3076 1 IN THE IOWA DISTRICT COURT FOR POLK COUNTY ----------------------------------------------- 2 JOE COMES; RILEY PAINT, ) 3 INC., an Iowa Corporation;) SKEFFINGTON'S FORMAL ) 4 WEAR OF IOWA, INC., an ) NO. CL82311 Iowa Corporation; and ) 5 PATRICIA ANNE LARSEN; ) ) TRANSCRIPT OF 6 Plaintiffs, ) PROCEEDINGS ) VOLUME XII 7 vs. ) ) 8 MICROSOFT CORPORATION, ) a Washington Corporation,) 9 ) Defendant. ) 10 ----------------------------------------------- 11 The above-entitled matter came on for 12 trial before the Honorable Scott D. Rosenberg 13 and a jury commencing at 8 a.m., December 6, 14 2006, in Room 302 of the Polk County 15 Courthouse, Des Moines, Iowa. 16 17 18 19 20 HUNEY-VAUGHN COURT REPORTERS, LTD. 21 Suite 307, 604 Locust Street 22 Des Moines, Iowa 50309 23 (515)288-4910 24 25 3077 1 A P P E A R A N C E S 2 Plaintiffs by: ROXANNE BARTON CONLIN 3 Attorney at Law Roxanne Conlin & Associates, PC 4 Suite 600 319 Seventh Street 5 Des Moines, Iowa 50309 (515)283-1111 6 RICHARD M. HAGSTROM 7 AARON McPARLAN MICHAEL R. CASHMAN 8 Attorneys at Law Zelle, Hofmann, Voelbel, 9 Mason & Gette, LLP 500 Washington Avenue South 10 Suite 4000 Minneapolis, Minnesota 55415 11 (612)339-2020 12 ROBERT J. GRALEWSKI, JR. Attorney at Law 13 Gergosian & Gralewski 550 West C Street 14 Suite 1600 San Diego, CA 92101 15 (619) 230-0104 16 17 18 19 20 21 22 23 24 25 3078 1 Defendant by: DAVID B. TULCHIN 2 STEVEN L. HOLLEY SHARON L. NELLES 3 JOSEPH E. NEUHAUS JEFFREY C. CHAPMAN 4 Attorneys at Law Sullivan & Cromwell, LLP 5 125 Broad Street New York, New York 10004-2498 6 (212)558-3749 7 ROBERT A. ROSENFELD Attorneys at Law 8 Heller Ehrman, LLP 333 Bush Street 9 San Francisco, California 94104 (415)772-6000 10 STEPHEN A. TUGGY 11 HEIDI B. BRADLEY Attorneys at Law 12 Heller Ehrman, LLP 333 South Hope Street 13 Suite 3900 Los Angeles, CA 90071-3043 14 (213) 689-0200 15 DANIEL S. SILVERMAN Attorney at Law 16 Heller Ehrman, LLP 4350 La Jolla Village Drive 17 Seventh Floor San Diego, CA 92122-1246 18 (858) 450-8400 19 20 21 22 23 24 25 3079 1 BRENT B. GREEN Attorney at Law 2 Duncan, Green, Brown & Langeness, PC 3 Suite 380 400 Locust Street 4 Des Moines, Iowa 50309 (515)288-6440 5 RICHARD J. WALLIS 6 STEVEN J. AESCHBACHER Attorneys at Law 7 Microsoft Corporation One Microsoft Way 8 Redmond, California 98052 (425)882-8080 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 3080 1 MR. CASHMAN: Good morning, Your 2 Honor. 3 THE COURT: Good morning. 4 You had some issue you wanted to take 5 up before we went into Williams? 6 MR. CASHMAN: Yes. 7 Michael Cashman for the Plaintiffs. 8 And this morning I would like to 9 address one exhibit which Ms. Conlin would like 10 to use in her opening statement. If I may 11 approach, please. 12 THE COURT: Yes. 13 MR. CASHMAN: I'm handing the Court a 14 copy of Plaintiffs' Exhibit 3034. 15 THE COURT: Was this the Phase 6? 16 MR. CASHMAN: No. And I'm also 17 handing the Court some agreements that the 18 parties entered into on November 29, 2006. 19 And, in particular, I'm directing the Court's 20 attention to paragraph A1. 21 The reason I've given these two 22 documents to the Court is Plaintiffs' Exhibit 23 3034 has embedded hearsay in it on the first 24 page where that objection was sustained by the 25 Special Master. 3081 1 That embedded hearsay is indicated in 2 the material that's bracketed in yellow on the 3 first page. 4 Ms. Conlin wishes to use this document 5 in opening statement, but is not going to be 6 displaying or discussing any of that embedded 7 hearsay. She is going to be referring to the 8 last page of this document. 9 And I have bracketed the information 10 which Ms. Conlin is going to use in her opening 11 as you can see, near the top of the last page. 12 THE COURT: In pencil there? 13 MR. CASHMAN: Pardon me? 14 THE COURT: It's in pencil? You 15 bracketed in pencil? 16 MR. CASHMAN: Yes, I bracketed in 17 pencil. 18 That information is not subject to any 19 objection that has been sustained by either the 20 Special Master or the Court. 21 Microsoft is objecting to the use of 22 this exhibit on the basis of a collateral 23 estoppel objection. 24 The Plaintiffs believe initially that 25 collateral estoppel objection is not well 3082 1 taken. And at the appropriate time during 2 trial, when and if this exhibit is proffered, 3 we submit, would be the time when to take up 4 that objection. 5 But, more importantly for the moment, 6 is the agreement that I've handed up to the 7 Court regarding the use of evidence in opening 8 statements. 9 And as the Court will see in paragraph 10 A1, the parties have agreed that any evidence 11 that was designated in the course of pretrial 12 activities essentially, as indicated here, may 13 be used in opening statement if the proponent 14 of the evidence believes in good faith they 15 will be admitted for the purpose that the 16 evidence is being used in opening. 17 And then the agreement describes the 18 good faith standard. And it's undisputed that 19 collateral estoppel objections do not preclude 20 the use of such evidence in opening statement. 21 So, the Plaintiffs believe that on the 22 basis of our agreement with Microsoft, we 23 should be able to use this document, 24 Plaintiffs' Exhibit 3034, without objection. 25 And I also note for the Court's 3083 1 information, paragraph A3, which says 2 objections to evidence used in opening 3 statements are not waived if they are not 4 asserted during the opening. 5 Therefore, again, if and when this 6 exhibit is proffered, during the course of the 7 trial is the right time to take up the merits 8 of any collateral estoppel objection. 9 Thank you. 10 THE COURT: Mr. Tuggy? 11 MR. TUGGY: Yes, Your Honor. 12 With respect to Plaintiffs' Exhibit 13 3034, the portion Plaintiffs seek to use -- I 14 don't have the bracket Mr. Cashman provided to 15 you, but I believe it's the first paragraph on 16 the last page. 17 Is it the one that begins Bengt 18 Akerlind and ends at the word them? 19 THE COURT: Starts Gates. 20 MR. TUGGY: I see. Underneath that 21 line. 22 What this handwritten note says, and 23 we don't -- I don't know who within IBM wrote 24 this note. It says Gates, and then an arrow, 25 why are we working with IBM when they are doing 3084 1 these deals? Don't work with them. 2 It's unclear, based on the document, 3 whether that's a statement attributed to Gates, 4 a statement attributed to Akerlind. It's just 5 unclear what this evidence is just from the 6 face of the document. 7 Now, it does appear it was marked as 8 an exhibit in the government case, and perhaps 9 there's testimony there that has not been 10 provided to us that they intend to use to 11 enlighten us on the meaning of this document, 12 but I say all of that to point out that there 13 are ambiguities about this statement and what 14 force it ought to have at trial. 15 Now, this statement -- as you know, 16 the issue of whether Microsoft put improper 17 pressure on IBM not to preload SmartSuite was 18 part of the findings of fact that have been 19 read to the Jury, and that this relates to 20 SmartSuite can be seen from the line above the 21 part that the Plaintiffs intend to use. In 22 that line, it references SmartSuite. 23 In paragraph 132, or Finding of Fact 24 Number 132, Judge Jackson wrote, in sum, from 25 1994 to 1997, Microsoft consistently pressured 3085 1 IBM to reduce its support for software products 2 that competed with Microsoft's offerings. 3 And this note is dated February 19, 4 1997. 5 This document offered as an exhibit in 6 the government case is specifically in support 7 of Finding of Fact Number 132 and the preceding 8 findings of fact relating to this issue and 9 SmartSuite. 10 So it's Microsoft's position that this 11 evidence is barred by the collateral estoppel 12 issue. Both because under the collateral 13 estoppel doctrine, the preclusion of evidence 14 works both ways and Plaintiffs are not 15 permitted to present evidence supporting 16 findings of fact unless they can identify some 17 alternative purpose which they haven't done 18 here. 19 And, also, that it will result in 20 prejudicial effect. I fully expect that if 21 Microsoft attempted to explain this document, 22 examine witnesses on what it meant, Plaintiffs 23 would object that Microsoft shouldn't be 24 permitted to do that based upon the preclusive 25 effect of collateral estoppel. 3086 1 So Microsoft asserts that this ought 2 not be shown to the Jury because this part of 3 the document is barred by the collateral 4 estoppel. 5 Now, as to the agreement, under the 6 agreement, it is true that the parties have 7 agreed that where a party has a good faith 8 belief that a document is going to be admitted 9 at trial, they can use it in their opening 10 statements. And there are certain definitions 11 of what constitutes a good faith belief. 12 In our view, this issue is now before 13 you. And if this Court should rule that this 14 exhibit ought to be barred by the collateral 15 estoppel doctrine for being presented at trial, 16 then Plaintiffs cannot have that good faith 17 belief and use it in their opening statement. 18 So the issue is being presented to you 19 before the opening statement, whether it ought 20 to be admitted. And if you rule that it 21 shouldn't be because of the collateral estoppel 22 doctrine, then, they can't have a good faith 23 belief in its admissibility to use it in 24 opening statements. 25 Based on those considerations, we 3087 1 respectfully request that Plaintiffs not be 2 permitted to show Plaintiffs' Exhibit 3034, 3 and, particularly, the section bracketed by 4 Mr. Cashman in their opening statements. 5 THE COURT: Anything further? 6 MR. CASHMAN: Yes, Your Honor. 7 Plaintiffs believe Mr. Tuggy is just 8 plain wrong on all counts. 9 First of all, the agreement is clear 10 that we are entitled to use this. This 11 document is not being moved into evidence at 12 this time. Therefore, their objections to it 13 are not right and should be taken up at the 14 time the document is proffered. 15 They are not entitled to object under 16 our agreement to the use of this document 17 during opening statement as a preliminary 18 matter. 19 Turning to the collateral estoppel 20 objection that Mr. Tuggy claims. First, he 21 started out by saying the document is unclear 22 and that it has ambiguities. 23 Well, if that were so, first of all, 24 that's a matter to clarify -- that Microsoft 25 could clarify through whatever witness it comes 3088 1 in at trial or on its own case, if they choose. 2 But that claim that it's ambiguous or 3 unclear undercuts their objection of collateral 4 estoppel because Plaintiffs don't agree with 5 this, obviously, as we've said before, but, 6 even under Mr. Tuggy's claim standard, that 7 they have to make a clear link between a 8 collaterally estopped finding and the evidence 9 at hand. If you're talking about evidence 10 which is ambiguous or unclear, they can't even 11 make their own claimed burden. 12 Furthermore, Mr. Tuggy attempts to say 13 that this is collaterally estopped under 14 Finding 132. 132 makes no mention of this 15 document or anything in Plaintiffs' Exhibit 16 3034. 17 Furthermore, Mr. Tuggy himself 18 acknowledges that Plaintiffs' Exhibit 3034 19 talks about SmartSuite applications. And we've 20 already been through the fact that applications 21 is a separate, independent claim in this 22 lawsuit and that Plaintiffs are certainly 23 entitled to use underlying evidence from the 24 government case to support claims such as its 25 applications claim. 3089 1 So the Plaintiffs, as explained 2 before -- I think all these collateral estoppel 3 objections are without merit, but for the 4 reasons I've addressed specifically relative to 5 3034, this exhibit should be permitted during 6 opening statement. 7 Thank you. 8 THE COURT: Anything further? 9 MR. TUGGY: Nothing further, Your 10 Honor. 11 THE COURT: I'll allow you to use it 12 during opening statement. 13 Do you want these back? 14 MR. CASHMAN: Yes. Thank you, Your 15 Honor. 16 THE COURT: Do you want to go with 17 Williams now? 18 MR. GRALEWSKI: If Your Honor is so 19 inclined. I know there's ten minutes. 20 THE COURT: The Court looked over the 21 four objections or, rather, appeals, I guess, 22 in regard to pages 218 of the deposition up 23 through page 222. 24 As far as 218, lines 5 through page 25 220, line 2, Court sustains the ruling of the 3090 1 Special Master, except as to page 219 where 2 lines 16 to 22 may be admitted. 3 Page 220, lines 13 through 25, the 4 appeal is sustained or, rather, the ruling on 5 appeal is affirmed. I'm sorry. 6 Next one was page 221, line 5, to 222, 7 line 5. Affirmed except for lines 5 through 14 8 may be used. 9 Page 222, lines 6 through 17, ruling 10 of the Special Master is affirmed. 11 So now you are on 227; right? 12 MR. GRALEWSKI: Yes, Your Honor. 13 227, 3 through 228, 12. 14 THE COURT: Okay. 15 MR. GRALEWSKI: We've rounded second 16 base. We've picked up the third base coach and 17 we are being waved home and have six left in 18 Williams. 19 227, 3 to 228, 12, Your Honor, is, 20 first of all, not hearsay by and large. The 21 chunk from 227, 3 to 228, 1, there are no 22 statements there by Mr. Williams being offered 23 for their truth. 24 The Plaintiffs would submit that to 25 the extent there is hearsay, it picks up at 3091 1 228, 2 through 12, and we believe it fits 2 squarely within Callahan because Williams is 3 repeating statements by his customers 4 explaining their motive, namely, why the 5 customers are going with Microsoft over DRI. 6 He repeats what he's being told by OEMs. 7 In sum and substance, there is a huge 8 machine behind Microsoft's products, et cetera, 9 et cetera. 10 That's 228. 11 THE COURT: Mr. Tuggy? 12 MR. TUGGY: Yes, Your Honor. 13 First section of this -- well, just by 14 way of background, the reason why there is no 15 ruling by the Special Master on this excerpt is 16 because the Plaintiffs added this after the 17 Special Master process. 18 The adding of this testimony was 19 pursuant to agreement, and we don't object to 20 that. 21 The first part of the testimony, 22 beginning at page 227, line 3, is a question 23 that references paragraph 158 of Mr. Williams' 24 declaration. 25 The question asked, in paragraph 158 3092 1 of your declaration you say that, quote, Egg 2 Head wrote DRI on May 16th, 1991, indicating 3 they would no longer carry the DRI product, end 4 quote. 5 As you know, this declaration by 6 Mr. Williams was submitted to the FTC in 1993, 7 and a copy of that has been provided to you. 8 If needed, I have a copy that I can 9 provide to the Court, but I think for the 10 purposes of this argument carrying forward what 11 we discussed last time about that declaration, 12 it itself is hearsay, and the examiner here is 13 trying to determine the foundation for 14 Mr. Williams' statement to this effect in the 15 declaration. 16 And Mr. Williams, in responding says 17 he does not have a copy of the Egg Head letter, 18 he does not recall what was said in the letter 19 or why, and doesn't recall being told by any 20 distributors the things that are said in the 21 letter. 22 So Plaintiffs apparently are offering 23 this for the truth of the statement in 24 Mr. Williams' declaration which is hearsay, 25 and, in addition, are doing so even though 3093 1 Mr. Williams admits he doesn't have personal 2 knowledge of the matters stated in his 3 declaration. 4 So, for that reason, that first 5 Section 227, line 3, to 228, line 1, is 6 inadmissible as hearsay within hearsay. 7 With respect to the second section 8 Mr. Gralewski referred to, that is page 228, 9 lines 2 through 12, these statements are not 10 statements by an OEM of an OEM's state of mind. 11 What Mr. Williams says is that he had 12 conversations with OEM and retail distribution 13 executives outside the company, he says, saying 14 that one of their real obstacles was the huge 15 machine behind Microsoft's products. 16 That's simply a statement of fact, a 17 belief about Microsoft's process for marketing 18 its products, and it's being offered for the 19 truth of that fact. 20 It's not being offered for an OEM 21 state of mind, and in particular, there is no 22 statement by the declarant here, Mr. Williams, 23 that an OEM made a statement of its state of 24 mind. 25 It's just OEMs making statements about 3094 1 Microsoft's marketing process. That is not 2 state of mind evidence and under Callahan 3 should not be admitted as such. 4 In addition, even if this were somehow 5 state of mind evidence, this certainly isn't 6 evidence of a then existing state of mind. 7 In other words, there's no statement 8 here by an OEM of its then existing state of 9 mind regarding its motives or plans with 10 respect to the marketing or sale of products by 11 Microsoft or Digital Research. 12 For those reasons, Microsoft's hearsay 13 objection to this entire section of testimony 14 should be sustained. 15 MR. GRALEWSKI: Your Honor, taking the 16 second chunk of testimony. 17 I fail to see the difference between 18 statements in Callahan where the customer said 19 I didn't buy beer because, and the statements 20 here where the customers say I didn't buy DRI's 21 products because. 22 In Callahan, the because was or the 23 motive was the prices were higher. In this 24 chunk of testimony, Your Honor, here the motive 25 or the reasons is because of the huge marketing 3095 1 machine that Microsoft has. 2 I think that these 11 lines of 3 testimony are directly analagous to the 4 Callahan situation. 5 As far as the first chunk of 6 testimony -- and often times during the meet 7 and confer process we accuse Mr. Tuggy of 8 having a photographic memory of all the 9 documents. So I think he'll correct me if I'm 10 wrong -- but this Egg Head document that's 11 referenced in paragraph 158 of Mr. Williams' 12 declaration, I believe, is clean and has been 13 or will be admitted as a DRI business record. 14 And I believe the document says, it's 15 a letter from Egg Head to DRI -- Egg Head was a 16 software distributor. I'm not sure they are 17 still around, but at this point in time they 18 were a software distributor. 19 And Egg Head wrote DRI and said, hey, 20 we're not carrying your products because we're 21 carrying Microsoft's products exclusively. 22 And so simply what's going on here is 23 that the examining attorney is asking the 24 witness about an otherwise admissible business 25 record. 3096 1 MR. TUGGY: If I may respond only to 2 the new argument about the otherwise possible 3 admissibility of the Egg Head letter. 4 I do recall that being an exhibit, but 5 I, frankly, don't recall the status of 6 objections to it. 7 Because -- but whether or not that 8 exhibit is admissible and could be the subject 9 of argument is not relevant to the 10 admissibility of this testimony because the 11 question is whether Mr. Williams' 12 characterization of the letter in his 13 declaration can be admitted for its truth, 14 which it can't both because he lacks personal 15 knowledge and because it's hearsay within 16 hearsay. 17 THE COURT: What's the state of mind 18 in lines 2 through 12 on page 228? Tell me the 19 state of mind. 20 MR. GRALEWSKI: The state of mind of 21 the OEMs in explaining the reasons why they are 22 not going to purchase DRI's products. 23 THE COURT: Where does it say they are 24 not going to purchase DRI's products? 25 MR. GRALEWSKI: Well, the whole -- 3097 1 those words probably aren't on those lines, 2 Your Honor, but this -- these 11 lines can't be 3 read in a vacuum. 4 This line of questioning concerns the 5 reasons why Microsoft -- I'm sorry, OEMs are 6 buying Microsoft's products instead of DRI's 7 products. 8 And Mr. Williams testifies, so let me 9 get to your answer basically, and he says, OEMs 10 and retail distribution executives -- so these 11 are his customers -- tell me that one of the 12 things that -- one of the real obstacles in 13 purchasing my product was this huge machine 14 behind Microsoft products. 15 And that's exactly the kind of quote 16 or a customer statement at issue in Callahan. 17 THE COURT: Anything else on this 18 issue? 19 MR. GRALEWSKI: Not from Plaintiffs, 20 Your Honor. 21 MR. TUGGY: Nothing further, Your 22 Honor. 23 THE COURT: Page 227, lines 1 through 24 12 are hearsay and not admissible. The 25 remaining lines on 227 to 228 are admissible. 3098 1 We're going to take a brief break 2 here, so I can get the Jury notebooks. Then we 3 will bring the Jury in. 4 MR. GRALEWSKI: Thank you, Your Honor. 5 MR. CASHMAN: Thank you, Your Honor. 6 (A recess was taken from 8:29 a.m. 7 to 8:35 a.m.) 8 THE CLERK: All rise. 9 THE COURT: Everyone else may be 10 seated. Thank you. 11 You may proceed. 12 MS. CONLIN: Thank you, Your Honor. 13 Good morning. 14 We were just about done with GO, this 15 little computer here. As a matter of fact, the 16 page that I'm beginning on is titled the end, 17 and it's the end of GO that I'm talking about. 18 One of GO's partners is a company 19 called EO, and that is a part of AT&T. 20 In mid-1991, GO spins the hardware 21 manufacturing part of its company off to EO, 22 and then a little bit later, AT&T proposes that 23 it buy GO. And GO merges with AT&T, and, 24 shortly thereafter, GO is gone. 25 But what about Microsoft's pen 3099 1 project? Microsoft's Winpad, the software that 2 Microsoft touted all over the world and sold to 3 or licensed to OEMs, it's vaporware. It is 4 never released. It never comes out. 5 No final version ever ships of 6 Microsoft's Winpad. 7 And the evidence will be that with no 8 competition from GO, Microsoft simply doesn't 9 finish the product. And that's the end of the 10 tablet computer until after the millennium. 11 You can get one now, as of November of 12 -- I beg your pardon. November 2002 is when 13 finally pen handwriting tablets of this sort 14 are finally available to consumers. 15 You can see the similarities with the 16 GO tablet back in the late '80s and early '90s. 17 The pen -- this is the GO, and that's the 18 current tablet PC. 19 So consumers wait for ten years until 20 -- for a convenient handwriting technology. 21 You may recall from Monday the 22 document, which is 310A, where Microsoft says 23 it will crush GO, and GO is crushed and 24 consumers wait. 25 What will Microsoft's evidence be 3100 1 about this GO PenPoint computer? Was it really 2 a good product? Of course, as with any 3 complicated, brand-new, groundbreaking 4 technology, there were lots of challenges. 5 There were lots of obstacles. 6 Handwriting recognition is very hard. 7 There's no question about it. Even today you 8 can't get 100 percent reading of handwriting. 9 It was also difficult, Mr. Kaplan will 10 say, to figure out how to build a screen. 11 We're kind of accustomed now with PDAs and that 12 sort of thing, you know, to a screen that can 13 sense whether a letter is a B or a D or an E or 14 an I, but at the time this was being developed, 15 that was very, very unusual. And, in fact, 16 didn't exist. 17 And, of course, it always takes time 18 and money to design any new product. 19 Like most products of the era -- I do 20 want to make one thing clear. The lugable that 21 we showed you -- I didn't ask Mr. Smith to 22 bring it back. He's perfectly capable of doing 23 that. But that was from the '80s. 24 By the early '90s, there were tablet 25 computers. I beg your pardon. There were 3101 1 laptop computers, not tablet computers, but 2 laptop computers. They were pretty clunky too, 3 but nothing like the 1987 lugable. 4 But like most products, the IBM 5 tablet, which State Farm used, I think it 6 cost -- I think Mr. Kaplan says that cost about 7 5,000. The manufacturer suggested retail price 8 was about 5,000. And -- no, I guess that was 9 about 3,000. And then it dropped to between 2- 10 and 4,000, the right price. You will hear 11 that, I think, a lot. 12 It was about the same as the bigger 13 laptops in 1992. 14 Microsoft will also produce evidence 15 that Mr. Kaplan wrote a book after he -- after 16 the business failed and he went on to other 17 things. The book is called Start-Up. 18 When he wrote the book, he did not 19 know -- he didn't have access to the documents 20 that you have seen and will see. He would 21 certainly write it differently today. 22 Microsoft will also point out that in 23 this GO computer, the operating system and the 24 user interface, the GUI, were combined, and the 25 implication is that GO is just like Microsoft, 3102 1 like Microsoft Windows in '95, but GO was a 2 little start-up company and Microsoft was and 3 is a monopoly controlling access to an 4 essential part of computers. 5 Also, be clear that Mr. Kaplan makes 6 no claim to having invented laptop computers. 7 This is a tablet computer, and he does claim, 8 and has a right to claim, that he and his 9 company were the creators of this. 10 He conceived it, he designed it, and 11 he's entitled to say so. 12 Microsoft's examination of Mr. Kaplan 13 may seek to confuse you on that point. 14 There were also -- besides the heavier 15 laptops, there were also little PDAs that came 16 out right around the same time. But, of 17 course, you couldn't use the laptop with 18 handwriting. And the PDAs at that time had 19 very little functionality. 20 Microsoft also makes the point -- we 21 believe their evidence will be that Mr. Kaplan 22 does not think he is developing a product for 23 the desktop market. 24 Even though it runs on the desktop, 25 Mr. Kaplan thinks he's developing a product 3103 1 that is a tablet computer. 2 It is Microsoft itself who targets GO 3 as a competitor on the desktop. And, indeed, 4 in some of the documents that we've already 5 seen, Microsoft says GO is trying to take over 6 the laptop. 7 So even though Jerry Kaplan doesn't 8 want to take over the desktop -- that's what I 9 meant, the desktop -- Microsoft seems to think 10 he does or he might or he could. 11 Microsoft will also focus on an 12 agreement, which we've touched on, between IBM 13 and GO. 14 GO asks IBM to build a dedicated 15 computer that only runs PenPoint because if 16 it's built to run both Pen Windows and 17 PenPoint, it will have to be heavier. It will 18 have to be larger in size. It will have to 19 have a disk. It will be more fragile, more 20 likely to break, and clumsier. 21 GO asking IBM to build a dedicated 22 machine is nothing like Microsoft telling OEMs 23 that they can only get the operating system if 24 they buy Windows or only get Windows if they 25 buy the operating system. 3104 1 It is not an exclusive arrangement 2 between GO and IBM of the sort that we believe 3 the evidence will show Microsoft forced OEMs 4 into. 5 When little GO asks IBM not to 6 cooperate with Microsoft on the pen computer, 7 it has no leverage. It can't make threats that 8 will kill IBM's business. Unlike with 9 Microsoft, IBM is free to agree or disagree. 10 When Microsoft, on the other hand, 11 says do not cooperate with so-and-so, they can 12 and have, in fact, backed up those demands by 13 withholding products or withholding information 14 or betas. You've also seen some of those 15 documents. 16 We believe that the evidence will show 17 that Microsoft used against GO several 18 exclusionary tactics. 19 The exclusionary license practices, 20 vaporware, the beta blacklist, difficulties 21 honoring its contracts with GO, FUD, threats 22 and intimidation, and technology espionage. 23 Does GO fail because it's a poor 24 product, its marketing isn't good enough or it 25 doesn't sell many systems or the schedule slips 3105 1 or all of the other things? We believe the 2 evidence will show that Microsoft's conduct was 3 a cause of GO's failure. 4 Instruction 14, the Court said, quote, 5 the Plaintiffs do not have to prove that an 6 antitrust violation was the sole cause of their 7 injuries. 8 What we have to show is that it was a 9 cause and that it was a material cause, but it 10 need not be the only cause. And we will prove 11 that Microsoft's conduct was a cause of GO's 12 demise. 13 The evidence will show that Microsoft 14 was unwilling to compete with GO on the merits. 15 And part of the reason for that, 16 according to its own executives, it was a year 17 or more behind. Indeed, no product was 18 released until 2002. And consumers did not 19 have access to tablets for ten years. I 20 believe the evidence will be that that was 21 because of Microsoft's anticompetitive conduct. 22 Let me move on to OS/2. 23 In the mid-1980s, IBM and Microsoft 24 agree to jointly develop a new desktop 25 operating system to be called OS/2 with the 3106 1 stated goal between Microsoft and IBM of making 2 the world's most advanced commercially 3 available operating system. 4 Returning for a moment to DRI. 5 Almost from the beginning, almost from 6 the very beginning, MS-DOS was kind of an aging 7 operating system. 8 When originally written by Tim 9 Paterson of Seattle Computer, QDOS, quick and 10 dirty operating system, that Mr. Paterson 11 wrote, he wrote that as a clone of CP/M, which 12 was the DR, Digital Research product, and CP/M 13 was the operating system of the '70s. 14 So the core of the MS-DOS technology 15 is not designed for advanced functionality. 16 When MS-DOS was taken from QDOS, which 17 was taken from CP/M, it was -- there was no 18 advanced functionality. 19 The APIs for MS-DOS are not robust, 20 nor are there standard ways of creatively 21 solving some of its inherent limitations. 22 This results in some software just not 23 playing very well together because the various 24 software manufacturers, the various software 25 developers, are doing things differently. 3107 1 For another thing, PC users running 2 MS-DOS cannot multitask. They cannot do more 3 than one thing, run more than one application 4 together. And MS-DOS doesn't have protected 5 memory, which means that one program may 6 overwrite the data from another program at any 7 time. 8 Finally, MS-DOS suffers from this 640 9 K barrier. We've talked a bit about the memory 10 limitations. These are the memory limitations. 11 And that hampers an ISVs ability to harness all 12 the power that hardware has, computer hardware 13 has. 14 And even as MS-DOS is aging, there are 15 huge improvements in the underlying hardware. 16 By the mid-1980s, computer chips, Excel, the 17 processors, Excel-compatible processors are 18 getting a lot faster. 19 In order to take advantage of these 20 improvements in the chips, in the hardware, a 21 computer's operating system must be coded in a 22 way that supports those new features. And 23 MS-DOS is not. 24 As part of the joint agreement, the 25 JDA between Microsoft and IBM, a brand-new GUI, 3108 1 graphical user interface, is developed. It's 2 called PM or presentation manager, and it's 3 designed specifically for the new operating 4 system OS/2. 5 Microsoft already has a GUI. It is 6 Windows. And it was developed prior to the 7 joint development agreement between IBM and 8 Microsoft. 9 It was -- Windows was, of course, 10 originally designed to run only on MS-DOS, and 11 MS-DOS -- or Microsoft is already actively 12 marketing Windows. 13 And the first Windows -- I don't know 14 if we've talked about this -- the first Windows 15 1.1, 1.0 came out in 1985. But even early on, 16 Microsoft toys a bit with the idea of making 17 Windows the dominant desktop GUI system. 18 And despite concurrent efforts by IBM 19 to develop presentation manager for OS/2, 20 Microsoft actually spends a good deal of time 21 and research and money prior to 1989 trying to 22 make Windows run on OS/2. 23 Nathan Mhryvold, who you've heard from 24 before, who is sort of the strategic technology 25 guru at Microsoft, gets very frustrated with 3109 1 Microsoft's efforts in connection with putting 2 Windows on OS/2 and with respect to the 3 confusion at Microsoft between presentation 4 manager, PM, and Windows. 5 And he writes to Bill Gates and Steve 6 Ballmer complaining about Microsoft's inability 7 to properly plan and execute on this product. 8 And this is Plaintiffs' Exhibit 19, 9 and it is a very lengthy memo, and it is -- 10 there's quite a bit of coarse language in this 11 document. 12 And I'm going to read a relative -- 13 well, given the size of the memo, this is a 14 relatively brief part that I want to focus your 15 attention on. 16 Mr. Myhrvold's e-mail, lengthy e-mail 17 dated -- now, this is one of those documents 18 that has two dates on it. Can you see that at 19 the top? There's your top date of June 10th of 20 1988 and the bottom date of March 9th of 1992. 21 So from the context, we believe this 22 is an older -- that the older date appears to 23 be given what seems to be going on in the 24 company at the time. 25 Here's what he says. Our systems 3110 1 software strategy is a rambling litany of 2 failure. We have patched together a bunch of 3 shitty technology with the incredible personal 4 performance -- performances of a couple of 5 outstanding individuals. 6 I could list endless examples of this, 7 but one will suffice. The near miss we had 8 with presentation manager. We lost the fucking 9 business, and then got it back, but as a guy 10 who was there, let me tell you, it was close. 11 This scared the shit out of me. 12 Stop and think for a moment what it 13 would have meant to have lost that, what the 14 fuck -- what the fuck would that leave us? We 15 would be way up shit creek without PM, 16 presentation manager, the GUI for OS/2. 17 Recall that this would not just have 18 meant losing to IBM. They pitched Lotus about 19 doing it -- making the PM GUI. 20 The only reason that we got the PM 21 business was that we had Windows, a piece of 22 technical that we could both threaten IBM with, 23 and also position as something we could bring 24 to the party. In terms of its origin, however, 25 Windows was a complete accident. 3111 1 In my view, Microsoft is a 2 nontechnical company. The one thing that we 3 know how to do is compete with an existing guy 4 who might have some vision, and that is how we 5 got into the Windows business. 6 The Windows story was not all happy. 7 We had absolutely no idea as to what it would 8 take to put it on OS/2. Absolutely none. 9 No one had thought through even the 10 most basic issues for making this stuff work 11 together. 12 There has been an incredible amount of 13 work in both the base and in PM that we have 14 had to stumble our way through. There was also 15 a lot of work that IBM made us commit to which 16 we did not understand. 17 The clear message that I got from the 18 PM experience was that our system software 19 strategy was virtually nonexistent. We have a 20 neat bag of tactics to limp between one IBM 21 crisis to another, but no idea of how to keep 22 them in the long term. 23 We can lie, cheat, or steal in the 24 heat of the moment to get over a critical 25 issue, like the PM business. Doing this 3112 1 repeatedly without coming up with any solutions 2 or even any plan for a solution can only be 3 called one thing, mismanagement. 4 We have seriously and stupidly 5 mismanaged the most valuable software asset in 6 the world. 7 This goes out to executives. But even 8 after this happens, there's still confusion 9 about how Microsoft should move forward. 10 Should Windows be with OS/2 or should it be PM? 11 And where should Microsoft be? Does it stay 12 with its MS-DOS, even though it has its JDA, or 13 does it commit its resources solely to OS/2? 14 Let's talk a bit about the technical 15 superiority of OS/2. 16 The DOS plus Windows combination, two 17 separate products, MS-DOS, Windows, is inferior 18 technology to OS/2 and presentation manager, 19 and everyone, including Microsoft, including 20 Bill Gates, is aware of that. 21 In fact, Bill Gates refers to OS/2 as 22 the big brother of Windows and admits it's 23 better than Windows. 24 This is Plaintiffs' Exhibit 126. 25 E-mail re: Windows PR. 3113 1 He says, as far as OS/2 goes, we are 2 working internally on how we position OS/2 as 3 the big brother to Windows. Maybe we should do 4 some naming changes. 5 OS/2 should be viewed as Windows plus. 6 We need to make it clear that Windows success 7 ensures OS/2 success since OS/2 is better. 8 Nathan Mhryvold also knows that the 9 Windows plus DOS combination is inferior 10 technology, but Mr. Mhryvold, Nathan Mhryvold 11 -- there are two Mhryvolds, they are brothers. 12 Both work at Microsoft. 13 Nathan is just the one we've heard 14 from and Cameron we will hear from more in 15 connection with the applications business. 16 This is Nathan. Nathan thinks it's 17 important to come up with some product that can 18 achieve market dominance because of how much 19 money Microsoft can make on a monopoly. 20 This is Plaintiffs' Exhibit 258. It's 21 called Microsoft system software focus. 22 Mr. Mhryvold says, we are presently 23 engaged in a number of battles. The most 24 important one, in my mind, is to retain our 25 dominant position for desktop PC operating 3114 1 systems in the midst of changing to new 2 technology base to escape the inherent 3 technical limitations of DOS. That is what I 4 just talked to you about. 5 This is proving much harder than we 6 originally thought. 7 OS/2 will not be the near term 8 successor to DOS, and it may never be. Windows 9 is only a temporary solution to the problem 10 because it still has severe internal 11 limitations, IE based on DOS, hard to put 12 security in, so on. 13 Nevertheless, this is our calling. We 14 are in a terrific position to achieve this, and 15 it would be crazy to abandon it. The size of 16 the opportunity is enormous. There is a huge 17 value to a monopoly, and we have the position 18 and skills that it makes sense for us to shoot 19 for it. 20 OS/2 breaks new ground in the PC 21 industry by introducing support for a lot of 22 advanced new features. Preemptive 23 multitasking, multithreading, interprocess 24 communication features, virtual memory support, 25 fully protected memory, dynamic linking library 3115 1 support. 2 I can see that you all know all about 3 this. Don't worry. There will be explanations 4 for these differences, these advances that OS/2 5 includes in the operating system. They will 6 not, however, come from me at this time. 7 And also I wanted to -- OS/2, this is 8 OS/2 2.0. This was produced by IBM. We'll get 9 to the part of the story where they go their 10 separate ways, IBM and Microsoft. 11 None of these things, preemptive 12 multitasking and multithreaded and DLL support 13 and so on, none of those things are supported 14 at first by MS-DOS. 15 The difference between the two 16 operating systems is very, very great. 17 One, OS/2 is well designed, stable, 18 and easy to use. 19 The other, the MS-DOS Win combination 20 is unstable, tricky, and as Anthony Short, who 21 was the director of OS/2 development, writes 22 bluntly to Steve Ballmer and Brad Silverberg in 23 1991, OS/2 just works, whereas, MS-DOS requires 24 what Mr. Short calls black magic. 25 This is Plaintiffs' Exhibit 601. 3116 1 Finally, with OS/2, you just install 2 it and it works. You don't have to fiddle 3 around with UMBs. That's upper memory blocks. 4 These things are completely black magic. 5 You have to try loading different 6 combinations of things up there and watch to 7 see if your machine crashes. This is the most 8 important difference between DOS and OS/2. You 9 should try to install DOS and Windows and 10 Landman on PS/2. That's personal systems 2, 11 which was an IBM piece of hardware at the time. 12 For corporations and governments, 13 large clients, the advanced features of OS/2 14 are particularly important and corporations and 15 governments adopt OS/2. They make -- at least 16 they get internal plans. They make plans to 17 adopt OS/2. 18 The high end versus the low end. So 19 virtually everybody agrees that OS/2 is a 20 better, more advanced, more robust operating 21 system. 22 And one of the things that makes OS/2 23 more advanced is that it is optimized for 24 Excel's newest processors, the 386 and 486 25 series. 3117 1 However, Microsoft decides to abandon 2 OS/2 on the desktop in favor of the MS-DOS 3 Windows combination. 4 You will hear more about the reason 5 for Microsoft's decision, but let me give you a 6 quick preview. 7 Microsoft sees a Windows' centric 8 strategy as a means not only to dominate the 9 operating system market, but the evidence will 10 show that clear back, Microsoft saw Windows as 11 a way of seizing control of the important 12 applications markets as well. 13 That's the word processing and 14 spreadsheet markets, which are a part of this 15 case. 16 On August 14, 1990, Paul Maritz 17 distributes a memo to Gates, Ballmer, and other 18 members of the executive staff containing the 19 brand-new strategy as of August of 1990 for 20 committing to Windows and dead-ending 21 presentation manager. 22 This is Plaintiffs' Exhibit 365. And 23 Mr. Maritz says, do not forward a copy -- do 24 not forward or copy. We do not want press 25 speculation on the contents of this memo. 3118 1 We have decided to use Windows instead 2 of PM. This change raises -- you know, I'm 3 skipping these things because, otherwise, we'll 4 just be here even longer than we have been. 5 You will have the whole document. I 6 am trying to keep the context clear. 7 We've decided to use Windows instead 8 of PM. This change raises a question about 9 what our strategy is for Windows, OS/2, and NT 10 OS/2. That is -- NT means their new 11 technology. Has to do with a networking 12 system. 13 In light of our ongoing discussion 14 with IBM, many factors are subject to change. 15 The following points are not public and all of 16 our comments to ISVs, applications developers, 17 OEMs, the press, et cetera, should not 18 communicate any change in strategy for Windows 19 or OS/2. 20 He goes on to say, our long-term 21 product strategy is Windows on DOS and Windows 22 on OS/2. 23 Our goal is to simplify our message to 24 ISVs and focus all of our development and 25 marketing energy behind the Windows API. 3119 1 Without announcing a change -- without 2 announcing a change in strategy, we are telling 3 ISVs to write for Windows. 4 The model we want to present to 5 customers and ISVs is a single, consistent 6 Windows interface running on DOS, on the 32-bit 7 OS/2 base, and eventually on NT. 8 As radical as this may sound, very 9 little has changed near term and we want to 10 avoid giving the appearance to customers or 11 ISVs that we are abandoning OS/2. 12 Then he does this sort of series of 13 questions and answers in this document. 14 Does this mean that PM is dead? And 15 the answer he provides is no. It is 16 dead-ended. 17 Microsoft PM development will not 18 continue beyond Cruiser. Cruiser is OS/2 -- 19 these code names are -- Cruiser is OS/2 2.0, 20 this OS/2 2.0. 21 And we will discourage or stop 32-bit 22 PM applications development as we will not 23 migrate that API forward. 24 Does IBM buy into dead-ending PM? No. 25 That is why we should be careful what we say in 3120 1 public to ISVs and OEMs. 2 Does this mean we are killing OS/2? 3 No. We are starting down a path toward merging 4 Windows and OS/2 functionality. 5 What should we tell corporate 6 accounts? For the time being, nothing 7 different. 8 What should we tell ISVs, the software 9 developers? Absolutely nothing. We must be 10 extremely careful that our change in strategy 11 not be leaked to the press, particularly prior 12 to reaching an agreement with IBM. 13 What do we plan to tell the press? 14 Nothing. We have a lot to lose and nothing to 15 gain by communicating that we are changing our 16 strategy before we know what the implementation 17 will be. 18 On August 15, 1990, Silverberg, Brad 19 Silverberg, Paul Maritz, and a guy whose name 20 you haven't heard yet I don't think, Bob -- I 21 think it is Muglia, M-u-g-l-i-a, produce a memo 22 further planning their upcoming dealings with 23 IBM in which they discuss PM and what to do 24 with OS/2. 25 The stated objective of the memo is to 3121 1 figure out how to make Windows dominant. And 2 this is Plaintiffs' Exhibit 9654. 3 And in that document Microsoft -- 4 written here, Microsoft says assertion. We 5 should absolutely minimize our support for PM. 6 In the end, what Microsoft decides to 7 do is give OS/2 back to IBM. As this memo 8 discusses, there are a number of reasons for 9 this choice. 10 By transferring OS/2 to IBM, Microsoft 11 can free up over 150 people to work on projects 12 which provide significant benefit for the 13 company. This also provides us with a great 14 way to get out of the PM mess. Don't do 15 anything except develop and announce MS plans 16 to enhance Windows. 17 For OS/2 2.0, this means: 18 One, transfer control of OS/2 2.0, and 19 all future release to IBM ASAP. 20 Number 2, don't waste any energy on 21 OS/2 content. 22 And, Number 3, don't do an OS/2 2.0 23 packaged product. 24 The decision to abandon OS/2 and focus 25 on Windows, according to Microsoft during this 3122 1 time frame, must be kept a secret. 2 Paul -- the orders that Mr. Maritz 3 gave in his Microsoft memorandum on August 4 14th, 1990, are followed by Microsoft's 5 marketing and PR forces because more than a 6 month later, on September 17th, 1990, Collins 7 Hemingway of Weggener Edstrom, the outside PR 8 firm, distributes to U.S. marketing a rude Q 9 and A about Microsoft strategy. 10 Rude Q and A. We've seen these 11 before. Microsoft's attempt to prepare for any 12 questions they think might come. 13 And this certainly reflects 14 Mr. Maritz's admonition not to disclose their 15 change in strategy. 16 This is Exhibit 389. IBM Q and A. 17 Is the JDA, joint development 18 agreement, dead? No. The JDA between IBM and 19 Microsoft is extended. Q, does this signal a 20 change in the joint IBM/Microsoft fall COMDEX 21 1989 announcement? No, this does not signal a 22 change. 23 And that is said despite 365 in which 24 Mr. Maritz says we have decided to use Windows 25 instead of PM. 3123 1 Doesn't this signal that Microsoft is 2 not fully committed to OS -- not as fully 3 committed to OS/2 as we thought? No. We are 4 fully committed to OS/2. 5 Though Mr. Maritz says a month earlier 6 it's dead-ended. 7 Is Microsoft giving up on OS/2 2.0 8 development because of the success of Windows 9 3.0? We are not giving up on OS/2 2.0 10 development. 11 Though Mr. Maritz says our goal is to 12 focus all our development and marketing energy 13 behind the Windows API a month earlier in 14 Exhibit 365. 15 Question: The industry sees a 16 competition between OS/2 and Windows. Doesn't 17 this signal that Microsoft is betting on 18 Windows in the short run? 19 No. We will provide technology that 20 will leverage the popularity of Windows and 21 Windows applications to OS/2. 22 And this, despite Mr. Maritz's 23 direction, we will discourage or stop 32-bit 24 applications development. 25 How many copies of OS/2 have shipped 3124 1 to date? Shipments aren't what you expected, 2 right? 3 Answer: While our initial 4 expectations for OS/2 were very aggressive, 5 OS/2 is meeting and exceeding our current 6 expectations. 7 Q. Is Microsoft still committed to 8 doing presentation manager applications? You 9 bet. 10 That, of course, is directly contrary 11 to what has been decided more than a month 12 earlier. 13 And this strategy of stalling and 14 misrepresenting goes on from September until 15 early March of 1991. 16 And this time, without any fireworks, 17 is best summed up by a March 4th, 1991, e-mail 18 exchange between Cameron Mhryvold, Nathan's 19 brother, and Bob Muglia, and Bob Jones, who is 20 a special marketing manager at Microsoft. 21 This is Plaintiffs' Exhibit 592, and 22 Mr. Jones says, do you think we can lull IBM to 23 sleep for another year by telling them that we 24 believe this while we steal more oil wells with 25 Windows? 3125 1 Microsoft declares war on OS/2. Those 2 are not my words. Those are Microsoft's words. 3 It starts with an OS/2 slogan called 4 better Windows than Windows. 5 In early March 1991, Pam Edstrom of 6 Weggener Edstrom is notified that the Wall 7 Street Journal is going to do an article on 8 OS/2 and Windows. 9 And, from Microsoft's perspective, 10 it's not going to be a friendly article. And 11 for one thing it will highlight some of the 12 problems with Windows, but, more importantly to 13 Microsoft, it will contain a claim by Lee 14 Riesweig, an IBM executive who is in charge of 15 OS/2, that OS/2 is a, quote, better Windows 16 than Windows because of its ability to run 17 applications, Windows applications out of the 18 box. 19 One of the reasons Microsoft is 20 worried about this claim is that OS/2 is 21 developed all along to be an advanced operating 22 system. That was what they intended from the 23 very beginning. 24 Windows, on the other hand, suffers 25 from the problems we talked about. The 3126 1 problems of robustness, memory, and so on. 2 On April 24th, 1991, Silverberg writes 3 to Gates, to Ballmer, and others expressing his 4 concern about the quality of Windows. 5 This is Plaintiffs' Exhibit 699. 6 April 24th, 1991. 7 He says -- Mr. Silverberg says, while 8 we are all very upset with IBM and may severely 9 doubt their ability to deliver better DOS than 10 DOS, better Windows than Windows, one thing is 11 very clear. IBM has hit a very raw nerve with 12 our customers about Windows robustness. People 13 are sick and tired of UAEs. Unexpected 14 application errors, UAEs. Not good things. 15 I never thought I'd see people 16 actually rooting and cheering for IBM. But 17 it's happening now. 18 A few weeks after the Wall Street 19 Journal article appears and the big IBM better 20 Windows than Windows announcement, the two 21 companies, Microsoft and IBM, meet to discuss 22 what to do going forward. 23 On April 20th, 1991, Tony Audino sends 24 an e-mail recap of the meetings to Gates, 25 Silverberg, Joachim Kempin, Paul Maritz, and 3127 1 Steve Ballmer, and writes in Plaintiffs' 2 Exhibit 689A, subject IBM briefing. Jim and 3 Joe, IBM people, made it clear IBM was going it 4 alone. It is now undeniably clear that we have 5 a major competitor in the OS business. 6 Knowing that IBM is now competing 7 against them, Microsoft circles the wagons and 8 quite literally declares war on IBM. 9 On April 22nd, 1991, Silverberg writes 10 an e-mail to Ballmer asking him to give a 11 speech. 12 This is Plaintiffs' Exhibit 691. 13 There are other documents about this, and I was 14 going to read some others of them, too, but I 15 cut them out. 16 What's going on here is they are 17 planning to have a big meeting, you know, of 18 lots of employees, and Mr. Silverberg, who's 19 the planner of the meeting, wants Mr. Ballmer 20 to come along personally. 21 He says, I would really like you to be 22 there to give the declaration of war speech. 23 And, apparently, that does occur. 24 Mr. Ballmer comes along and does give the 25 declaration of war speach. But because of what 3128 1 we've already discussed in connection with DRI, 2 the per processor licenses, the minimum 3 commitments, the PPBs, the long terms, the war 4 is over before it even begins. 5 It's a foregone conclusion that 6 Microsoft, because of these exclusionary 7 contract terms, is going to prevail. 8 The per processor pricing scheme, as 9 you will recall, offers significantly lower 10 prices if the OEM is willing to pay for every 11 machine of a particular processor type that 12 goes out the door, whether or not the computer 13 has the Microsoft operating system on it. 14 Therefore, if an OEM did want to ship 15 machines with OS/2 on them, they would have to 16 pay for two licenses. The per processor 17 license they're paying to Microsoft and then 18 they would have to pay IBM for the OS/2 19 license. 20 Microsoft clearly realizes how 21 important these exclusionary licenses -- these 22 exclusionary license restrictions are. 23 Mr. Kempin, worldwide director of OEM 24 sales at Microsoft, says -- this is an e-mail 25 between he and others. 3129 1 And he responds to an observation that 2 our revs margin and OEM business will decrease. 3 Our revenues, our margins in the OEM business 4 will increase because of OS/2. And here he 5 says no. He says, no, our revenues and margins 6 will not decrease. This is Plaintiffs' Exhibit 7 690. 8 He says, Mr. Kempin says, decrease in 9 margin only if people vote to buy retail from 10 OS -- retail OS/2 from IBM, which I doubt. 11 All key OEMs are committed to DOS 5.0 12 and Win 3.x for the next two, three years. 13 Most of them per processor or at least per 14 system. This will give us time to respond 15 technically. 16 In other words, Microsoft recognizes 17 that IBM is effectively locked out of the OEM 18 channel. And that is the most important 19 channel for the distribution of operating 20 systems. 21 We will prove these effects are not 22 accidental. They are designed, they are 23 implemented specifically to lock out 24 competition and to preserve Windows market 25 share. 3130 1 On April 12th, 1991, shortly before 2 the declaration of war against IBM, there is a 3 management conference, Microsoft management 4 conference, held in a place called Sudden 5 Valley, and one of the topics for discussion is 6 how to push Windows, separate GUI product, how 7 to push the Windows market share up to 90 8 percent. 9 Jeremy Butler writes a summary of the 10 conference. 11 This is Plaintiffs' Exhibit 675. Memo 12 re: Presentations given at the Sudden Valley 13 management conference. 14 Our vision is that all personal 15 computers use a graphical user interface. For 16 the vast majority of workstations, that means 17 Windows. 18 How can we assure that Windows is used 19 on 90 percent of all machines as quickly as 20 possible? The answer to that question is 21 provided on the next page of the memo, the next 22 page of 675, and it includes bundling. 23 This is -- the title of this bullet 24 point is Microsoft levers -- what Microsoft has 25 that can get it to 90 percent share of the 3131 1 Windows -- on the desktop of the Windows GUI. 2 Bundling. Put it together with the 3 already dominant, already monopoly DOS. Dealer 4 incentives. Pricing. One Windows worldwide. 5 And then on the next page -- actually, 6 we skip two pages. What must be done in the 7 market? Strong identification with DOS. A DOS 8 5.0 RUP, R-U-P, bundle. We think that's retail 9 upgrade. We've got different opinions about 10 what it was from Microsoft folks. 11 And it may mean different things in 12 different contexts too. But that appears to be 13 a retail upgrade bundle. Put the two products 14 together. 15 Combined setup with DOS. Promote as 16 preferred DOS shell. Remove DOS shell. 17 DOS comes with a little shell of its 18 own which pretty soon is removed. This is 19 bundling strategy, of putting the two products, 20 the monopoly product, the then monopoly 21 product, DOS, MS-DOS, together with Windows GUI 22 is, again, mentioned as an effective technique 23 against Microsoft in a memo from Tony Audino to 24 Gates, Silverberg, Ballmer and others about the 25 failed negotiations with IBM on April 20th, 3132 1 1991. 2 After Mr. Audino recounts IBM's 3 statements of continued commitment to OS/2, he 4 lists a few tactics that can be used against 5 them. 6 He writes in Exhibit -- in Plaintiffs' 7 Exhibit 689A, 8, we need to aggressively pursue 8 OEMs with DOS/Windows bundle deals as a 9 preemptive strike against IBM. 10 If they, these are the OEMs, have per 11 system deals with us on Windows now, they will 12 be hard-pressed to do deals with IBM on OS/2 13 2.0. 14 The reason, once again, that OEMs 15 would be hard-pressed to work with IBM is 16 because of the nature of per processor pricing. 17 Nobody wants to pay for a product they 18 can't use. If they are already paying 19 Microsoft for every machine sold, then Windows 20 is what they'll have to bundle. They can't 21 afford to seek out another licensing deal with 22 IBM at the very same time. 23 Microsoft knows that. 24 Another tactic that Microsoft adopts, 25 and one that we have seen before, is the 3133 1 creating incompatibilities between products. 2 At the outset of Microsoft's war 3 against IBM, it is, as I have mentioned, clear 4 that OS/2, IBM, has a superior operating system 5 on the merits. 6 Steve Ballmer is trying to find a way 7 to defeat the better Windows than Windows 8 claim. 9 On March 8, 1991, David Cole sends 10 Mr. Ballmer an e-mail, which appears to be in 11 response to some prior request from 12 Mr. Ballmer, informing Mr. Ballmer that it will 13 be very hard to prevent IBM from running 14 Windows on OS/2, even if they do all the tricks 15 and traps that they know of. 16 Here's Mr. Cole's memo. And it's 17 Plaintiffs' Exhibit 601, and the title of it is 18 e-mail, Subject: Win 3.1 and DPMI. 19 DPMI -- again, we will leave it to the 20 experts to explain that, but the letters mean 21 DOS protected mode interface. And that is a 22 method prescribed by Microsoft for creating 23 compatibilities between OS/2 so Windows 24 applications could be run on the OS/2 operating 25 system. 3134 1 And here is what he says. Just to let 2 you know what's up/down on our end. Preventing 3 IBM from running Windows in protect mode on 4 OS/2 2.0, will be very hard, if not impossible. 5 We can do all the tricks and traps we 6 know of to give them a hard time, but the fact 7 that they get source to everything means they 8 could figure it out. 9 Because of the agreement between IBM 10 and Microsoft, IBM gets access to the source 11 code for the products. 12 The best we can do is stall them. Let 13 me know what we are really trying to do here 14 and I'll -- probably I'll do the best I can on 15 this end to accommodate. 16 Mr. Cole continues to hunt for ways, 17 for tricks and traps, that will harm OS/2. 18 And just as Mr. Ballmer leans on him 19 to find ways to do it, he's asking others. 20 He's asking for suggestions from others. 21 And one of these people is a 22 programmer named Tony G. He writes to David 23 Cole on March 7th, 1991. This is Plaintiffs' 24 Exhibit 7558. And the subject is they don't 25 stand a chance. 3135 1 He says, On further investigation, 2 OS/2 doesn't stand a chance of running the 3 current kernels for the following reasons. 4 One, selector tiling. And selector 5 tiling comes up several times. 6 The second is a suggestion that we 7 should put a version check in this routine and 8 abort the boot under OS/2, exclamation point. 9 Yes, that exclamation point. 10 Okay. A version check in this routine 11 and abort OS/2 is probably apparent to you, but 12 the idea is to figure out whether or not what's 13 going on is that OS/2 is running and if OS/2 is 14 running or is the underlying operating system, 15 then it won't -- it won't work. 16 We remember this with DRI, Bambi, 17 nested task. This is a suggestion to do the 18 same kind of thing in this. 19 And Aaron Reynolds, who we saw, Aaron 20 knows some other gotchas, which they would have 21 to work around too. 22 David Cole thinks that the selector 23 tiling issue may have some potential. He 24 responds to the e-mail from Tony G. and copies 25 a bunch of people including another programmer 3136 1 named Earl Horton asking them to do some more 2 work on the issue. 3 It's a continuation in 7558 with the 4 subject matter. They don't stand a chance. 5 Can you guys do me an additional favor 6 on this and follow up with one of the old 7 Cruiser guys about the tiling system? Scott 8 Lu. maybe -- Scott Ludwig is an old Cruiser 9 guy. This is incredibly key. 10 Now, that would be -- check with a 11 Microsoft employee who has worked on OS/2 and 12 see what we can do with the selector tiling 13 issue to abort OS/2. 14 On March 8th, 1991, Earl Horton sends 15 Scott Ludwig, e-mails Scott Ludwig about the 16 problem, and he says, we are being asked to 17 make it impossible to run Windows 3.1 in 18 protected mode under any DPMI -- DPMI server on 19 OS/2 1.x. 20 The evidence will be that what 21 Mr. Ballmer intends to do and what he has David 22 Cole and others searching for is the ability to 23 -- is to destroy the ability of OS/2 to run 24 Windows applications. 25 Okay. You remember applications 3137 1 barrier to entry is something that Judge 2 Rosenberg has talked to you about the findings 3 in fact. We'll be talking about it quite a bit 4 more this morning. 5 But there are all kinds of 6 applications out there for Windows. And the 7 Windows applications run on OS/2 2.0, then the 8 applications barrier will not be effective to 9 protect Microsoft's monopoly. And so what 10 they're searching for is a way to make that not 11 happen. 12 And this technology, this DPMI, 13 Microsoft has just recently pushed the industry 14 to adopt this DPMI. And it is important, 15 critical to IBM's ability to run the Windows -- 16 to run Windows and Windows applications on 17 OS/2. 18 What this would allow is for Windows 19 users to make the jump to OS/2 without 20 abandoning all of their old applications. 21 One of Microsoft's tactics to prevent 22 this from happening is to simply stonewall the 23 delivery of the Windows source code to IBM 24 despite a contractual obligation to do so. 25 Ballmer, Mr. Ballmer, knows how 3138 1 important this DPMI is to IBM. So there's a 2 brief e-mail exchange over the issue, and 3 Mr. Ballmer himself decrees that Microsoft 4 should hold back that code, the DPMI code. The 5 exchange starts on March 1st, 1991, in an 6 e-mail from Sherrie V. to Mr. Ballmer. 7 This is Plaintiffs' Exhibit 3447. 8 March 1st, 1991, Windows source code 9 to IBM. And this Sherrie V. says to 10 Mr. Ballmer, what they really want is 3.1 11 source to evaluate the DPMI client code. We 12 didn't send it to them. 13 Brad S-I, Silverberg, said you wanted 14 to hold off and mentioned the concern about IBM 15 incorporating 3.1 as WABBC, which means binary 16 compatibility, onto OS/2 2.0 and not paying any 17 royalty on it. 18 I see three options. 19 Number 1, send it to them. They know 20 the DPMI functionality is in there, and have a 21 license to preliminary source code no less 22 frequently than monthly, so are making a 23 legitimate request. 24 Two, don't send it to them. I tried 25 to buy some time with DeAnn Safford, at IBM by 3139 1 telling her the code was still prebeta and not 2 in a suitable state to send out, but she, IBM's 3 representative, refused to wait and insisted on 4 immediate delivery. 5 Number 3, do something slimey like 6 send it to them, but break key parts of the 7 code. 8 On March 3rd, a couple of days later, 9 Mr. Ballmer tells Sherrie V. to withhold the 10 code. This is Plaintiffs' Exhibit 3448. 11 And this is what Mr. Ballmer says. 12 Take out the DPMI functionality and send them 13 3.1 sources in a month. We only owe them 14 source once a month. 15 Microsoft originally announces that 16 Windows 3.1 will be what is called a DPMI 17 client. They incorporate this DPMI, obviously, 18 because they believe this will help Windows 3.1 19 in the market. 20 David Cole writes Mr. Silverberg back 21 on January 30th, 1990. This is before the 22 change in plans with respect to OS/2. This is 23 back more than a year. 24 And this is Plaintiffs' Exhibit 9508. 25 Windows everywhere, right? Making 3140 1 Windows a DPMI client will make this easier. 2 But then after Microsoft's decision 3 concerning DPMI and Windows 3.1 changes as 4 their strategy with respect to OS/2 changes. 5 And on April 3rd, 1991, Mr. Silverberg 6 writes in an internal e-mail, which is 7 Plaintiffs' Exhibit 9449. I see in -- who's he 8 writing to here? David, I think that's David 9 Wozniak. 10 I see in PC Week this week you mention 11 that Win 3.1 will be a DPMI client. Not so. 12 At least the retail version of 3.1 will not be 13 a DPMI client. This is a change as of a few 14 weeks ago. 15 Why make this fundamental technical 16 change? The value to Microsoft of its change 17 in position is explained by reading these 18 internal e-mails from Mr. Silverberg in August 19 of 1991. And this is Plaintiffs' Exhibit 4180. 20 Which OEMs are asking for DPMI in Win 21 3.1 and why? It really only makes sense when 22 you're trying to host Windows on non-DOS 23 systems. It's not something we're eager to do. 24 What he's asking there is -- or what 25 he's saying there is you really only need for 3141 1 Win 3.1 to be a DPMI client if you want to run 2 it on something other than MS-DOS. So who's 3 asking for it. 4 And, eventually, Windows 3.1 is not a 5 DPMI client. 6 Microsoft realizes that DPMI might 7 allow consumers to move easily from Windows to 8 OS/2. So August 22nd, 1991, Plaintiffs' 9 Exhibit 9507 says, we are willing to produce 10 DPMI-client versions of Windows for OEMs who 11 license Windows and make corresponding 12 commitments to Windows. We will not be 13 providing the DPMI-client version as the retail 14 product. 15 From this and other documents, you 16 will see that Microsoft is prepared to limit 17 distribution of the DPMI client version of 18 Windows 3.1 to those OEMs who can be counted on 19 not to use it with OS/2. 20 Other Microsoft customers who might 21 want to use it on OS/2 receive Windows 3.0 22 without the DPMI, which makes it possible or 23 makes it easier to run the Windows and Windows 24 applications. 25 Now, what happens, of course, is IBM 3142 1 has to change -- has to change horses in the 2 middle of the stream. And they do that, but it 3 substantially delays the release of OS/2 2.0. 4 One of our technical and industry 5 experts, Mr. Ron Alepin, will talk to you 6 further about the background and the 7 significance of this DPMI issue and will, as I 8 said, explain to you what that means. 9 On April 29th, 1991, Steve Ballmer, 10 again, receives some ideas about how to create 11 incompatibilities in OS/2. This time from a 12 guy named Stu K. Stu's proposal is to check 13 whether a Windows application is being run on 14 OS/2. If it is, he suggests they display a 15 warning about the application not being 16 certified. 17 He writes in Plaintiffs' Exhibit 695, 18 to discount the better Windows than Windows 19 claim. Put in OS/2 2.0 and WLO-detection into 20 Win 3.1. If an app is -- I used to know how to 21 pronounce -- how they pronounce that, but I 22 just can't remember. So we'll just say WLO, 23 then run it. 24 If OS/2 2.0 exists and the app is not 25 a WLO app, then put up an error message 3143 1 indicating that the app is not certified by MS 2 and that it can be run at one's own risk. Hard 3 for IBM to make a better Windows claim with 4 such a scheme in place. 5 Once we detect the existence of OS/2 6 2.0, there may be other special warnings we can 7 emit that highlight legitimate problems with 8 Windows apps running on top of OS/2 2.0. 9 For example, a message each time one 10 prints, that says the Windows and OS/2 printer 11 models are different and that output will be 12 different between the two systems. 13 What Stu K. is suggesting here is that 14 Windows display warnings to users of Windows 15 applications when they try to run them on OS/2. 16 Like the verified DOS that Mr. Gates ordered 17 back in 1988 for DRI. 18 And like the one -- like the AARD code 19 that went into the Christmas beta in December 20 of 1991, a few months after this exchange. 21 Now, we do not know whether or not 22 this suggestion was implemented. In his 23 examination of Microsoft's source code for 24 MS-DOS 6.0, Andrew Schulman found code in the 25 combined setup for Windows and MS-DOS that 3144 1 tests for OS/2 and displays a warning. 2 On October 30th, 1991, Bill Gates 3 himself weighs in. He knows what to do and he 4 wants to do it. 5 This is Plaintiffs' Exhibit 1059. 6 Mr. Gates says, we should design some 7 of our extensions explicitly so that IBM can't 8 run them under OS/2. We need to put some real 9 thinking into this. 10 Just a month later, there is a formal 11 OS/2 attack plan -- it is 20 pages long -- by 12 Joseph Crawczak, the product manager, which 13 includes the following ideas of OS/2, attack 14 plan, draft only. Plaintiffs' Exhibit 1116. 15 He says, it is therefore critical that 16 we prevent OS/2 2.0, from eroding Windows 17 momentum and defeat IBM's attempt at gaining a 18 foothold on the desktop with OS/2. 19 To do this we must: Pursue a product 20 development strategy that prevents IBM from 21 claiming Windows compatibility. Prevent 22 Windows applications from running correctly on 23 OS/2. 24 Next, set a new bar for Windows 25 functionality and requirements for 3145 1 compatibility. 2 And reposition OS/2 as impractical and 3 incompatible in the minds of customers. 4 We have seen these tactics before. 5 The technical sabotage, FUD, deception, 6 Microsoft makes its products incompatible on 7 purpose and then uses the incompatibilities 8 that it has created to claim that OS/2 is a bad 9 product. 10 In early February -- and we're a 11 couple of years beyond -- I guess, a year 12 beyond. In early February 1993, Mr. Gates 13 writes to Silverberg and Maritz about how good 14 OS/2 still is. OS/2 2.0 has come out now, and 15 how Microsoft will continue to have trouble 16 with its message as long as OS/2 is perceived 17 as a high-end product in the marketplace. 18 He writes in Plaintiffs' Exhibit 1554, 19 the subject of which is OS/2 2.1. I am really 20 concerned over everything I am hearing about 21 OS/2 2.1. 22 As usual, they get total credit for 23 being crash-proof. Our inability to explain 24 they are not crash-proof is going to kill us. 25 I don't understand what they did 3146 1 technically, but I am very, all caps, 2 impressed. I thought the mapping they had to 3 do would prevent them from ever getting this 4 good. 5 Gates himself drives the campaign 6 against OS/2 and puts executives on -- puts 7 pressure on executives to keep on it. 8 In a June 27th, 1991, e-mail to 9 Silverberg and Ballmer, Gates asks them about 10 some features that could be added to Windows to 11 make it more robust and discusses the 12 effectiveness of discrediting OS/2. 13 He writes in Plaintiffs' Exhibit 808, 14 IBM is putting our lack of protection between 15 applications in their ads, and there is no good 16 response. We can try to say that OS/2 isn't 17 robust either, but that is a hard point to 18 make. 19 And it is because, as you have seen 20 from the documents, Microsoft knows that OS/2 21 is robust. 22 Back to 808 -- and this is a response 23 from Brad Silverberg to what Mr. Gates says. 24 Mr. Silverberg tells him that adding 25 protection to Windows would be hard from a 3147 1 development perspective, but he has another 2 thought. He suggests that attacking OS/2 3 robustness might not be as hard as Gates 4 thinks. 5 This, again, is 808. Yes, this would 6 be a fairly disruptive thing to do. On the 7 other side, though, we do have some apps 8 developed by Eric Fo -- that is Eric Fogland, 9 which will crash OS/2 2.0. We will do more of 10 these. 11 What the app that Mr. Silverberg 12 refers to is -- referred to by the executives 13 as the bad app. It is not a real application. 14 It is developed by Earl Fo specifically to 15 allow Microsoft to spread FUD about the 16 robustness of OS/2. 17 Just a few days before Silverberg's 18 e-mail to Gates, Eric Fo, himself, 19 Mr. Fogland, briefly describes what the bad app 20 does. And this is Plaintiffs' Exhibit 797. 21 Do you see the subject matter -- this 22 is to Mr. Maritz, called one bad app. 23 This is June 26, 1991. 24 I have written a PM, presentation 25 manager, app that hangs the system. Sometimes, 3148 1 quite graphically. You can look at it any 2 time, just let me know. 3 The reason Microsoft creates or has 4 Mr. Fogland create this bad app isn't to 5 distribute it to users, isn't to distribute it 6 to customers. It doesn't do anything useful 7 for customers. This bad app is entirely for 8 FUD purposes. Fear, uncertainty, and doubt. 9 And with this application in hand, 10 Microsoft sends Mr. Ballmer out to do a tour 11 demonstrating that OS/2 is unstable by 12 showing -- putting in Mr. Fogland's bad app and 13 showing how it crashes the system. 14 While on July 24th, when Mr. Ballmer 15 is on the tour, Mr. Gates reads a Seybold 16 report on office computing that praises OS/2. 17 He, Mr. Gates, feels apparently they are not 18 succeeding in their FUD efforts and criticizes 19 his executives and the PR folks in Plaintiffs' 20 Exhibit 860, which is an e-mail re: Seybold 21 report on office computing. 22 This report highlights our failure to 23 get our message out. It praises OS/2 2.0 24 endlessly, using the charts from the IBM white 25 papers. It praises the better Windows than 3149 1 Windows capability. 2 Turning the page. I am sure I will 3 get back some message about how Steve will see 4 these people and it will all be better. It 5 won't be better. No one is taking 6 responsibility for getting our message out 7 broadly. We should recognize we are losing 8 this battle and we need a lot more creativity 9 to get on top of it. 10 And, of course, he does get an e-mail 11 back about Mr. Ballmer's tour and the efforts 12 to crash OS/2. 13 On July 25th, 1991, Claire Lematta, 14 you've heard her before, from Weggener Edstrom, 15 responds to Mr. Gates, Silverberg, Allchin, 16 Maritz, Ballmer, with the details of the tour. 17 And she writes in Plaintiffs' Exhibit 18 860, the title of which is Steve B. press tour 19 trip report, long mail. 20 Steve B. went on the road to see the 21 top weekly's industry analysts, and business 22 press. The meeting included demos of Windows 23 and OS/2, including a performance comparison to 24 Windows and a bad app that corrupted other 25 applications and crashed the system. 3150 1 It was a very valuable trip and needs 2 to be repeated by other Microsoft executives 3 throughout the next month so we hit all 4 publications and analysts. 5 The demos of OS/2 were excellent. 6 Crashing the system had the intended effect, to 7 FUD OS/2 2.0. 8 The bad app, one purpose only, to FUD 9 OS/2 2.0. 10 So what has happened here is 11 Mr. Ballmer goes around the country with the 12 bad app and shows that OS/2 can be made to 13 crash. 14 Now, he does not reveal that the app 15 that crashes OS/2 was especially written to do 16 one thing and one thing only, and that is to 17 crash OS/2. 18 No consumer will ever use it. No 19 consumer will ever see it. It won't be 20 crashing their computers. It is Microsoft FUD. 21 On July 28th, 1991, Mr. Muglia writes 22 Mr. Maritz from the road about how popular the 23 bad app is with Microsoft salespeople. They 24 just can't hardly wait to get their hands on 25 it. 3151 1 This is Plaintiffs' Exhibit 7603. PM 2 bad app is a great thing. When I mentioned 3 this app to the sales force, they practically 4 begged me for a copy. 5 However, I agree with Eric Fo that we 6 absolutely should not give out this app. The 7 potential backlash of Microsoft spreading OS/2 8 viruses is not worth it. A really good video 9 like this will blow apart whatever momentum IBM 10 has achieved with 2.0. 11 This goes on, this bad app -- and 12 eventually it becomes known as the 13 Terminator -- I'm sure named after Arnold 14 Schwarzenegger's movies -- for its ability to 15 destroy OS/2. 16 Rob Horowitz recounts a Ballmer 17 demonstration of the Terminator in a July 30th, 18 1991, e-mail summary. This occurs at a 19 financial analysts meeting. 20 This is Plaintiffs' Exhibit 5330. 21 Robustness? Not a lot more robust than 22 Windows. Demoed an OS/2 app (Terminator) 23 running under OS/2 2.0, which corrupts other 24 apps and eventually brings down the system. 25 I do want you to note there that this 3152 1 says that it brings down OS/2 2.0. I'm going 2 to show you a little bit more about that. 3 Despite the success of Ballmer to the 4 bad app and other FUD efforts, Silverberg is 5 still pushing hard on doing more against -- 6 more anti-OS/2 propaganda. 7 On December 25th, Christmas day, 1991, 8 he writes to Ballmer, Cole, Maritz, Pam 9 Edstrom, of the need to blitz OS/2 in order to 10 avoid a repeat of the DR-DOS situation. And 11 you will see that the title of the e-mail is 12 blitzing OS/2 2.0. 13 This is Plaintiffs' Exhibit 1153. 14 As OS/2 is starting to come near -- 15 and he means OS/2 2.0 -- we need to build our 16 artillery against it and take the battle to 17 them. I don't want a repeat of the DR-DOS 18 situation where the reviews came out first and 19 then we were forced to react. 20 DR-DOS got good reviews for 21 compatibility, momentum, and then we're placed 22 on the defensive having to demonstrate and get 23 the word out on its problems. 24 We can't let this happen with OS/2. 25 We need to know the problems, especially with 3153 1 Windows compatibility and performance and let 2 the reviewers know before they do their 3 reviews. 4 Let's blitz them before they get 5 started. 6 By February of 1992, the bad app has 7 become known as TIII, Terminator 3, and 8 Silverberg is still keen on how effective it is 9 in the fight against OS/2. 10 He believes that it is effective FUD 11 and wants a list of what it does, and he writes 12 to a guy named Doug Hogarth in Plaintiffs' 13 Exhibit 1236. 14 Re: OS/2 protection myth. 15 And the first thing he asks is what 16 are the things that TIII exploits? Later in 17 the day on February 24th, Mr. Hogarth responds 18 to Silverberg. But Mr. Hogarth is 19 uncomfortable with the bad app. 20 And what it does, especially since it 21 has actually been developed for an earlier 22 version of OS/2. Not OS/2 2.0, but OS/2 1.3, 23 the one that comes out before OS/2. 24 And Ballmer has implied on his tour 25 that it was being used on the new version of 3154 1 OS/2 2.0. 2 And in 1236 Mr. Hogarth says due to 3 the sensitive nature of the bad app, I'd prefer 4 to demonstrate/discuss only in person. 5 Ballmer went on a press tour with this 6 thing and lots of the press perceived it as an 7 I can crash the OS/2 2 beta when it really is a 8 version of a 1.3 style app. 9 I can't stress enough how sensitive 10 the code is considered. It is -- I think that 11 meant -- it is meant only to point out what you 12 said, that OS/2 is not bullet proof. 13 On February 27th, there's another 14 meeting of Microsoft executives and the 15 Weggener Edstrom PR folks in which they discuss 16 an escalation of the FUD campaign. Now they're 17 calling it crush OS/2. 18 Claire Lematta writes the following to 19 Silverberg, Maritz, and others. And this is 20 Plaintiffs' Exhibit 1254. 21 This is what we agreed to. Next week 22 the OS/2 swat team are going to aggressively 23 put together the crush OS/2 data. We need to 24 demonstrate that it is not better Windows than 25 Windows. 3155 1 At the end of the week, PR brains are 2 put -- PR brains are put to determining what is 3 the best way to present, leverage, communicate 4 this information. 5 Mr. Marty Towcher is an in-house PR 6 person at Microsoft, and he's not too happy 7 with the plan. He feels that it makes them 8 look bad, and he writes to Mr. Silverberg, 9 Maritz, and others. Puts in his two cents. 10 My two cents is that we look bad and 11 uncredible when we demonstrate OS/2 problems. 12 We can talk about them and hand evaluation 13 guides, but actually showing the product makes 14 us look bad. 15 But Silverberg continues his campaign 16 against OS/2 for the rest of 1992. 17 On August 12th, 1992, he writes an 18 e-mail to Gates in which he reiterates the need 19 to keep up the pressure on OS/2 by hammering -- 20 by, quote, hammering a few more nails in its 21 coffin. 22 This is Plaintiffs' Exhibit 1393. 23 Update. We keep trying to put nails in the 24 OS/2 coffin. They keep pushing the lid up. We 25 should hammer a few more nails in now and not 3156 1 let them get momentum. 2 And hammer in a few more nails is just 3 what they do. There's a relentless, continuous 4 FUD campaign against IBM. 5 Would now be a good time, Your Honor? 6 THE COURT: Sure. 7 Ladies and gentlemen, we will take a 8 break right now. 15 minutes. Remember the 9 admonition previously given. 10 Please leave your notebooks here and 11 we'll see you in about 15 minutes. 12 (A recess was taken from 10 a.m. to 13 10:18 a.m.) 14 THE CLERK: All rise. 15 THE COURT: Everyone else may be 16 seated. 17 My court attendant told me some of you 18 wanted some legal pads. That's fine. There is 19 a thing in the back you can get them in. I'll 20 give you a minute to get organized. Do you 21 want to go ahead and put those on your thing? 22 And other problems with the notebooks? 23 We'll wait. 24 [Juror Name]: Oops. 25 MS. CONLIN: Those little things 3157 1 always stick to my fingers. 2 THE COURT: Got it? Hope those work. 3 [Juror Name]: Yep. 4 THE COURT: All right. You may 5 continue, ma'am. 6 MS. CONLIN: Thank you, Your Honor. 7 We were talking about OS/2, and I'm 8 going to tell you -- I've told you about the 9 bad app and some of the other things, 10 intentional incompatibilities, but there was 11 more. 12 Microsoft uses access to Windows 95 13 betas, Chicago betas, and programming 14 information as a lever against OS/2, as it used 15 the beta of MS-DOS, betas of MS-DOS and Win 3.0 16 against DR-DOS. 17 ISVs and OEMs are very keen to get 18 their hands on this Chicago beta. And they 19 need early betas because applications, ISVs, 20 applications vendors, and others, need them to 21 test the system and make sure their own 22 applications are going to work with this new 23 Chicago -- with the Chicago system and that 24 their applications can come out at about the 25 same time that the Windows 95 hits the market. 3158 1 On May 13th, 1993, Barry Spector 2 writes Joachim Kempin about an announced deal 3 between IBM, AST, OS/2. This is Thursday, May 4 13th, 1993. 5 Mr. Spector says -- this is 6 Plaintiffs' Exhibit 1661 -- as we suspected, 7 AST will stand on stage with IBM next week, 8 announce that they will sign a license with IBM 9 for OS/2. I tried to convince them not to 10 stand on stage with them. 11 Kempin's response is to withhold 12 cooperation and code from AST and IBM. He 13 writes -- this is also 1661 -- very clear to 14 me. No Chicago. No cooperation. No beta. No 15 alpha code. Total war. 16 Yet another way in which Microsoft 17 uses Chicago as a lever against IBM is to 18 withhold a license to IBM -- you remember the 19 two parts of IBM, the hardware part that must 20 have Windows to survive and the software part 21 which develops OS/2 and other software that 22 competes with Microsoft. 23 The software department creates and 24 sells OS/2. The hardware part needs Windows 95 25 to sell its PCs. 3159 1 OS/2 has a very small market share. 2 And despite the fact that it's IBM's own 3 operating system, IBM still needs to be able to 4 license Windows or it won't be able to sell its 5 machines. And IBM knows this. 6 And on August 14th, 1994, Kempin 7 proposes that Microsoft invoke its ability 8 under its contracts to audit IBM. 9 Until this audit is over, they do, in 10 fact, withhold a license to Windows 95 from 11 IBM. 12 This is back August 14th, 1994. This 13 is Mr. Kempin, and he is -- this is Plaintiffs' 14 Exhibit 9624, confidential MS OS/2 royalty 15 numbers, quick analysis. 16 Mr. Kempin says let's do it, audit 17 them. Reason to say, when auditing them, 18 according to their bragging in the press, we do 19 not believe they are reporting correct OS/2 20 numbers and we do not believe their last Win, 21 Windows report. 22 As Judge Rosenberg told you, and as 23 we'll discuss in a minute, as he told you when 24 reading the findings from the government case 25 that are already proven, this audit is a part 3160 1 of