high, and the merger
will produce anti-competitive results.  Oracle's own discount forms demonstrate
that.  And from the perspective of economic effects, the result is equally
clear.  Dr. McAfee's analysis showed that PeopleSoft's competition has resulted
in lower prices to customers and that without PeopleSoft's competition, prices
would go up significant amounts.  Three different lines of Dr. McAfee's analyses
support the same conclusion.

I expect that Oracle will continue to pick at the studies by the Government
experts.  Perhaps they included an erroneous Oracle form or two.  Perhaps they
should have drawn the upmarket line as $600,000 as opposed to $500,000.
Perhaps, perhaps, perhaps.  The body of the Government's case has come into the
record unscathed. No Government customer witness under cross-examination ended
up agreeing with Oracle's lawyers.  They all held their ground and rejected
Oracle's contentions.  There were no gaffes in the testimony of the three
Government experts.  Perhaps Oracle can continue to "nibble around the edges"
but the experts themselves stood by their powerful conclusions.

[END]


TRIAL DAY 12:  JUNE 23, 2004

The judge released an order today, denying the media's request to unseal all
highly confidential documents previously submitted into the record under seal
and to unseal, as well, the report of Judge Legge, appointed as a special master
in the case. The order pointed out that Judge Legge's report includes many
confidential documents that did not end up becoming part of the record. The
order agreed that many companies designated as "highly confidential" documents
that really did not deserve that characterization, but the order said that the
court had remedied that practice by permitting only appropriate documents to be
kept confidential, and if some documents have been improperly designated, the
court will deal with them on a case-by-case basis.

The first witness of the day was Richard Knowles, the vice president of
Operations -- North America, for SAP America. Knowles was put on by Oracle. One
of the SAP lawyers told me at the beginning of the trial that if Oracle called
SAP as a witness, SAP's testimony would help the Government just as much as
Oracle. That is precisely what happened.

TRIAL DAY 12


On direct examination, Oracle elicited testimony that SAP was a large global
competitor, sufficient at least in the witness's view to compete for any
customer in any vertical. The witness explained that SAP separated customers
into categories based on revenue size, rather than functional needs. The witness
claimed that an Oracle acquisition of PeopleSoft would result in more rather
than less competition, and the witness claimed that SAP was "neutral" with
respect to the Oracle proposal to acquire PeopleSoft, not taking sides on the
issue. Finally, the witness identified internal documents that evidenced concern
about competition from Microsoft.

In one puzzling bit of direct testimony, Oracle elicited testimony that SAP's
internal discount request forms showed relatively little final round bid
competition with either Oracle or PeopleSoft. Since Professor Elzinga has
already shown that Oracle's forms show a great deal of head-to-head competition
between Oracle and PeopleSoft, the testimony elicited by Oracle from Knowles
appeared to confirm a key component of the Government's case -- that Oracle and
PeopleSoft compete directly against each other far more often than either
company competes with SAP.

On cross-examination, the SAP witness said a number of things that support the
Government's position. First, with respect to



TRIAL DAY 12


Microsoft, the witness admitted that with all of Microsoft's efforts, that
company could compete for only those customers up to $300 million in revenue --
the lower part of the mid-market. Next, the witness admitted that SAP's
marketshares for the U.S. were different than its shares worldwide, bolstering
the Government's view of a U.S.-only market. And the witness admitted that SAP
had failed to capture significant U.S. business in a number of verticals, until
including utilities and banking.

The witness testified that it might take three years for SAP's alliance with
Accenture to generate a fully functional product, bolstering the Government's
position that the product markets at issue in this case are hard to enter. The
witness acknowledged under cross-examination that the view that an Oracle
acquisition of PeopleSoft would produce more competition was simply the
witness's personal view, and conceded that internal SAP documents recognize that
a combined PeopleSoft/J.D. Edwards will be SAP's key competition for application
software. The witness also acknowledged that SAP was not, in fact, neutral --
SAP had testified in the E.U. on Oracle's behalf, and the CEO of SAP had
publicly criticized the DOJ for bringing this case.

All in all, it is difficult for me to see how the SAP appearance helped Oracle's
case all that much.

TRIAL DAY 12


In the afternoon, Doug Burgum, senior vice president of Microsoft Business
Solutions testified, out of order, as the final live Government witness. Burgum
had been the CEO of Great Plains. In fact, he testified how he had mortgaged his
family farm in North Dakota to seed the company and how he had taken the company
public. When the company was sold to Microsoft, he became a key executive of
that company, running all four Microsoft back-office product lines.

Burgum produced internal Microsoft documentation showing precisely the market
segments by size and description that Microsoft products are aimed at -- all
within the small enterprise and mid-market space. Burgum explained that not all
customers fit neatly into a category; nevertheless, it is clear what markets
Microsoft's products are intended to address. Microsoft mid-market products may
also be targeted at subsidiaries and divisions of large corporations, provided
the subsidiaries or divisions need a low level of functionality.

The witness went on to explain (and showed internal Microsoft documents to
support his claim) that Microsoft has no intention to ever sell to large complex
customers in the up-market. That market, according to the witness, requires a
direct sales force



TRIAL DAY 12


and large amounts of support, and includes relatively few potential customers,
as compared with the mass distribution markets that Microsoft focuses on.

Burgum described, and showed documents about, Microsoft's acquisition talks with
SAP. Microsoft started those talks in response to Oracle's hostile tender offer,
according to the witness. Of significant interest was the fact that Microsoft's
internal analysis, made at the beginning of the SAP talks, concluded that
Oracle's unsolicited tender offer, even if unsuccessful, would weaken PeopleSoft
and distract its executives. The witness testified that the talks with SAP were
abandoned when Microsoft focused on how difficult integration of such a large
acquisition would be.

Most significant was Burgum's testimony about the heretofore secret details of
Microsoft's "Project Green." According to the witness, Microsoft is writing a
common code base to integrate all four of its product lines (each product line
came to Microsoft through acquisition of a company). Release of the first
version of this new code base is dependent upon the prior release of a new
Microsoft operating system, and a new Microsoft server product, both code named
"Longhorn." According to Burgum, the new application product, Project Green,
will not likely be



TRIAL DAY 12


initially released until 2008 or 2009, at best, and the new product will not get
even the functionality of the existing Microsoft mid-market products until about
2011.

Oracle's cross-examination did not seem well focused. Oracle first attempted to
show Microsoft's bias, by arguing that Oracle's tender offer would endanger
Microsoft's database business. Then, Oracle attempted to prove that Microsoft
was already selling into the up-market--something Oracle would have to prove in
order to demonstrate the potential of timely entry by Microsoft, given the fact
that Microsoft's new product won't be ready until 2011. But Dan Wall, Oracle's
attorney, could show, at best, only one or two large Microsoft application
customers, and some internal prospect lists, claiming a handful of
opportunities. In any case, the witness explained that he had set up an internal
clearance procedure to make sure that Microsoft salesmen did not sell its
product into large accounts, beyond the capabilities of those products. And the
witness stuck to what the internal document showed -- that Microsoft does not
target anything larger than the upper mid-market.

Wall also attempted to get the witness to agree that the up-market is saturated,
with few new procurement opportunities. But out in the hall, after the court
session, the press wondered how



TRIAL DAY 12


Oracle could argue on the one hand that Microsoft wanted to enter the up-market,
but at the same time claim that the up-market was saturated. Why would Microsoft
invest hundreds of millions of dollars to enter a saturated market?

Oracle originally planned to call its own Microsoft witness, but Wall suggested
that Oracle would forego that opportunity. Given the state of the record, it is
difficult to see how Oracle can prevail on the issue of Microsoft's potential
entry. It is one thing to suggest to the press that Microsoft is really lying
and actually intends to go into the up-market, however implausible that seems.
But proof in court is something altogether different. Oracle must prove that
entry by Microsoft is "likely," "timely," and of sufficient competitive scale
before the law will recognize Oracle's claims. Right now, I believe Oracle is
nowhere near to satisfying these requirements.


[End]



TRIAL DAY 13:  JUNE 24, 2004


Oracle lawyer Dan Wall began the day by announcing that Oracle will eliminate a
number of the witnesses it previously intended to call, and that the trial might
conclude the following Wednesday or Thursday.

The first Oracle witness of the day was Michael Sternklar, of FESCo, a division
of Fidelity Investments. Oracle has argued that outsourcing is a major source of
competition for the three (Oracle, PeopleSoft, and SAP) application software
vendors, and FESCo, a large outsourcer, was expected to support Oracle's claim.

The witness did assert that business processing outsourcers ("BPO") have
competitive advantages over traditional software vendors in terms of service
offerings, but between the direct and cross, the witness admitted:

     1.   FESCo only has 10 large customers, including its own parent company,
          and has plans to add only one new customer next year and five the
          following year;

     2.   FESCo cannot provide services to employees located outside the U.S.
          even if its potential customers might be headquartered inside the
          U.S.; and

TRIAL DAY 13


     3.   Most BPO companies run software provided by Oracle, SAP, or
          PeopleSoft. In fact, most run PeopleSoft software; FESCo itself runs
          Oracle software.

The witness testified that FESCo made an acquisition of some new software from
IBM and that FESCo is in the process of rewriting and upgrading its software.
The witness said that FESCo will offer this new product within 18-24 months
and will migrate existing customers (all of which are running on Oracle) to that
new platform within five years. However, he conceded that these goals are
"aspirational." Under questioning from the judge, the witness admitted that
there have already been significant delays in the "major rewrite" of its new
product, currently in process.

Oracle's second witness of the day was Ken Harris, a consultant who markets his
services under the name "Retail.In.Genius." Harris testified he had prior CIO
experience at Taco Bell, Nike, and Gap. He expounded at great length about all
the techniques he used to get vendors to lower their prices to him, including
the alternative of "doing nothing." He emphasized with great



TRIAL DAY 13


gusto all the alternatives he claimed enterprise customers have for back-office
software needs.

The direct testimony was punctuated by a chess match between Dan Wall and the
Government attorney, Steven Kramer. Whenever Wall would ask a conclusary
question about what enterprise customers do, Kramer would object on the ground
that the witness could not testify beyond his experience at the three customers
where he had been employed. Only witnesses qualified by the court as "experts"
can testify more broadly, and this witness, Mr. Harris, was not qualified as an
expert, argued Kramer. Wall kept suggesting that he might offer Mr. Harris as an
expert, but he never did. So, the record of this witness's testimony is what is
called "lay opinion," the weakest and least credible form of testimony.

Far worse for Oracle was the judge's response to Kramer's various objections.
When Dan Wall argued that Mr. Harris should be able to give the same kind of
testimony that the Government's customer witnesses had, the judge himself
pointed out that, unlike those customers, Mr. Harris was a "paid witness."
Touche. Equally telling to me was the fact that the judge had no questions of
his own for this witness.




TRIAL DAY 13


I would have been tempted to pass on the opportunity to cross-examination
someone who even the judge understood was a "paid witness." But Kramer had even
more for the record. Mr. Harris admitted under cross that only one of the three
procurements he talked about was a major procurement, and that occurred in 1989.
More importantly, he conceded that during his entire job tenure for each
procurement, he selected either PeopleSoft, Oracle, or SAP as a final bidder and
winner. He also admitted that he never evaluated Lawson in any detail.

I have no idea what Oracle's lawyers were thinking when they decided to call the
witness. I think it would have been better for them to have no customer
testimony than to have, in opposition to all of the Government's customer
witnesses, this paid witness.

The third Oracle witness of the day was the director of Personnel Service
Delivery at Bank of America. He testified that prior to its merger with Fleet,
the witness used PeopleSoft core HR products, along with many point solutions.
Thereafter, as a result of Fleet's prior use of outsourcing, Bank of America has
used FESCo, running Oracle software. Indeed, on cross-examination, the witness
admitted that despite all this testimony about outsourcing, what basically
happened with adoption of the Fleet system is that Oracle replaced PeopleSoft.

TRIAL DAY 13


Finally, DOJ officially concluded its case by playing a videotape of the
deposition testimony of Orlando Ayala, a senior VP at Microsoft. Dan Wall had
previously observed that the witness, like Madonna, goes by only a single name,
in this case "Orlando." This piqued the judge's interest -- he subsequently
commented about the interesting appellation.

Orlando's testimony hammered home the points the Government has made about
Microsoft's effect on the high-function software market for HR and financial
services. The witness testified:

     1.   Non-established vendors have no sales opportunities with the largest
          customers, as it takes many years and substantial investment to enter
          the up-market, even if the vendor has a technically capable product;

     2.   On the other hand, Oracle, SAP, and PeopleSoft, which are established
          in the up-market, have substantial opportunities;

     3.   Microsoft's Axapta product is not capable of meeting the needs of
          multinational customers;

     4.   Microsoft has scaled back its marketing messages in order to avoid
          overselling its product and producing failed implementations;

TRIAL DAY 13


     5.   While Microsoft may occasionally get a large customer, Microsoft's
          offerings are not suitable for all such customers;

     6.   Microsoft has over-emphasized its business software, and its real life
          experience in trying to break into even the mid-market as an
          application vendor has been "a humbling experience."

Following the Orlando videotape, the Government officially rested its case. Dan
Wall moved for a "directed verdict" -- a routine motion at this stage of the
case. The motion argued that the Government hasn't proven the basic elements of
its case, so there is no point in making Oracle put on a defense.

Claude Scott responded by going down chapter and verse many of the proof points
of the Government's case. The judge asked a couple of questions and took the
motion under submission, to be ruled upon at a later time, perhaps not until a
final decision on the merits.


[End]



TRIAL DAY 14:  JUNE 25, 2004


The judge began the day by telling both sides that he is going to order them to
provide definitions to certain terms. They are to work together to provide
agreed-upon definitions, if possible, and otherwise, they can each submit a set
of definitions. The terms include "high-function software," "function," "ERP,"
and "EAS."

I believe the judge asked for these definitions because he has begun to think
about writing his opinion in this matter. Typically, a judge's opinion in a
complicated technology case will start with definitions of key terms used in the
decision, so the reader does not have to be versed in industry jargon to get the
gist of what the judge is saying.

The judge gave no indication as to how he will rule on this matter. But it is
hard to view the fact that he is asking for a definition of "high-function
software" -- something Oracle claims does not exist -- as a happy development
for Oracle. Nevertheless, out in the hall at the break, Oracle's lawyers told
the press, and some journalists dutifully reported, that the judge's request
"appeared to be questioning the basis of the Government's case."

TRIAL DAY 14


Oracle called its own Executive Vice President of Application Development, Ron
Wohl, as the first witness of the day. Wohl made three basic points in his
direct testimony:

     1.   Scalability: Wohl said scalability of application code is "not a
          primary constraint" on the ability of enterprise software to serve
          large numbers of users and hence, a company like Microsoft can scale
          its mid-market product to serve much larger concerns;

     2.   Integration: Wohl said that EAS suites are best suited to smaller
          companies, while larger companies operate a wide variety of software;

     3.   Innovation: Wohl claimed that competition is but one driver for
          innovation and that Oracle adds new features not so much to compete
          with PeopleSoft, but to satisfy customer demand.

Before the Government lawyer could even begin his examination, the judge began
to ask Wohl questions. Under questioning from the judge, Wohl conceded that an
application can scale easily to larger numbers of employees only if other needs
remain constant. The Government lawyer followed this up, getting Wohl to concede
that complex scalability issues cannot be solved through




TRIAL DAY 14

architecture alone (as he had testified on direct) but also must be addressed in
the software code.

The Government lawyer, Phil Malone, next showed Wohl internal Oracle documents
and got him to admit that PeopleSoft's innovations (and superior HR products)
forced Oracle in turn to innovate. This is an important concession. The
Government has already shown that head-to-head competition with PeopleSoft
produces lower Oracle prices to customers. Now the Head of Application
Development of Oracle conceded (and his own documents confirmed) that
head-to-head competition with PeopleSoft produces greater innovation by Oracle.
Showing that competition between the merging companies has produced lower prices
and greater innovation is the key to the Government's case.

Anticipating arguments that I expect Oracle will make in the coming days, Malone
also forced Wohl to concede that Oracle's products, such as application servers
and integration products, are not substitutes for enterprise applications.
Finally, under cross, Wohl admitted that he was not consulted regarding plans
for R & D head count post-acquisition and that head count numbers he provided in
that regard are "speculation."

TRIAL DAY 14


Oracle's final witness of the day was Safra Catz, its Co-President and a board
member. Oracle's other Co-President, Chuck Phillips, was publicly leading the
effort to acquire PeopleSoft until about the time his rather unfortunate
"oligopoly" language came to light, at which point Catz seemed to become the
primary spokesperson on the issue.

Catz tended to give long-winded answers to nearly every question, so the
significance of what she said, particularly on cross-examination, seemed to be
lost in media reports. Catz's first major point was that Oracle sales for what
the Government calls "high-function software" (more precisely, Oracle's sales of
HR and financial platforms to large corporations) are "chump change." Such sales
totaled about $18.6 million for the first three quarters of 2003. Her point was
that any anti-competitive effects of the proposed acquisition on such sales
would be more than offset by the large efficiencies the deal will net for
Oracle.

But on cross, the Government lawyer, Claude Scott, showed data making it clear
that the Oracle-PeopleSoft rivalry is hardly over "chump change." Scott showed
the witness internal Oracle win-loss data revealing that for a single quarter
during the relevant period, Oracle won only three deals against PeopleSoft. But
PeopleSoft won 11 deals against Oracle, and the two companies were continuing to
compete over no fewer than 59 deals in the pipeline.

TRIAL DAY 14


Second, on direct examination, Catz contrasted her view of the trivially small
affected product market with the vast savings she claims Oracle will achieve in
the transaction. She presented a "model" -- a set of formulas and relationships
- -- to demonstrate what would happen to the two companies post-acquisition. Her
point was that the post-acquisition company will be able to function at a higher
scale. She said she got this idea after PeopleSoft proposed that it acquire
Oracle's application business in 2002. Catz said the idea for a more efficient,
combined company was first proposed to her at a meeting with PeopleSoft CFO
Kevin Parker.

Catz's testimony on the post acquisition company goes to the antitrust issue of
"efficiencies." If a proposed deal will cause untoward concentration and likely
decreases competition, the defendant can still prevail if it can show that the
deal will make the post-acquisition company so much more efficient -- in certain
specified ways -- and these efficiencies will be passed on to consumers in the
form of lower prices.

TRIAL DAY 14


There are several unusual aspects of Catz's model. Of course, it is normal in an
antitrust case for a defendant to argue that the proposed deal will make the
merged company more efficient. But this type of testimony usually comes into the
record as a learned study prepared by a distinguished expert. Here, Oracle is
presenting its efficiencies study through a lay witness. Indeed, Claude Scott
established on cross that none of Oracle's distinguished economic experts, nor
even Ron Wohl, the Oracle executive in charge of application R & D, were
consulted with regard to the numbers in the model. From the testimony, it
appears that Catz and Ellison just took a pre-existing model developed to
evaluate potential acquisitions of J.D. Edwards and Lawson, and made up numbers
for a proposed PeopleSoft acquisition. Scott established on cross that there are
no extant work papers that show how Catz and Ellison calculated the various
numbers put into the model.

Catz's model showed that Oracle can save over $1 billion a year post-acquisition
by firing about 6,000 PeopleSoft/JDE employees. Scott established that under the
model, the post-acquisition firm would keep virtually no PeopleSoft sales and
marketing personnel (because it won't be selling and marketing PeopleSoft
products) and only about half of PeopleSoft's development personnel (who would
work on the next generation Oracle product). The large numbers of fired
employees got most of the headlines.

TRIAL DAY 14


But to me, the most important part of the testimony was Catz's admission, under
cross, that there are no features that PeopleSoft products have that Oracle
could not develop on its own -- without acquiring PeopleSoft. Indeed, Catz
admitted again and again that Oracle and PeopleSoft HR and financial management
offerings are virtually identical. She also admitted that Oracle wants
PeopleSoft for its customers, and perhaps for some of its developers, but not
for any of its technology.

The key foundational claim of the Government's case -- found in the Government's
trial brief among many other places -- is that by this tender offer, Oracle is
trying to acquire PeopleSoft's customers without competing for them in the free
market. Catz admitted the point. And further, the Government argues that Oracle
will inflict upon customers enormous switching costs, and upon thousands of
people the burdens of job loss, not to get some greater functionality that will
be used to create a better product, but to be able to sell PeopleSoft customers
all the various products Oracle makes.

Catz admitted that post-acquisition, new Oracle application products, like the
current products, will only run on the Oracle database. She said PeopleSoft
customers who choose to migrate to



TRIAL DAY 14


the new Oracle applications will be given an Oracle database free at first (this
is a wholly new claim, not found on the Oracle website, as far as I know) and
that Oracle will do whatever possible to facilitate the migration to the new
platform. Of course, Catz admitted that Oracle will not pay the costs of
re-implementation and integration for both applications and a new database --
costs that are at least seven to eight times the cost of the software itself,
according to prior witnesses.

Catz also admitted that had Oracle acquired J.D. Edwards or Lawson, Oracle would
have increased maintenance prices to the customers of those companies.


[End]



TRIAL DAY 15:  JUNE 28, 2004

Oracle lawyer Dan Wall began today by announcing that Oracle will not be calling
PeopleSoft CEO Craig Conway as a witness after all. Oracle has had Conway on its
witness list since Oracle's initial witness designation in April. Oracle claimed
that it would question Craig for six hours, more than any other witness. When I
asked the Court to cut the amount of time Oracle proposed to interrogate Conway,
Oracle insisted that it needed the full six hours. Now it turns out that Oracle
has decided that it doesn't need Craig's testimony at all.

Perhaps Oracle never did intend to call Craig -- and only went through the
process of subpoenaing him for PR reasons or to attempt to distract him from
focusing on business. Or perhaps Oracle's lawyers actually intended to call
Craig, but finally came to their senses. We'll never know. But it seems clear to
me that Craig's testimony would have severely damaged Oracle's case. Craig would
have explained that PeopleSoft has gone through the past year, retaining the
trust of customers and employees, in the face of Oracle's unsolicited tender
offer. It would have been powerful testimony, but we won't see it, because
Oracle apparently decided it was afraid to examine Craig in open court.

TRIAL DAY 15



After the announcement abandoning Craig's interrogation, Oracle put on its first
witness of the day, John Coughlan, CEO of Lawson Software. Throughout the trial,
Oracle has touted Lawson as a rising upmarket competitor, sufficient in its own
right to prevent Oracle from raising prices following an acquisition of
PeopleSoft.

On direct, Coughlan took issue with statements made by government attorneys to
the effect that Lawson is "not a serious player" in the upmarket and is "not
likely to be" in the future. Coughlan spoke of his successes among large,
complex customers, especially in health care, retail, and the public sector,
listing several marquee customers in each vertical. Coughlan claimed that Lawson
has competed with Oracle and PeopleSoft on "almost every account."

But the cross-examination turned into a disaster for both Lawson and Oracle.
Here is the lead from INFORMATIONWEEK:

     CEO John Coughlan, testifying at the Oracle antitrust trial, tried to paint
     his company as more than a mid-market business software vendor, but was
     confronted



TRIAL DAY 15


     with examples of enterprise customers that the Justice Department used as
     examples of failure.

     Lawson Software, Inc. and CEO and president John Coughlan had a bad day
     Monday.

If anything, I think that INFORMATIONWEEK understates the extent of the carnage.
On direct, Coughlan hyped the fact that he had taken his company public. On
cross, he admitted that he had taken the company out at $14 a share, and it was
now trading at $7 a share. In 2003, it was revealed, Lawson had a net operating
loss. Government attorney Claude Scott then showed example after example of
Lawson problems that left customers dissatisfied, and having to turn to Oracle
and PeopleSoft. He backed up each example with internal Lawson documents and
customer memoranda.

MasterCard complained to Lawson about inadequate payroll technology, for
example, and concluded that it was not interested in Lawson's software because
of lack of functionality. Scott next showed that the City of Dallas, after
installing Lawson's product, complained of not being able to pay its police and
firemen. The witness acknowledged that the City of Dallas stopped paying Lawson.

TRIAL DAY 15



Scott showed that both Johnson & Johnson and McGraw-Hill, used as examples of
Lawson trophy customers in the direct examination, rejected further Lawson
purchases, because of Lawson's inadequate functionality. And so it was, account
after account. Problems at Hyatt Hotels, problems at Northrop Grumman, etc. And
more product deficiencies. Lawson can't support Asian languages, Lawson has
little revenue from international customers, etc. Coughlan had claimed in direct
testimony that his company had sales to many customers in excess of $500,000 in
licensing revenues, but on cross it was revealed that his list included
services, not just licensing revenue. Lawson, it turns out, has an R&D budget
about 7 percent of that of PeopleSoft.

Finally, Scott turned to full-page ads in the WALL STREET JOURNAL which Lawson
took out to publicly thank Oracle for putting Lawson in the same market as
PeopleSoft and SAP, based on Oracle's rhetoric in this trial. The witness
admitted that his company had spent more than $300,000 on these "Thank you
Oracle" ads.

In the afternoon, Oracle put on one of its principal economic experts, MIT
Professor Jerry Hausman. Hausman testified for several hours, attacking the
testimony of both of the government's principal experts, Ken Elzinga and Preston
McAfee.



TRIAL DAY 15


Hausman first criticized Elzinga's work, finding that the relevant market should
be worldwide (or at least Europe and the United States), rather than just the
U.S. He next claimed that the product market was poorly and vaguely defined, and
claimed that the market should include competition from outsourcers, and from
Microsoft, AMS, and Lawson. Actually, the government experts included AMS and
Lawson in their analysis, but Hausman claimed those companies were
insufficiently represented in the government's data set.

Hausman said his own studies showed that between AMS and Lawson, the two
companies covered half of the relevant market, and would keep Oracle
post-acquisition from raising prices. He went on to explain that he thought
Lawson would constrain Oracle post-acquisition, even though Lawson is present in
just a couple of verticals, because Lawson would move to additional verticals if
Oracle tried to raise prices. How Lawson would be able to do this was never
explained.

Hausman also claimed that because few financial or HR products are sold without
the inclusion of products from other pillars, resulting in blended pricing,
these sales "in bundles" would in some way constrain Oracle from raising prices
on financial and HR software. And Hausman testified that the largest customers
get



TRIAL DAY 15


the largest discounts, so there really isn't a class of customers to
price-discriminate against.

Most of Hausman's testimony was directed against Professor McAfee's work. McAfee
had testified extensively as to the anticompetitive effects of the proposed
acquisition -- likely substantial price increases and a diminution in
innovation, in McAfee's opinion. Hausman, on the other hand, argued that this
case was not proper for a "unilateral effects" approach because the elimination
of PeopleSoft would not affect price, post-acquisition. According to Hausman,
customers involve only 1-2 vendors in pricing negotiations. Furthermore, he ran
his own regression analyses, getting results different from McAfee; according to
Hausman, the presence of additional vendors does not result in lower prices.

As the day went on, Hausman's testimony seemed to get more and more extreme. He
claimed that his study showed that SAP constrains Oracle's prices just as much
as PeopleSoft -- notwithstanding the fact that Oracle's own discount forms show
much greater competition from PeopleSoft. He claimed that just two vendors, SAP
and Oracle, would provide sufficient competition to keep Oracle from raising
prices post-acquisition, despite the testimony of customers, conspicuously
including Oracle's



TRIAL DAY 15


customers, to the contrary. Hausman concluded that prices would not rise at all
after the acquisition, and innovation would not be affected by the absence of
PeopleSoft.

Hausman's direct testimony ended at the conclusion of the court day. He will be
cross-examined tomorrow.

[END]

TRIAL DAY 16:  JUNE 29, 2004


There was only a half-day of court time available today, and the
cross-examination of one of Oracle's principal economic experts, Jerry Hausman,
consumed all of it. Hausman was cross-examined by Government attorney Sandy
Adler. Sandy came into town from Washington just for the cross-examination; he
has no other role in the case. In fact, whenever Jerry Hausman testifies as an
expert against the Government, Sandy Adler comes into town to cross-examine him.
This is like having a "defense specialist" on the bench of your basketball team
that you bring into the game just to shut down the other team's shooting guard.

Before the cross-examination started, Hausman stopped by to talk to someone in
the row behind me. In a stage whisper everyone could hear, Hausman said, "[the]
last time Adler cross-examined me, I made minced meat out of him." Well, not
today, Jerry.

Adler's manner is scholarly rather than accusatory. When he burrows in on the
wording of the expert's written report, it sounds like he is exploring the
subtleties of a scholarly tract. He sounds well versed in economics and
statistics -- and he looks like he could be an expert witness himself.

Today, aided by the witness's own demeanor, Sandy started with a few real
zingers. Sandy's first job, I believe, was to attack the witness's credibility
- -- to make the witness sound like he is



TRIAL DAY 16


given to overstatement and exaggeration, and therefore should not be believed.
Sandy showed on the screen a sentence from Hausman's written report in this
case, in which Hausman claimed that Professor Elzinga's reliance on the
testimony of PeopleSoft executives is:

     "Akin to using the testimony of a drug dealer to decide that crack cocaine
     is good for you."

"Did I read that correctly?" Sandy asked. "Yes," Hausman replied.

Next, Sandy raised the issue of how Oracle's "efficiency story" was prepared. In
an earlier journal installment, I explained that in all cases I have been
involved in, efficiency justifications are made through learned economic
studies, put into the record through expert testimony. Here, Safra Catz herself
did the efficiency study. This is extremely unusual.
Sandy asked Hausman if Hausman did any original research on Oracle's efficiency
report. Hausman claimed that he had. So Sandy read into the record Hausman's
deposition testimony in which Hausman said, "I did no original research. I took
what Oracle had done and tried to look at it...."

Next, Sandy read the witness a statement from Oracle's white paper submitted to
the Department of Justice during the investigatory phase of this matter. During
that phase, Oracle used a different economic expert, Fredrick Warren-Boulton,
but



TRIAL DAY 16


the earlier expert report was still submitted in Oracle's name. The Oracle
economist at the earlier phase (Warren-Boulton) submitted a report saying that
the "markets for EAS software functions similarly to auction or other bidding
markets in important respects." This is precisely the point that Professor
McAfee, a Government expert witness, made earlier in the trial.

Everyone in the courtroom knew that Hausman would have to contradict the expert
report Oracle itself submitted to the Department of Justice. That was
embarrassing enough. But Hausman went further and said:

     "I want to say this in the nicest possible way, but Dr. Warren-Boulton, in
     my view, is an extremely poor economist. So there is no reason I would
     agree with him."

Wasn't Warren-Boulton formerly the head of the Economics Department of the
Antitrust Division, Adler asked. "That's true," replied Hausman. Asked whether
he even bothered to read the earlier economist's report, Hausman said that he
started to, "[b]ut I thought it was of extremely poor quality, so I stopped very
early on."

So you reject the view of Oracle's previous economic expert, Sandy wanted to
know. "Not only rejected that view, I rejected all his views. ...I wouldn't take
very seriously what he had to say."

TRIAL DAY 16


Adler next turned to the issue of whether Oracle post-acquisition would be able
to raise prices. On an earlier trial day, Oracle expert witness Tom Campbell
said he could not say whether, if Oracle, PeopleSoft, and SAP all merged into a
single firm, that firm could raise prices.

Hausman did Campbell one better. Hausman started by claiming that if all three
vendors merged into a single firm, that firm COULD NOT raise prices. Actually,
when Hausman equivocated on the stand, Adler played his deposition testimony on
the court monitors.

Then Hausman claimed that even if all vendors of packaged software joined
together, they COULD NOT raise prices. Again, the witness equivocated, so Adler
played the witness's deposition testimony in which Hausman claimed that even if
all vendors of packaged software joined together, they would not satisfy the
"hypothetical monopolist test" of the Merger Guidelines. On the witness stand,
Hausman retracted his earlier position (also given under oath), and claimed that
the companies could raise prices for financial software, but he was not sure
about HR software.

Adler pressed the point. Did the witness ever correct his prior erroneous
testimony that he was now retracting? After all, the witness made many other
corrections to his deposition transcript.




TRIAL DAY 16


The witness said he just corrected the typos in his transcript. "But you
couldn't tell the other side that you've changed your entire view of the
market?" asked Adler. The witness said, "No," and provided explanation.

The next subject in the testimony was a key to the Government's case. Hausman
said in his direct testimony that he had conducted his own regression studies
and those studies showed that when additional bidders were added to the mix,
prices did not go down -- that is, increasing the number of competitors does not
result in greater competition in the form of lower prices. This is Oracle's
fundamental position. But through elaborate and difficult interrogation of a
hostile witness, Adler got Hausman to admit that whatever the general truths of
Hausman's statements on adding bidders, Hausman's own studies showed that adding
PeopleSoft as a competitive bidder induces Oracle to give over a 10 percent
greater discount.

Then the witness went on to contradict the next Oracle expert, Dale Kutnick, who
hasn't even testified yet. Hausman said he doesn't expect "technological
breakthroughs in the industry." But Kutnick is expected to argue that there will
be changes in technology so dramatic that the Government should stop worrying
about whether Oracle can increase prices post-acquisition.




TRIAL DAY 16


There was an extensive discussion about whether large customers actually pay
lower prices for the same software. On direct, Hausman said they do. But on
cross, his testimony was equivocal. Even the judge repeatedly asked questions
about this. Hausman agreed that large corporate customers pay more per seat than
smaller customers, but he also argued that they are not buying the same product.
The Judge seemed baffled by that explanation.

Finally, Adler asked Hausman when certain data he presented in his direct
testimony had been prepared. Adler made it clear that the data had not been
provided to the Government, as required by the court's procedures. Hausman
admitted he had made the calculations days ago. Adler made it clear that Hausman
had done a lot of work since his deposition was taken and his expert report was
filed, but none of that had ever been given to the Government. Adler then
revealed that Hausman had previously been sanctioned by a federal court for not
providing information to the other side in an appropriate manner. The court in
that case struck Hausman's entire testimony and the court of appeals affirmed
that decision.

[End]

Important Additional Information
PeopleSoft has filed a Solicitation/Recommendation Statement on Schedule 14D-9
regarding Oracle's tender offer. PeopleSoft stockholders should read the
Schedule 14D-9 (including any amendments or supplements) because these documents
contain important information. The Schedule 14D-9 and other public filings made
by PeopleSoft with the SEC are available without charge from the SEC's website
at www.sec.gov and from PeopleSoft at www.peoplesoft.com.

EX-99.(E)(23)(A)

exv99wxeyx23yxay