11750 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 XLIII 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 8:30 a.m., February 2, 14 2007, 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 11751 1 A P P E A R A N C E S 2 Plaintiffs by: BRADLEY M. BEAMAN 3 Attorney at Law Roxanne Conlin & Associates, PC 4 Suite 600 319 Seventh Street 5 Des Moines, IA 50309 (515) 283-1111 6 ROBERT J. GRALEWSKI, JR. 7 Attorney at Law Gergosian & Gralewski 8 550 West C Street Suite 1600 9 San Diego, CA 92101 (619) 230-0104 10 KENT WILLIAMS 11 Attorney at Law Williams Law Firm 12 1632 Homestead Trail Long Lake, MN 55356 13 (612) 940-4452 14 15 16 17 18 19 20 21 22 23 24 25 11752 1 Defendant by: DAVID B. TULCHIN 2 SHARON L. NELLES Attorneys at Law 3 Sullivan & Cromwell, LLP 125 Broad Street 4 New York, NY 10004-2498 (212) 558-3749 5 STEPHEN A. TUGGY 6 Attorney at Law Heller Ehrman, LLP 7 333 South Hope Street Suite 3900 8 Los Angeles, CA 90071-3043 (213) 689-0200 9 DAVID E. JONES 10 Attorney at Law Heller Ehrman, LLP 11 One East Main Street Suite 201 12 Madison, WI 53703-5118 (608) 663-7460 13 BRENT B. GREEN 14 Attorney at Law Duncan, Green, Brown & 15 Langeness, PC Suite 380 16 400 Locust Street Des Moines, IA 50309 17 (515) 288-6440 18 19 20 21 22 23 24 25 11753 1 STEVEN J. AESCHBACHER Attorney at Law 2 Microsoft Corporation One Microsoft Way 3 Redmond, WA 98052 (425) 882-8080 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 11754 1 (The following record was made in the 2 presence of the jury at 8:33 a.m.) 3 THE COURT: Everyone else may be 4 seated. 5 Before we begin, Plaintiffs' Exhibit 6 1369 was previously admitted I was told by 7 Mr. Green, so -- 8 MR. TULCHIN: That's correct, Your 9 Honor. 10 THE COURT: So that's admitted. Okay. 11 MR. GRALEWSKI: Thank you, Your Honor. 12 Good morning. We are going to 13 continue with the videotape deposition of 14 Mr. Barrett. 15 THE COURT: Thank you. 16 (Whereupon, the following video was 17 played to the jury.) 18 Question: You've been handed Exhibit 19 9 to your deposition, which is another string 20 of e-mails produced to us by Microsoft. 21 I'm going to direct your attention to 22 the e-mail that begins at the upper left-hand 23 corner of the first page. 24 Do you recognize this as an e-mail 25 from a David Cole, C-o-l, to Brad Chase, 11755 1 yourself, and others at Microsoft dated July 2 17, 1991? 3 Answer: Yes, I do. 4 Question: And who is David C-o-l? 5 Answer: David Cole, C-o-l-e, is his 6 last name, was the -- I don't recall the exact 7 title, group program manager on Windows 3.1, 8 sort of my counterpart. 9 Question: And what was the 10 responsibility of a group program manager? 11 Answer: Well, program management is 12 -- I think can be classified in some ways as 13 project management, where they -- they 14 basically do all the things that are necessary 15 to get the project started and completed, other 16 than develop the software, write the code and 17 that sort of thing. 18 Question: I see. So you worked with 19 him on the development side, or the two of you 20 worked together to develop a 3.1, Windows 3.1? 21 Answer: Yes. Together we were 22 responsible for delivering Windows 3.1. 23 Question: All right. Mr. Cole says 24 at the beginning of his e-mail, I think we 25 should use Windows to get a Microsoft operating 11756 1 system back on to the NetWare clients which 2 will bundle or require DR-DOS. 3 What was NetWare? 4 Answer: NetWare is -- NetWare is a 5 product from Novell that basically was a 6 network operating system, for lack of a better 7 term, that allowed PCs to talk to one another 8 over a network. 9 NetWare client -- well, you didn't ask 10 that question, sorry. 11 Question: Tell us what a NetWare 12 client is. 13 Answer: NetWare client is the 14 software that runs on a PC, which is basically 15 the software that does the connecting of the PC 16 and the PC operating system to the network. 17 Question: At this time, did Microsoft 18 have products that were competitive with 19 NetWare and NetWare clients? 20 Answer: Yes. 21 Question: And what were those 22 Microsoft products? 23 Answer: It was called LAN manager and 24 OS/2 LAN manager. 25 Question: Mr. Cole goes on to say, we 11757 1 should alter our plain a bit and move all the 2 DOS 6 improvements directly into Windows. 3 When the user starts Windows, they get 4 the Microsoft operating system (including 5 networking) and all the other cool features 6 that go with that. 7 When they quit, they get NetWare and 8 DR-DOS and no Windows applications. The key is 9 getting a piece of Microsoft system software on 10 the client so that we can deliver our strategy 11 and vision. We can leverage Windows and 12 Windows applications to do this. 13 Can you explain to me in lay terms 14 what the strategy was that Mr. Cole was 15 proposing here? 16 Answer: Well, the vision was 17 basically Microsoft software doing all the 18 functions. And strategy is to use various 19 pieces of Microsoft software to do that. 20 Question: How did this work? How 21 could Windows and Windows applications be 22 leveraged to make everything run on Microsoft 23 software? 24 Answer: Well, basically -- it's 25 actually outlined in this e-mail fairly 11758 1 succinctly. If you use -- Microsoft DOS 2 requires Windows to have, required to be 3 running for Windows to run, then basically 4 you've cut DR-DOS out of the loop and made it 5 essentially a useless appendage. 6 An additional cost item, if you've 7 standardized on Windows applications, which was 8 happening at that time, that that would 9 basically make DR-DOS an ordered expense, an 10 unnecessary added expense. 11 Question: And is there some way that 12 moving all of the DOS 6 improvements directly 13 on to Windows accomplished that goal? 14 Answer: Yes. 15 Question: How is that? 16 Answer: Yes. That's exactly -- 17 that's exactly to the point by essentially 18 putting DOS 6 into Windows that caused -- that 19 would cause that short-circuit to happen, that 20 DR-DOS would no longer be germane. 21 Question: Now, when Mr. Cole talks 22 about putting DOS 6 on to Windows, is that a 23 form of merging, technically merging Windows 24 and DR-DOS? 25 Answer: Yeah, that's exactly -- 11759 1 exactly it. 2 Question: Mr. Cole ends his e-mail 3 message by saying, if we are going to take over 4 the desktop when Windows starts, it must be all 5 Microsoft written software since Novell won't 6 help us do that. 7 Do you understand what was meant by 8 that sentence? 9 Answer: Well, yeah. Again, it goes 10 back to what I said earlier. The whole point 11 is to have only Microsoft software running and 12 taking over the desktop means precisely that. 13 Question: Mr. Barrett, you've been 14 handed Exhibit 10 to your deposition which is 15 another conglomeration of e-mail messages that 16 was produced to us by Microsoft. 17 I apologize for the small print, but 18 this is how it was produced to us. 19 Are you able to read this? 20 Answer: With some effort. 21 Question: There are a number of 22 e-mail messages here, but I'm actually 23 interested in the e-mail message that begins at 24 the bottom of the second page of the exhibit 25 that's Bates stamped MS-PCA 1178453. 11760 1 Do you recognize this as an e-mail 2 message from you to Brad Silverberg dated July 3 16th, 1996, subject Novell? 4 Answer: Yes, I do recognize this is 5 my e-mail. 6 Question: Was this e-mail generated 7 in response to the e-mail that we were just 8 looking at from Mr. Cole, Exhibit 9? 9 Answer: Yes, I -- let me look at that 10 just briefly. 11 Yes, I believe it was. 12 Question: You started out your e-mail 13 by saying, I think we need to carefully measure 14 our response. Totally freezing them out will 15 force them to compete on a wider basis and 16 could cause the disaster scenario to occur. 17 Do you know what disaster scenario you 18 were referring to? 19 Answer: Yeah. Disaster scenario is 20 just basically forcing, causing our competitors 21 to get together and up the playing -- basically 22 make the stakes much higher by when they get 23 together, they could become more significant 24 force. 25 Question: You go on to say, not to 11761 1 mention the thin FTC ice that we would be on. 2 FTC refers to Federal Trade 3 Commission? 4 Answer: Yes, it does. 5 Question: And what -- what did you 6 mean by thin FTC ice that you would be on? 7 Answer: Well, I'm not a lawyer, so 8 it's been pointed out to me, had been pointed 9 out to me that I really didn't have a basis to 10 make a judgment on. 11 Although there was a lot of -- there 12 were a lot of FTC issues going on there. 13 Question: During this time period? 14 Answer: During that time period. So 15 I was sensitive to it. 16 Question: All right. And I know 17 you're not a lawyer, but I'm just simply 18 asking, what was it that you were referring to 19 here in your lay opinion? 20 Answer: It was -- it was my belief -- 21 my belief that we had won. I mean, we had 22 basically sold at that point, I don't know, 23 four to six million copies of Microsoft Windows 24 3.0 and the corporations were flocking to us in 25 droves and we had basically won. 11762 1 And basically freezing out our 2 competitors might have been a bad thing to do. 3 But, again, not being a lawyer, I really -- you 4 know, probably was not -- should not have even 5 made that comment. 6 Question: Then you go on to say, on 7 the other hand, we have an advantage with 8 Windows and should press it in the systems 9 area. 10 What did you mean by that? 11 Answer: Well, because Windows was 12 becoming very, very successful at that point. 13 It was -- basically that software would wind up 14 on every corporate desktop. 15 And so not unlike what David Cole was 16 saying, that's our entry. That's our, if you 17 will, our springboard to do things that would 18 allow us to dominate the market. 19 Question: All right. 20 Answer: In case -- sorry. In this 21 specific case, it was about going to war with 22 Novell. 23 Question: All right. You go on to 24 say, one approach would be to deny enhanced 25 mode to all but real mode DOS customers. 11763 1 What is enhanced mode and what is real 2 mode? 3 Answer: Enhanced mode refers to 4 protected mode. There are a number of terms 5 for it. But basically it's the mode that 6 Windows would run in that would give the most 7 benefit by creating a large amount of memory. 8 Windows applications were very memory 9 hungry, and real mode limited them to what's 10 called the 640 K box, which was a very, very 11 constrained box at the time. 12 Small amount of memory forced them to 13 do some complicated and slow things in order to 14 fit all these features into memory. And 15 because of that, any application that ran in, 16 that didn't run in enhanced mode was at a 17 significant advantage. 18 Question: And enhanced mode or 19 protected mode Windows came out with which 20 release of Windows? 21 Answer: Windows 3.0. It was the 22 biggest feature of Windows 3.0. The thing that 23 caused Windows 3.0 to be successful. 24 Question: Now, what were you 25 referring to here when you said, one approach 11764 1 would be to deny enhanced mode to all but real 2 mode DOS customers? 3 Answer: Well, looking at it, I think 4 I probably didn't write this very clearly. 5 But what I was trying to say was that 6 MS-DOS, anybody using MS-DOS as the operating 7 system that Windows was running on would be our 8 -- we would only allow MS-DOS -- sorry. I'm 9 trying to say this clearly and I'm not doing a 10 good job. 11 Question: Take your time. 12 Answer: By making essentially MS-DOS 13 a requirement underlying Windows running in 14 enhanced mode would basically force anybody who 15 wanted to use enhanced mode Windows, which 16 basically is everybody, they would have to use 17 MS-DOS. 18 Question: As opposed to any competing 19 DOS? 20 Answer: That is absolutely correct. 21 As opposed to any other DOS from some other 22 company. 23 Question: Do you know if that 24 proposal was ever implemented? 25 Answer: I don't believe it was. 11765 1 Question: You go on to say, this can 2 be accomplished in a number of ways, but the 3 single most effective one is to move kernel (or 4 major components of it) into a VxD and thus 5 significantly up the cost of entry to running 6 full Windows. 7 What was it that you were proposing 8 there? 9 Answer: Okay, this is a very complex 10 and highly technical. Put it in a nutshell, 11 the kernel is the innermost piece. It's the, 12 if you will, the heart and soul of Windows, a 13 small piece of code. 14 Putting it into a VxD is a fairly 15 technical term for putting it into a 16 specialized -- making it be a specialized piece 17 of the Windows system. 18 Question: So the kernel you were 19 referring to here was the DOS kernel, the 20 MS-DOS kernel? 21 Answer: The Windows kernel. 22 Question: The Windows kernel, all 23 right. 24 Answer: The Windows kernel. 25 By doing that, there's other -- some 11766 1 other advantages, some technical advantages to 2 doing that, but the whole idea was to put it 3 into what's called a virtual device driver, 4 VxD, and thus making it difficult for people to 5 get at the kernel who do things with it. 6 Question: And that's what would 7 prevent other DOS competitors from making use 8 of Windows? 9 Answer: Yeah, it would prevent DOS 10 competitors from running Windows. Although, it 11 would require them, as I say here, it would 12 require them to do a bunch of work. 13 Wouldn't necessarily prevent them, but 14 it would cause them to have to do a bunch of 15 work and thus delaying their product plans. 16 Question: And is that work, that 17 extra work what you're referring to, is that 18 what was meant by significantly up the cost of 19 entry to running full Windows? 20 Answer: Yes, that's exactly what I 21 meant by that statement. 22 Question: Do you know whether 23 Microsoft ever implemented this idea of moving 24 the kernel or major components of it into a 25 VxD? 11767 1 Answer: I believe it did not happen. 2 Question: And VxD is virtual device 3 driver? 4 Answer: Yes, that's correct. 5 Question: Mr. Barrett, you've been 6 handed Exhibit 11, which is another 7 conglomeration of Microsoft e-mails which were 8 produced to us by Microsoft. 9 I'd like to direct your attention to 10 an e-mail that begins at the very bottom of the 11 first page from you to a L-a-w-r-e-n. 12 Do you know who that is? 13 Answer: I don't recall. 14 Question: Dated September 19th, 1991. 15 Subject re: Bambi and DR-DOS 6.0, many files 16 turned into crosslinked mess. 17 And then embedded within that e-mail 18 do you see a message from LAWREN to an alias 19 called Winbug dated the same date, September 20 19, 1991, on the beginning of the second page? 21 Answer: I'm sorry, the question? 22 Question: Do you see embedded within 23 your e-mail is an e-mail message from LAWREN to 24 an alias called Winbug on that same date? 25 Answer: Yes, I do. 11768 1 Question: Okay. Let's start with 2 that embedded e-mail that you were replying to 3 in your e-mail. 4 It begins, Windows bug report and then 5 it goes on to say, title, colon Bambi and 6 DR-DOS 6.0, many files turned into cross-linked 7 mess. 8 Do you know what that refers to? 9 Answer: Yes. 10 Question: Okay. Can you tell us what 11 that was? 12 Answer: Okay. Bambi I think was code 13 name for MS-DOS 6.0. And DR-DOS 6.0 is 14 obviously the Digital Research DOS. 15 Many files turned crosslink means -- 16 simply means that there was some problem that 17 occurred, that damaged the file system and 18 basically destroyed data. 19 Question: All right. And then do you 20 recall what the situation was where was this 21 crosslink mess? What caused that crosslink 22 mess? 23 Answer: No, I don't recall the 24 specifics of it. 25 Question: And then the e-mail at the 11769 1 bottom, the beginning of the bottom of the 2 first page was your response to the -- to 3 LAWREN's e-mail about the crosslink mess? 4 Answer: Yeah. Yes, it is my 5 response. 6 Question: Exhibit 12 has just been 7 handed to you, Mr. Barrett. And, again, this 8 is an e-mail, a series of e-mails produced to 9 us by Microsoft. If you would just take a 10 moment to look them over. 11 Answer: Okay. 12 Question: Now, there are some 13 references to Bambi in these e-mails also. 14 Was Bambi a beta version of a disk 15 cache driver that was being developed at 16 Microsoft? 17 Answer: Thank you for reminding me. 18 It, in fact, was that. It was a 19 technology we were developing to significantly 20 speed up performance of both Windows and DOS. 21 Question: Did that ultimately get 22 released, Bambi? Was it released as 23 Microsoft's SmartDrive disk cache driver? 24 Answer: Yes, that is exactly correct. 25 It was the early version of SmartDrive. 11770 1 Question: Were you involved in the 2 development of this disk cache driver that was 3 code named Bambi? 4 Answer: I was actually intimately 5 involved in it, yes. 6 Question: All right. Let me direct 7 your attention to the e-mail that begins at the 8 top of the page. 9 Do you recognize that as an e-mail 10 message from you dated September 28th, 1991 to 11 a Chuckst, Mikedr and Scottq? 12 Answer: Yes, I do recognize them. 13 Question: Who are those gentlemen? 14 Answer: Chuck Straub, I think, Mike 15 Dryfoos, and Scott Quinn. 16 And Chuck and Scott were the Bambi 17 developers, and Mike Dryfoos was the tech lead, 18 technical lead. 19 Question: And then your e-mail 20 message or response to an e-mail from Chuck 21 Straub to yourself and others at Microsoft 22 dated September 29, 1991? 23 Answer: Yeah, it was. 24 Question: All right. I'd like to ask 25 you about -- let's start with Mr. Straub's 11771 1 e-mail that you were responding to. 2 It's entitled -- the subject is Bambi 3 on DR-DOS 6.0. 4 Mr. Straub says, I tracked down a 5 serious incompatibility with DR-DOS 6. They 6 don't use the, quote, normal, close quote, 7 device driver interface for 32 M partitions. 8 What was he referring to there, if you 9 know? 10 Answer: Yeah. First of all, greater 11 than 32 M partitions means greater than 32 12 megabyte partitions. 13 There was a hard limit problem inside 14 of DOS, the file structures that DOS defined on 15 limiting sizes over 32 megabytes of disk space. 16 It seems like it's a trivial amount of 17 space now, but back then disks were in that 18 range. So this was cutting edge stuff back 19 then. 20 And the incompatibility is there is 21 simply the data structures on the disk that 22 people would use to essentially be able to get 23 access to greater than 32 megabytes. 24 MS-DOS did it one way, DR-DOS did it 25 another way. And if you tried to put MS-DOS on 11772 1 a machine that had a disk formatted by DR-DOS, 2 it wouldn't work over 32 megabytes. 3 Question: Okay. Now, did that have 4 some impact on this Bambi disk cache driver? 5 Answer: Certainly, certainly, because 6 Bambi understood -- Bambi understood at a very 7 low level how the disks were organized, the 8 data structures on the disks. So Bambi needed 9 to know the specifics of that to be able to 10 operate correctly. 11 Question: And this Bambi diskette 12 cache driver, was that incorporated into a 13 Microsoft product? 14 Answer: Yes, it was incorporated in 15 several Microsoft products, including Windows 16 and DOS 6. 17 Question: The third paragraph from 18 the bottom of Mr. Straub's e-mail begins, I've 19 patched a version of Bambi to work with DR-DOS 20 6 and it seems to run Windows 3.1 without 21 difficulty. 22 This same problem may have caused 23 other problems with Windows 3.1 and the swap 24 file under DR-DOS 6. It is possible to make 25 Bambi work, assuming we can come up with a 11773 1 reasonably safe method for detecting DR-DOS 6. 2 What was Mr. Straub telling you here? 3 Answer: Basically, what he's saying 4 here is that if you can figure out you're 5 running on DR-DOS 6, you can basically modify 6 Bambi to work in that environment, or I should 7 say you could modify Bambi to when it detects 8 DR-DOS 6 to work correctly. 9 Question: And he had developed a 10 patched version of Bambi that would allow Bambi 11 to work on DR-DOS 6.0? 12 Answer: That's precisely what he's 13 saying in this e-mail. 14 Question: I was asking you on Exhibit 15 12 about an embedded message from Chuck Straub 16 to you and others at Microsoft. And then you 17 replied to that message where Mr. Straub said 18 it was possible to make Bambi work with a 19 patch. 20 Do you see that at the top of the 21 page? 22 Answer: Yes, I do. 23 Question: And you said there the 24 approach we will take is to detect DR 6 and 25 refuse to load. The error message should be 11774 1 something like invalid device driver interface, 2 close quote. What were you saying there? 3 Answer: I was basically saying that 4 when SmartDrive, a/k/a Bambi, detects DR-DOS 5 that it wouldn't work. 6 Question: Do you know who made the 7 decision to do that? 8 Answer: Yes, I made that decision. 9 Question: Okay. Handing you Exhibit 10 13 to your deposition, which is another series 11 of e-mails provided to us by Microsoft. 12 I want to direct your attention to the 13 message that begins second from the bottom of 14 this exhibit from David Cole to you and Brad 15 Silverberg dated September 30, 1991. Subject, 16 Bambi on DR-DOS 6.0. 17 Do you see that message? 18 Answer: Yes, I do. 19 Question: All right. Do you 20 recognize this as a message that Mr. Cole sent 21 to you? 22 Answer: Yes, I do. 23 Question: In the message, it's just 24 one line, it says, it should say unsupported 25 version of DOS. 11775 1 Was this Mr. Cole's response to your 2 message in Exhibit 12 that we were just 3 referring to? 4 Answer: Yes, I believe it was his 5 response to me, to my message. 6 Question: So he wanted the error 7 message to say instead of saying invalid device 8 driver interface, he wanted it to say 9 unsupported version of DOS; is that correct? 10 Answer: That's what it appears to say 11 here, yes. 12 Question: And Mr. Cole, what was his 13 -- I'm sorry, I think I asked you this before, 14 but can you clarify what -- his position at 15 Microsoft was what? 16 Answer: He was the group program 17 manager on Windows. 18 Question: And did you report to him? 19 Answer: No, we were peers. 20 Question: You worked together with 21 him? 22 Answer: Yes. 23 Question: Do you know whether 24 Mr. Silverberg approved of this decision to put 25 up a warning message and have Bambi not run on 11776 1 DR-DOS? 2 Answer: Yeah, he did. 3 Question: You've been handed Exhibit 4 14 to your deposition, Mr. Barrett. 5 This is another set of e-mails 6 provided to us by Microsoft. 7 I want to direct your attention to the 8 e-mail that begins at the bottom of the first 9 page of the exhibit Bates stamped 1179289. 10 Do you recognize that as an e-mail 11 that you sent to Brad Silverberg, copy to David 12 Cole on September 28th, 1991? 13 Answer: Yes, I do. 14 Question: All right, okay. You say 15 at the bottom of your e-mail message, my 16 proposal is to have Bambi refuse to run on this 17 alias -- actually on this alien operating 18 system. Then you ask for comments. 19 Do you see that? 20 Answer: Yes, I do. 21 Question: Do you recall receiving any 22 comments from the recipients of this e-mail to 23 your proposal? 24 Answer: Not specifically. 25 Question: Do you know whether there 11777 1 was a period of time when the device -- I'm 2 sorry -- when the disk cache driver that 3 started out as Bambi did refuse to run on 4 DR-DOS? 5 Answer: I believe so. I believe so. 6 Question: Mr. Barrett, you've been 7 handed Exhibit 15 to your deposition. 8 This is another series of e-mails 9 produced by Microsoft. I want to direct your 10 attention to the last e-mail on the second 11 page. This is the page that's Bates stamped 12 MS-PCA 1143038. 13 And it's an e-mail message from -- do 14 you recognize this as an e-mail message from 15 Scott Quinn to Bambi and Georgef dated Monday, 16 September 30th, 1991? 17 Answer: Yes. 18 Question: The subject is Bambi 19 version 3.5. 20 Do you see that? 21 Answer: Uh-huh. 22 Question: And in his e-mail message 23 Mr. Quinn says, Bambi version 3.5 has passed 24 developer testing. The primary change fixes a 25 major problem with accessing logical units on 11778 1 external hard disks. 2 Then he goes on to say, also DR-DOS is 3 detected (needs testing) and Bambi refuses to 4 load. 5 Is this consistent with your 6 recollection of the status of Bambi at the time 7 that it passed developer testing? 8 Answer: Yes, this is. Yes, this was 9 consistent. 10 Question: And the e-mail alias here 11 that's shown in the to line Bambi, was that a 12 group of Bambi developers? 13 Answer: Yes, that was the group of 14 people developing Bambi. 15 Question: What happened to Bambi 16 after it passed developer testing? I'm trying 17 to find out what the next stage of development 18 was. 19 Answer: That would go into the QA 20 group for more full testing. Developer testing 21 is simply the developers that built it ensure 22 that they believe that it was going enough 23 quality to try to test to the next level of 24 quality. 25 Question: You've been handed Exhibit 11779 1 16 to your deposition, which is another 2 conglomeration of e-mail provided by Microsoft. 3 I'd like to refer you to the e-mail on 4 the first page, Bates stamped MS-PCA 1179461. 5 Do you recognize this as an e-mail 6 from Chuck Straub to you and Scott Quinn dated 7 February 23, 1993? 8 Answer: Okay. Yep. 9 Question: Mr. Straub says, this is 10 the mail which announces -- which announced the 11 fixing of the, quote, bug, close quote, which 12 had previously prevented DR-DOS 6 from working. 13 As of this point, the DR-DOS 6 check could have 14 been removed. 15 And then embedded in his e-mail is a 16 copy of an e-mail that Mr. Quinn sent out to 17 the Bambi group on October 16, 1991. 18 Do you see that? 19 Answer: Yes, I do. 20 Question: Do you know whether the 21 DR-DOS 6 check was removed after the bug which 22 prevented DR-DOS 6 from working with Bambi had 23 it been fixed? 24 Answer: I'm not really sure. It does 25 not -- this does not ring a bell one way or the 11780 1 other. 2 Question: All right. Do you know 3 what the bug was that appears -- it refers to a 4 bug in quotes. Do you know what bug he was 5 referring to? 6 Answer: Well, I think it referred to 7 a bug -- well, it referred to the way that 8 Digital Research was handling the greater than 9 32 megabyte partitions. 10 Question: We covered that earlier? 11 Answer: Yes. 12 Question: You've been handed Exhibit 13 17 to your deposition, which is more e-mail 14 provided by Microsoft in this litigation. This 15 is Bates stamped MS-PCA 1148484. 16 Do you recognize this as an e-mail 17 from David Cole to you and an Karlst dated 18 September 30, 1991? 19 Answer: Yes. 20 Question: The subject line says 21 supported DOSes. 22 And Mr. Cole begins his e-mail by 23 saying, it's pretty clear we need to make sure 24 Windows 3.1 only runs on top of MS-DOS or an 25 OEM version of it. 11781 1 Why was it that Microsoft needed to 2 make sure Windows 3.1 only ran on top of MS-DOS 3 or its OEM version? 4 Answer: Well, this is -- this goes 5 back to some of the other documents we looked 6 at. 7 Pretty much Bill Gates wanted to make 8 sure that we put our competitive operating 9 systems at a disadvantage, and this is one way 10 that we would do it. 11 Question: Did you hear that, 12 Mr. Gates say that in words or substance? 13 Answer: Yes, I personally sat in 14 meetings where he said that. 15 Question: All right. Mr. Cole goes 16 on to say, I checked with legal and they are 17 working up some text we are supposed to display 18 if someone tries to set up or run Windows on an 19 alien operating system. 20 We are supposed to give the user the 21 option of continuing after the warning. 22 However, we should surely crash at some point 23 shortly later. 24 Were you aware of any plan to cause 25 Windows 3.1 to crash after a warning message 11782 1 came up if it was being used in conjunction 2 with an operating system other than MS-DOS? 3 Answer: Yeah, I was aware of this. 4 Question: Mr. Cole goes on to say, 5 now to the point of this mail. How shall we 6 proceed on the issue of making sure Windows 3.1 7 requires MS-DOS? We need to have some pretty 8 fancy internal checks to make sure we are on 9 the right one. 10 Maybe there are several very 11 sophisticated checks so the competitors get put 12 on a treadmill. 13 Do you know what that reference is to 14 putting the competitors on a treadmill, what 15 that refers to? 16 Answer: Certainly. The whole idea, 17 in general, in competition is to cause your 18 competitors to constantly be playing catchup. 19 And that's why the reference to treadmill, 20 running fast to stay in place. 21 Question: Was putting the competitors 22 on a treadmill, was that a phrase that you 23 heard occasionally at Microsoft? 24 Answer: More than occasionally. 25 Yeah, it was a very common phrase. I used it 11783 1 myself quite a bit. 2 Question: All right. Do you 3 understand how this suggestion by Mr. Cole in 4 this e-mail would accomplish putting 5 competitors on a treadmill? 6 Answer: Certainly. The whole idea is 7 -- well, Digital Research was very, very good 8 at modifying their operating system to overcome 9 any barriers we have put in front of them. 10 In fact, the bug -- the quote-unquote 11 bug from a previous document is an example 12 where they figured out what we were looking for 13 and then they fixed it so that they looked more 14 like MS-DOS than they did before. 15 Basically, what you're trying to do is 16 cause them to continually have to rerelease new 17 versions of their operating system in order to 18 stay compatible. And that's precisely what 19 this was talking about. 20 Question: You've been handed Exhibit 21 19 to your deposition. This is another set of 22 e-mails provided to us in discovery from 23 Microsoft. 24 I want to direct your attention to the 25 e-mail that begins at the bottom of the page 11784 1 from a Jenk to Cliffg and Scott Quinn dated -- 2 and with a copy to the DOS development group, 3 dated September 30, 1991. 4 Answer: Okay. 5 Question: And embedded in that is -- 6 or it's replying to a message from Cliffg where 7 he discusses, quote, the official way to detect 8 DR-DOS. 9 Do you see that? 10 Answer: Yes, I do. 11 Question: Okay. And then he 12 enumerates three steps. 13 Do you recognize those as steps that 14 would allow the program to detect the presence 15 of DR-DOS? 16 Answer: That certainly looks like it. 17 Question: Okay. Do you recall 18 receiving this e-mail? 19 Answer: No, I don't recall receiving 20 this e-mail. 21 Question: Cliffg refers to, quote, 22 the official way to detect DR-DOS. 23 Is that a reference to -- well, what 24 is the official way to detect DR-DOS? What 25 does that mean? 11785 1 Answer: Basically, I think it simply 2 means the Digital Research -- Digital Research 3 recommended way of detecting Digital Research 4 DOS. 5 Question: You've been handed Exhibit 6 20 to your deposition, which is another letter 7 that was provided to us in discovery by 8 Microsoft Bates stamped MS-PCA 1143151. 9 And it appears to be a letter from a 10 Bradley Kerth, K-e-r-t-h, senior technical 11 support engineer at Digital Research, to a 12 Roger Sour or director of Windows development 13 at Microsoft dated October 24, 1991. 14 Answer: Yes. 15 Question: Did this letter ever come 16 to your attention? 17 Answer: Yes, it did. 18 Question: All right. And do you know 19 who Roger Sour is, S-o-u-r? 20 Answer: I don't know anyone named 21 Roger Sour. 22 Question: Tell me the circumstances 23 of how this came to your attention and what 24 happened, if you can recall. 25 Answer: Well, this came to me because 11786 1 of the director of Windows development. 2 Question: That was you? 3 Answer: Yes, that was -- well, yeah, 4 I was the closest thing to it. 5 Question: And what was the issue that 6 was -- that came to light as a result of this 7 letter? 8 Answer: Well, I'm not sure how to 9 answer your question, to be honest. 10 This came to -- there was some problem 11 that somebody obviously called Digital Research 12 about and wanted to get more information. 13 Question: All right. Well, Mr. Kerth 14 says in his letter, it has come to my attention 15 that on September 30th, 1991, you, meaning 16 Mr. Sour, had contacted the Digital Research 17 technical support department for assistance 18 with DR-DOS 6.0. 19 You provided the serial number and 20 then he recites the serial number to our 21 technical support analyst, Andrew Dyson, 22 D-y-s-o-n, and proceeded to ask if there was a 23 way for a program to detect if it is running 24 under that operating system. 25 While this information is not 11787 1 generally handed out, we try to maintain a very 2 cooperative policy towards software 3 manufacturers. In following that policy, 4 Andrew described the technique to do so. 5 Do you know if anyone at Microsoft 6 made the phone call that's described here? 7 Answer: I have no personal knowledge 8 of anyone actually making that call. 9 Question: All right. Was any 10 investigation made to see whether that, at 11 Microsoft, made to determine whether this call 12 was made? 13 Answer: You use the word 14 investigation, probably overstates it, but, 15 yeah, I did some asking. 16 Question: You looked into it? 17 Answer: I looked into it, yes. 18 Question: And what did you find out? 19 Answer: I found nothing. 20 Question: Okay. Was there -- did it 21 ever come to light that this Cliffg made the 22 phone call that's described here? 23 Answer: I -- no. At least never came 24 to light to me. 25 Question: Okay. Mr. Kerth's letter 11788 1 goes on to say, when Andrew asked why you 2 needed the information, you indicated that you 3 were developing portions of the new cache 4 software for the future Windows 3.1, and you 5 had found a, quote, problem in the DR-DOS 6.0 6 memory control blocks (MCB), close quote. 7 Now, that reference to the cache 8 software, was that the Bambi disk cache driver? 9 Answer: Well, it certainly sounds 10 like an accurate description of it. 11 Question: Okay. What are the memory 12 control blocks? What does that refer to? 13 Answer: I honestly don't know 14 specifically what the intent there is. I'd 15 only be speculating. 16 Question: Earlier when we were 17 talking about the Bambi, the e-mails concerning 18 the Bambi development, you made reference to 19 the fact that there was a difference between 20 the way DR-DOS and MS-DOS handled something 21 that affected this disk cache driver. 22 Does that have -- what you were 23 referring to earlier, does that have something 24 to do with the memory control blocks? 25 Answer: That would be a very logical 11789 1 assumption. 2 THE WITNESS: Could we go -- could I 3 confer with counsel for just a second? 4 Question: Sure. Absolutely. 5 Mr. Barrett, during the break, you 6 said you wanted to clarify something. 7 Answer: Oh, yes. I wanted to clarify 8 Cliffg was Cliff Garrett. He was a temporary 9 employee, G-a-r-r-e-t-t, I believe. And he was 10 the -- he basically ran the -- he definitely 11 ran the SmartDrive beta test program. I 12 believe he also was part of the DOS 6 beta test 13 program. 14 Question: All right. And just to 15 connect it all up, the SmartDrive product was, 16 the code name for that during development was 17 Bambi? 18 Answer: That is correct. 19 Question: Thanks for that 20 clarification. I appreciate that. 21 Continuing with Exhibit 20. I want to 22 refer you to the middle of the second paragraph 23 where Mr. Kerth says, as I understand it, your 24 goal is to identify the presence of DR-DOS 6.0 25 so that your software will terminate itself 11790 1 after warning the end user that, quote, an 2 unsupported DOS, close quote, is being used. 3 Usually when a software manufacturer 4 feels that something in our operating system is 5 preventing their application from running well, 6 that company works with us to resolve the 7 actual perceived or potential conflicts. 8 Was that your experience in the 9 industry, that when one software manufacturer 10 found some conflict with another software, that 11 the two companies would generally try and work 12 together to resolve those conflicts? 13 Answer: I wouldn't say it's the rule, 14 but it's quite common for two companies to work 15 to eliminate conflicts or other problems 16 between the two pieces of software that they 17 have. 18 Question: All right. And generally 19 when that is done, what's the rationale for 20 that? Why would the two companies work 21 together to eliminate any conflicts between 22 their two software products? 23 Answer: Typically the goal, good 24 companies care about their customers' 25 experience. And so they want to make sure that 11791 1 the customers have a good experience regardless 2 of what software they're running. 3 Question: Mr. Barrett, you've been 4 handed Exhibit 21 to your deposition. And do 5 you recognize this as a letter that you sent to 6 Mr. Kerth at Digital Research on November 1, 7 1991? 8 Answer: Yes. 9 Question: All right. You say, I am 10 the development manager for Microsoft Windows 11 3.1. As such, I was given the letter sent by 12 you and addressed to, quote, Roger Sour, close 13 quote, or director of Windows development. 14 This was a very odd piece of mail to 15 receive, that there is no one at Microsoft by 16 the name of Roger Sour. 17 Further, whoever this Mr. Sour is, he 18 certainly does not speak for Microsoft. 19 Perhaps you may have been the victim of a 20 prank. 21 Did you ever directly ask Cliff 22 Garrett whether he had sent this -- or whether 23 he had made this call under the name of Roger 24 Sour? 25 Answer: No. 11792 1 Question: You never asked him that? 2 Answer: Sorry. No, I did not. 3 Question: The detection code that 4 Mr. Garrett obtained from DRI, Digital 5 Research, is that used to detect DR-DOS? Was 6 it used in Microsoft's products? 7 Answer: I honestly don't recall. At 8 least at this moment. 9 Question: You've been handed Exhibit 10 22 to your deposition. 11 Do you recognize this as a letter that 12 Mr. Kerth, Bradley Kerth from Digital Research 13 sent to you on November 14th, 1991? 14 Answer: Yes, I do. 15 Question: At the end of his letter 16 Mr. Kerth says, if at any point Microsoft 17 detects a compatibility problem or potential 18 problem between Microsoft Windows 3.1 and 19 DR-DOS 6.0, please call myself or Brian -- I'm 20 sorry, Byron Tomingas immediately to discuss 21 the situation. 22 Did you or anyone else that you know 23 of at Microsoft call Mr. Kerth or Mr. Tomingas 24 at Digital Research to try and resolve the 25 conflict between the Bambi product and DR-DOS? 11793 1 Answer: No. 2 Question: Before you look at Exhibit 3 23, I just want to go back to 22 for a moment. 4 In Exhibit 22 Mr. Kerth says, the 5 first paragraph, I take it from your letter 6 that you do not know who Roger Sour is. That 7 there is no problem between Windows 3.1 and the 8 DR-DOS 6.0 memory control blocks, and then 9 Microsoft has no intention of identifying the 10 presence of DR-DOS 6.0 so that Windows 3.1 will 11 terminate itself or otherwise inconvenience the 12 end user because the end user is running DR-DOS 13 instead of MS-DOS. 14 He then asks you to let him know 15 immediately if his understanding is incorrect. 16 Do you see that? 17 Answer: Yes. 18 Question: Now, I want you to take a 19 look at Exhibit 23. 20 Was -- do you recognize Exhibit 23 as 21 a letter that you wrote to Mr. Kerth at Digital 22 Research on December 2nd, 1991, replying to his 23 letter of November 14th of Exhibit 22? 24 Answer: Yes, I do. 25 Question: And in it you say, thank 11794 1 you for your letter dated November 14, 1991. 2 Because Windows 3.1 is an unreleased product, I 3 cannot disclose details as to specific features 4 or functions in that product. 5 Was that a policy at that time at 6 Microsoft, that you did not release details as 7 to features or functions in unreleased 8 products? 9 Answer: No. 10 Question: Why were you telling 11 Mr. Kerth that you would not release the 12 details as to the specific features or 13 functions of the product? 14 Answer: Because -- well, because we 15 simply didn't want to go there. 16 Question: You didn't want to let him 17 know that you were going to continue to detect 18 DR-DOS? 19 Answer: That's correct. We did not 20 want to -- we didn't want to say anything that 21 could be perceived as tying our hands. 22 Question: Did you ever determine 23 where Digital Research was getting their 24 information about what Microsoft was doing with 25 the detecting Windows -- detecting the DR-DOS 11795 1 code? 2 Answer: I'm not 100 percent clear on 3 your question. Perhaps you'd like to rephrase 4 it. 5 Question: Yeah. Mr. Cole ends his 6 message by asking, you know, where the hell is 7 DRI getting their information. 8 Do you see that? 9 Answer: Yes, I do. 10 Question: All right. Did you or 11 anyone else that you know of at Microsoft make 12 an investigation to determine how Digital 13 Research was getting information about what 14 Microsoft was doing in this regard? 15 Answer: Yeah. I actually spent not a 16 lot of time, but I spent some time trying to 17 determine who this person might be, if there 18 was indeed a leak within Microsoft. 19 I went so far as to talk to the phone 20 people to see if there was a way to check phone 21 records, and also talked to a number of people 22 on the team and directly asked them if they 23 knew anything about it, of which they all said 24 no. 25 Question: Okay. So you never did 11796 1 find out how Digital Research was getting its 2 information? 3 Answer: No, did not. 4 Question: All right. As part of 5 looking into this, did you determine -- or did 6 you find out that Cliff Garrett had been in 7 touch with Digital Research to get the 8 detection code? 9 Answer: By the evidence document you 10 showed me earlier because I don't recall -- I 11 don't recall that specific situation, and yet I 12 would have thought that he would be the likely 13 candidate for this had I put two and two 14 together. 15 Question: Mr. Barrett, you've been 16 handed Exhibit 25 to your deposition, which is 17 a group of e-mails produced by Microsoft in 18 this case. 19 I'd like to direct your attention to 20 the second e-mail up from the bottom, the one 21 that appears to be from you. 22 Do you see that? 23 Answer: Yes, I do. 24 Question: Do you recognize that as an 25 e-mail message that you sent to Brad 11797 1 Silverberg, a Buckf and David Cole on November 2 6, 1991? 3 Answer: Yes. 4 Question: In that e-mail you say, 5 this is -- the subject line is re: FTC and 6 DR-DOS. 7 And was this a reply to the message 8 that you received from David Cole that we just 9 looked at in Exhibit 24? 10 Answer: Yes, it is. 11 Question: In the e-mail you say, they 12 are getting very accurate information from an 13 internal source. I don't want a witch hunt, 14 but this is out of control. 15 Buck, this is pretty important. I 16 talked to Brad and he is in agreement that we 17 should ferret out the leak. 18 Is this -- are you talking about the 19 leak to Digital Research? 20 Answer: Yes, that's precisely what 21 I'm referring to. 22 Question: At the end you say, I have 23 suspicions about one specific person, but don't 24 want to say in e-mail. 25 Who at that time did you suspect was 11798 1 leaking information to Digital Research? 2 Answer: Well, there was a -- and I 3 don't remember the guy's name, but there was a 4 difficult person, like another support person 5 on the beta program who I thought was -- I just 6 thought that they were disgruntled. I don't 7 believe that it was that person, though, and 8 I'm sorry, I don't recall their name. 9 Question: Okay. You've been handed 10 Exhibit 26, Mr. Barrett, which is yet another 11 set of e-mails produced to us in this case by 12 Microsoft. 13 I want to direct your attention to the 14 e-mail at the very top of the page. This is 15 Bates stamped MS-PCA 1179419. 16 Answer: Okay. The top piece of the 17 e-mail? 18 Question: The very top of the e-mail. 19 Answer: Okay, I see it. 20 Question: Do you recognize this as an 21 e-mail from Brad Silverberg to David Cole and 22 yourself on -- dated October 29, 1991? 23 Answer: Yep. Yes. 24 Question: The subject is DR and 25 Windows 3.1. 11799 1 And you'll see that embedded in 2 Mr. Silverberg's e-mail is an e-mail from Richf 3 to Brad Chase and Brad Silverberg. 4 Do you see that? 5 Answer: Yes. 6 Question: And who's Richf? Is that 7 Richard Fade? Do you know who Richardf is? 8 Answer: I believe Richardf is Richard 9 Fade. 10 Question: All right. In his e-mail 11 Mr. Fade says, I loaded DR 6 on my own system 12 about a week ago and have been using it since. 13 And then he goes on to talk about some 14 details of what he experienced while running 15 it. 16 Answer: Uh-huh. 17 Question: But then he concludes his 18 e-mail by saying, in short, I haven't seen any 19 basic kernel incompatibilities. 20 Are you able to tell what he's talking 21 about there? Incompatibilities between what 22 and what? 23 Answer: Incompatibilities between -- 24 the incompatibilities are between Windows 3.1 25 and DR-DOS. 11800 1 Question: Okay. And he wasn't able 2 to find any kernel incompatibilities? 3 Answer: Yes, although Richard Fade 4 wouldn't know a kernel if it hit him. He's 5 basically saying Windows 3.1. 6 Question: Windows 3.1 what? 7 Answer: With DR-DOS. 8 Question: Okay. And then Brad 9 Silverberg says in the e-mail that Mr. Fade's 10 e-mail is embedded in, he says, looks like 11 DR-DOS works with Windows 3.1. 12 Was that a conclusion that your team 13 reached at the end of October, that DR-DOS 14 actually did work with Windows 3.1 at that 15 point in time? 16 Answer: Yeah, I believe so. I think 17 they had made some changes or something. 18 I do want to point out that there are 19 a set of incompatibilities that still existed, 20 particularly using their EMM 386, which is 21 their enhanced memory manager, which was a way 22 of providing a set of features to DOS 23 applications so that they had more memory. 24 Question: All right. And that's what 25 he's talking about, what Mr. Fade is talking 11801 1 about in his first paragraph of the embedded 2 e-mail? 3 Answer: Yes, that's correct. 4 Question: Are you able to tell or did 5 you know how that particular incompatibility, 6 how much of an impact that had on being able to 7 run DR-DOS and Windows together? 8 Answer: It wasn't a big impact. 9 (Whereupon, the playing of the video 10 adjourned.) 11 THE COURT: We'll take our recess 12 right now. 13 Remember the admonition previously 14 given. 15 We'll be in recess for about ten 16 minutes. Leave your notebooks here. 17 Thank you. 18 (A recess was taken from 9:47 a.m. 19 to 10:02 a.m.) 20 THE COURT: Everyone else may be 21 seated. 22 You may continue. 23 (Whereupon, the following video was 24 played to the jury.) 25 Question: Exhibit 27 is additional 11802 1 e-mail provided to us from Microsoft, Bates 2 stamped MS-PCA 1143105. 3 And I'd like to refer you -- there are 4 two e-mails here. I'd like to refer you to the 5 bottom one. 6 Do you recognize this as an e-mail 7 from Brad Silverberg to Brad Chase, Steve 8 Ballmer and Tom Lennon? 9 Answer: I'm sorry. What was the 10 question again? 11 Question: Let me ask you a question 12 about the bottom e-mail. 13 Answer: Okay. 14 Question: You see it appears to be an 15 e-mail from Brad Silverberg to Brad Chase and 16 Steve Ballmer and Tom Lennon? 17 Do you see that? 18 Answer: Yes, I do. 19 Question: All right. And it's dated 20 October 8, 1991. Subject, thought you'd 21 appreciate. 22 And Mr. Silverberg says, the following 23 excerpt from Janineh's Windows 3.1 beta report. 24 Do you know who Janineh was? 25 Answer: Yes, I know who Janineh was, 11803 1 although I don't recall her last name. 2 Question: All right. What was her 3 responsibility or position at Microsoft? 4 Answer: She was a program manager in 5 charge of the 3.1 beta program. 6 Question: All right. And I think 7 you've mentioned before that you also worked on 8 the 3.1 beta program; is that correct? 9 Answer: Yes. 10 Question: Did you work -- did you 11 have any input to these beta reports that's 12 referred to here, the Windows 3.1 beta report? 13 Answer: Yeah, I'm sure I had some 14 input. In general, it was more reporting about 15 what was going on in the beta program, status 16 and that sort of thing. 17 Question: Now, Janineh, there's a 18 quote from her 3.1 beta report that's embedded 19 in the e-mail. 20 Do you see that at the bottom? 21 Answer: Yes, I do. 22 Question: And it says weekly -- under 23 the heading weekly activity in the report, it 24 says, Windows 3.1 doesn't run with DR-DOS. 25 And Janineh reports, we sent one 11804 1 person MS-DOS 5.0 to use and Randym is working 2 with another large account. This may stop them 3 from going to DR-DOS 6.0. 4 Do you know who Randym was? 5 Answer: No, I don't recall at this 6 moment. 7 Question: Do you know whether 8 Microsoft's sales and marketing people were 9 using the alleged incompatibilities between 10 DR-DOS and Windows to try and stop customers 11 from using DR-DOS? 12 Answer: I don't have any direct 13 personal knowledge of that. 14 Question: Did you learn of it 15 indirectly? 16 Answer: It was common knowledge that 17 the differences were being used by the sales 18 groups. 19 Question: Did people talk about that 20 fact at Microsoft? 21 Answer: It was fairly common, fairly 22 common discussion point. 23 Question: You've been handed Exhibit 24 28 to your deposition. 25 Do you recognize this as an e-mail 11805 1 message from David Cole to Brad Silverberg with 2 a copy to you dated November 4, 1991? 3 Answer: Yes. 4 Question: In this Mr. Cole talks 5 about where things stand on the wording of the 6 warning, what he refers to as the warning. 7 Do you know whether either of these 8 versions of the warning were actually 9 implemented in Windows 3.1? 10 Answer: I don't recall if these 11 specific ones were put in, but I know that 12 something very similar to the bottom one was 13 put into the beta release of 3.1. 14 Question: You've been handed Exhibit 15 29 to your deposition. 16 Do you recognize this as an e-mail 17 message from David Cole to Royha and Timbr with 18 a copy to you dated November 8th, 1991? 19 Answer: Yes, I do. 20 Question: Do you recall who Royha 21 was? 22 Answer: No, I don't. 23 Question: Or Timbr? 24 Answer: No, I don't. 25 Question: You mentioned earlier that 11806 1 -- well, this e-mail begins, in order to serve 2 our Windows customers properly, we are going to 3 warn them if they set up Windows or run MSD on 4 systems not using MS-DOS. 5 MSD refers to? 6 Answer: Microsoft diagnostics, I 7 believe. 8 Question: Okay. Mr. Cole goes on to 9 say, Aaronr -- is that Aaron Reynolds? 10 Answer: That is correct. 11 Question: Aaron Reynolds is 12 developing a piece of code which checks for the 13 absence of MS-DOS. Phil Barrett is working 14 with him. 15 Now, what was the code that Aaron 16 Reynolds was working on at this point in time? 17 Answer: That was a piece of code to 18 detect whether or not we're running on DR-DOS. 19 Question: And what work were you 20 doing with Mr. Reynolds in connection with 21 that? 22 Answer: Well -- 23 Question: Mr. Barrett, you've been 24 handed Exhibit 32 to your deposition. 25 Do you recognize this as an e-mail 11807 1 message from Aaron Reynolds to Karlst, yourself 2 and Sandyps? 3 Answer: Yes, I do. 4 Question: And it's dated February 24, 5 1993. 6 In the message, Mr. Reynolds begins by 7 saying, yesterday it was decided to do this 8 detection and display the message if 9 appropriate every time we run Windows. Aaron 10 has some code that you add to win.com. 11 Of course, the subject of this is 12 MS-DOS detection-hot job for you. 13 In December of 1991, was it, in fact, 14 decided that the Windows detection warning 15 message would be -- would be displayed every 16 time Windows is run? 17 Answer: Yes, it was. 18 I do want to point out that the date 19 on this is actually Friday, December 6th, 1991, 20 not Wednesday, February 24th, 1993. 21 Question: Thanks for making that 22 clarification. 23 You've been handed Exhibit 33, which 24 is another series of e-mail messages produced 25 by Microsoft in this case. 11808 1 I'd like to begin by asking you about 2 the message embedded in the first message. 3 You'll see that appears on page 2, and 4 it appears to be from Chuckst. Is that Chuck 5 Straub? 6 Answer: Yes, that's Chuck Straub. 7 Question: To you dated December 16, 8 1991, is that correct? 9 Answer: Yes. 10 Question: Mr. Straub in his e-mail 11 message to you says, sorry I wasn't able to get 12 these things done this evening. I had a prior 13 commitment, and he lists three things, one of 14 which was itemized Number 2, MS-DOS detection 15 code change. 16 Aaronr already came by my office with 17 a new object module which we checked in. And 18 then he goes on to talk about some details 19 about the MS-DOS detection code change. 20 And then what I want to ask you about 21 is whether this detection code was important 22 for the beta release of Windows 3.1? 23 Answer: Yeah, it was absolutely -- 24 yeah, it was absolutely considered important at 25 that time. 11809 1 Question: And was that the substance 2 of your message above where you said, business 3 as usual ain't good enough. We've got to 4 absolutely jump on this stuff. It can't wait 5 even a couple of hours, let alone a half a day. 6 I want you to make this your number one 7 priority. This beta release is critical to the 8 success of Windows 3.1. 9 Was that what you were trying to 10 convey to him? 11 Answer: Yes, that's exactly what I 12 wrote. 13 Question: Why the rush here? Was 14 there some urgency to getting this beta release 15 out? 16 Answer: Well, there were two things 17 going on. One was, yes, it was -- we had as a 18 management team decided that we were going to 19 do this. 20 Secondly, Chuck, who wasn't obvious -- 21 not terribly obvious in this mail, was a 22 performance problem and he was not doing his 23 job and he didn't come in. And so there were 24 two purposes of this mail. 25 One is to make it clear that we were 11810 1 going to put the MS-DOS detection code in the 2 beta and we were going to get it done. 3 And, secondly, was putting him on 4 notice that he was not doing his job, and that 5 he was on thin ice. 6 And this is called building a paper 7 trail when you have a performance problem. 8 Question: You've been handed Exhibit 9 35 to your deposition. 10 Mr. Barrett, do you recognize this as 11 an e-mail message from David Cole to Aaron 12 Reynolds, yourself, and others at Microsoft 13 dated February 11, 1992? 14 Answer: Yes, it is. 15 Question: Mr. Cole begins his e-mail 16 message by saying, after a long round of 17 debates, the executives have decided that no 18 message should be displayed when we fail to 19 detect MS-DOS. 20 The PR impact of showing a message 21 would be great and would consume much of our 22 time just dealing with the reaction from the 23 press and others. 24 First of all, were you involved in the 25 decision to not make use of the warning message 11811 1 in the final released product of Windows 3.1? 2 Answer: Yes, I was. 3 Question: And can you tell me how 4 that decision came about? 5 Answer: Well, I wasn't subject to all 6 of the discussions that Steve Ballmer and Brad 7 Silverberg had, but, basically, Steve Ballmer, 8 and I personally heard him say it, he just 9 thought it was a stupid thing to do. And that 10 went on for probably two weeks back and forth. 11 We finally reached consensus to not do it. 12 Question: All right. Now, at the 13 time that this decision was made, had the 14 Windows 3.1 beta release gone out to the beta 15 test sites? 16 Answer: Yes, it had. 17 Question: And the warning message did 18 appear in those beta copies; is that correct? 19 Answer: Yes, it did. 20 Question: Mr. Cole ends his message 21 by saying, I want to thank everyone for their 22 effort on this in putting up with the changes. 23 It was a close call on what to do. 24 Then he says, we will have a readme 25 item that explains we test it only on MS-DOS. 11812 1 What was readme? 2 Answer: Readme is a file, usually the 3 name of the file is read me.text and read 4 me.doc, which almost all software products, 5 when they ship, they include a readme file that 6 has very late breaking information that might 7 not have made it into the manual or something 8 like that. 9 The readme item specifically refers 10 to, you know, one item. They're typically 11 numbered or lettered, and it refers to one 12 specifically numbered or lettered item that 13 would be in there. 14 Question: All right. And what is the 15 purpose of a readme file? How does the end 16 user make use of that? 17 Answer: Typically, it's to 18 communicate very late breaking information or 19 incompatibilities or problems that have been 20 found and gives the users ways to work around 21 those problems. Basically, it's a 22 communication vehicle for very, very recent 23 information that you hadn't been able to 24 formalize. 25 Question: Exhibit 36 has just been 11813 1 handed to you, and this is, again, a series of 2 e-mail messages that was produced to us in this 3 litigation by Microsoft. 4 Let me ask you about the top e-mail 5 message. 6 Do you recognize this as an e-mail 7 message sent by Richab to Brad Silverberg, 8 David Cole, and yourself dated Friday, November 9 30, 1990? 10 Answer: Yes, I do. 11 Question: And was this a response to 12 the e-mail messages that appear below it? 13 Answer: Yes, I believe it is. 14 Question: In the e-mail message at 15 the bottom of the page from it appears to be 16 Amitc. Do you know who that is? 17 Answer: Yes, that's Amit Chaterjee. 18 He's a software engineer on Windows. 19 Question: To a Danq. Who's Danq, do 20 you know? 21 Answer: Doesn't -- doesn't ring a 22 bell. 23 Question: Dated November 27, 1990. 24 Amit Chaterjee says, I have been told 25 that I will not need to answer support 11814 1 questions on the VxD anymore. 2 That is virtual device driver? 3 Answer: Yes, that's correct. 4 Question: Some people still have my 5 number and they call me. I told them that we 6 now have a different group handling support 7 issues and that Tim is the person to talk to. 8 However, I think it has been decided 9 that Digital Research will not be supported. 10 Rich Abel should have the list of vendors who 11 we do not want to support. 12 Let me ask you, why did Microsoft 13 decide not to support Digital Research? 14 Answer: Because they were a 15 competitor. 16 Question: He makes reference to a 17 list of vendors who we do not want to support. 18 Are you familiar with that list? 19 Answer: I'm familiar that it existed. 20 Question: Was that sometimes referred 21 to as the beta blacklist? 22 Answer: I never heard it called that 23 when I was at Microsoft. 24 Question: What was this list called? 25 Answer: The list of companies we 11815 1 didn't want to support. I don't believe it 2 actually had a specific name. 3 Question: All right. Do you know who 4 was in charge of keeping that list, maintaining 5 it? 6 Answer: I honestly don't know. I'm 7 not sure that anyone really was. I think it 8 existed more in the form of, you know, word of 9 mouth and sort of common understanding. 10 Question: Do you know how long the 11 list was? 12 Answer: At this point I can only 13 recall of one company that would have been on a 14 list, if it existed in physical form, and 15 that's Digital Research. 16 Question: All right. It goes on say, 17 quite some time back DRI was sent a very early 18 version of VxD, which, again, you said is 19 virtual device driver. 20 I don't know what to tell them. I 21 guess we must somehow politely let them know 22 that we don't want to support them. I don't 23 feel very comfortable in this situation and 24 would not want to deal with Digital myself. 25 Do you know why Microsoft gave Digital 11816 1 Research a copy of the digital device driver if 2 Microsoft was not going to support Digital 3 Research? 4 Answer: I believe that it was a 5 mistake that it got sent to them in the first 6 place. 7 Question: I see. 8 Answer: I know that it was a mistake. 9 Question: All right. You've been 10 handed Exhibit 38 to your deposition, 11 Mr. Barrett. 12 I want to refer you only to the e-mail 13 at the bottom of the first page. 14 I'm sorry, I take that back. 15 There's some embedded e-mails, but 16 it's an e-mail about a third of the way up from 17 the bottom from Brad Silverberg to a number of 18 people, including yourself, on July 26, 1991. 19 Answer: Yes. 20 Question: Subject re: Digital 21 Research. 22 Do you recognize this as an e-mail 23 that Mr. Silverberg sent to you and others on 24 July 26, 1991? 25 Answer: Yes, I do. 11817 1 Question: Mr. Silverberg asked the 2 question how does DRI have Windows 3.1 in the 3 first place? They are on the beta blacklist. 4 I was just -- I wanted to show this to 5 you to ask you whether this refreshes your 6 recollection as to whether there was something 7 called the beta blacklist at Microsoft? 8 Answer: I guess this is pretty black 9 and white. I guess I never recalled it as the 10 beta blacklist, but, obviously, Brad 11 Silverberg's words speak for themselves. 12 Question: Was there someone who had 13 authority to determine who would be denied 14 support, who would appear on this, 15 Mr. Silverberg calls it the beta blacklist? 16 Answer: Sorry. Yes, Silverberg. 17 Question: Mr. Silverberg made the 18 determination as to who would be denied 19 support? 20 Answer: Yeah. 21 Question: Just to satisfy counsel, 22 were you personally present at a meeting where 23 Mr. Silverberg talked about denying Digital 24 Research the 3.1 beta? 25 Answer: Yes, I was. 11818 1 Question: And what did he say? 2 Answer: He said that there -- well, I 3 won't say precisely the way he said it, but he 4 basically said that DR would never receive a 5 copy of the 3.1 beta. They're a competitor. 6 Question: Do you recall precisely how 7 he put that? 8 Answer: Yes, I do. 9 Question: And would you tell us, even 10 though if -- even if it is -- 11 Answer: He said there's no fucking 12 way in hell that Digital Research will ever 13 receive a copy of the 3.1 beta. They're a 14 goddamn competitor. 15 Question: Thank you. 16 You never saw a physical list of 17 people who were on this beta blacklist? 18 Answer: No, I never personally saw 19 one. 20 Question: You've been handed Exhibit 21 39 to your deposition, Mr. Barrett. 22 This is another series of e-mail 23 messages provided by Microsoft in this case. 24 You'll see in the middle of the page 25 is an e-mail message that appears to be from 11819 1 Brad Silverberg to a Jancl and a Pattys, 2 subject Digital Research. 3 Do you see that? 4 Answer: Uh-huh. Yes. 5 Question: And it starts by saying, we 6 should not be providing Digital Research any 7 assistance getting their operating system to 8 work with our software. Our software supports 9 MS-DOS, not DR-DOS. It's completely up to them 10 to figure out how to resolve any problems that 11 may occur. 12 Did that statement there reflect 13 statements, other statements that you heard 14 from Mr. Silverberg with respect to supporting 15 Digital Research? 16 Answer: I think this is a very polite 17 way of saying we don't support our competitors. 18 It's their problem. 19 Question: And then above that is -- 20 at the very top of the page is an e-mail 21 message. 22 Do you recognize that as an e-mail 23 message sent by Mr. Silverberg to you and 24 others at Microsoft on Monday, July 29th, 1991? 25 Answer: Yes. 11820 1 Question: There Mr. Silverberg says, 2 you bet. DRI is on the DOS/Win blacklist. 3 Nothing gets shipped from this BU without the 4 blacklist being checked. 5 First of all, what was the reference 6 to a BU? 7 Answer: That's business unit. The 8 Windows and DOS business unit. 9 Question: All right. Now, he makes a 10 reference to the blacklist being checked. Was 11 there -- what was the process, if you know, for 12 checking the beta blacklist? 13 Answer: I don't think there was a 14 process. 15 Question: Okay. And then embedded in 16 that message is a message about a third of the 17 way down the page. 18 Do you recognize that as an e-mail 19 message from Steve Ballmer to Brad Silverberg, 20 yourself, and others at Microsoft dated July 21 28, 1991? 22 Answer: Yes, I do. 23 Question: He starts that out by 24 reference to BARD. Do you know what that is? 25 Answer: Yeah, that's a misspelling of 11821 1 Brad. 2 Question: Oh, okay. 3 Brad, please make sure we are not 4 supporting DRI anywhere in the company with 5 this stuff, thanks. 6 Other than receiving this e-mail from 7 Mr. Ballmer, did you ever hear Mr. Ballmer make 8 statements of that nature? 9 Answer: I don't recall specifically 10 hearing him say that. 11 Question: Was this company policy at 12 Microsoft that you are not going to support DRI 13 anywhere in the company? 14 Answer: I'm really not sure how to 15 answer your question. You know, policy makes 16 it sound formal. 17 Question: There was nothing written 18 other than this e-mail that you're aware of? 19 Answer: No, I'm not -- sorry. I'm 20 not aware. 21 Question: Mr. Barrett, you've been 22 handed Exhibit 42 to your deposition. 23 This is, again, another series of 24 e-mails produced to us by Microsoft. 25 I'd like to direct your attention to 11822 1 the e-mail that begins at the bottom of the 2 first page. This is Bates marked X 575838. 3 Do you recognize this as an e-mail 4 message sent to you and others at Microsoft by 5 Ben Slivka on November 26, 1990? 6 Answer: Yes, I do. 7 Question: The subject line says, to 8 share or not to share, that is the question. 9 And the next series of exhibits is 10 going to -- Mr. Barrett, is going to concern 11 Microsoft's actions with the independent 12 software vendors. 13 Here Mr. Slivka begins his e-mail by 14 asking or stating the issue should Microsoft 15 applications (and other ISVs) have access to 16 Windows source code. 17 And you had stated earlier that ISVs 18 refers to independent software vendors? 19 Answer: That's correct. 20 Question: Mr. Slivka says -- you do 21 recognize this is an e-mail sent by Mr. Slivka 22 to you? 23 Answer: Yes, I do. 24 Question: All right. Mr. Slivka goes 25 on to say under the heading complications, and 11823 1 he says, number one, if Microsoft applications 2 have access to the Windows sources, then all 3 independent software vendors should have 4 access, else we are subject restraint of trade 5 complaints (to say nothing of the morality of 6 the situation). 7 Now, what was Mr. Slivka's position at 8 Microsoft? 9 Answer: Ben Slivka was a -- probably 10 still is, a software engineer, fairly senior 11 one. Project lead, technical lead type at that 12 point. I don't recall his specific role at 13 that time, though. 14 Question: At the time of this e-mail 15 message in 1990, were some independent software 16 vendors complaining that Microsoft applications 17 had some unfair advantage because they had 18 access to Windows source code? 19 Answer: I don't specifically recall 20 if it was happening at this time frame, but it 21 was happening. 22 Question: Well, let me ask you this. 23 Was it an advantage for applications developers 24 to have access to Windows source code? 25 Answer: Absolutely, yes. 11824 1 Question: And why is that? 2 Answer: Well, as it pointed out in 3 this memo, which I completely agree with, 4 Windows is poorly documented and there are 5 undocumented APIs, application program 6 interfaces, and there's behavior that is poorly 7 understood even by the Windows developers. 8 Question: And how does access to the 9 source code provide an applications developer 10 with some advantage? 11 Answer: Application developers, by 12 looking at the source code, while a fairly 13 arduous and difficult process, one that a small 14 ISV probably couldn't do or couldn't afford to 15 do, it would allow them to dedicate a person to 16 become a Windows internal expert and to be able 17 to advise their development teams on the best 18 way to structure their applications to perform 19 the best on Windows, in the Windows 20 environment. 21 Question: Does access to Windows 22 source code provide applications developers 23 with a time advantage? Does it make it easier 24 or faster to develop an application if you have 25 that access? 11825 1 Answer: Certainly time advantage and 2 efficiency advantage can be gotten by 3 understanding the internals of Windows. 4 Question: Does access to Windows 5 source code provide application developers with 6 any sort of cost advantage? 7 Answer: I'd have to say yes, although 8 it's not as simple a formula. 9 Question: If you'll look at 10 Mr. Slivka's e-mail message, as I said, it 11 starts with the heading issue and then it goes 12 on with complication. We read complication 13 number one. 14 He goes on in complication number two 15 to say, if outside independent software vendors 16 have access to our sources, then we make it 17 much easier for another company to come along 18 and clone Windows. 19 Let me ask you this. Were there 20 companies outside of Microsoft that were given 21 access to Windows source? 22 Answer: Yes, there were. 23 Question: And did Microsoft protect 24 themselves from the potential of cloning 25 Windows with a nondisclosure agreement? 11826 1 Answer: I can't answer one way or the 2 other. I simply don't know. 3 Question: All right. Now, when you 4 worked at Microsoft, you had access to Windows 5 source code, correct? 6 Answer: Yes. 7 Question: And you currently work for 8 a competitor of Microsoft; is that correct? 9 Answer: That's correct. 10 Question: Did you sign a 11 nondisclosure agreement at Microsoft that would 12 prevent you from disclosing Windows source code 13 to a competitor that you might later work for? 14 Answer: Yes. 15 Question: And I assume you have not 16 violated that agreement? 17 Answer: Absolutely not. 18 Question: Mr. Slivka goes on in his 19 e-mail message about a quarter of the way down 20 the page to say, the state machine that is 21 user.exe is barely documented in SDK. 22 Let's see if we can define some terms 23 here. SKD is the software development kit? 24 Answer: Software development kit, 25 yes. 11827 1 Question: And what is user.exe? 2 Answer: User.exe is one of the key 3 modules of Windows, parts of Windows. User 4 basically implemented all of the user 5 interface, Windows, menu, sliders, all that 6 stuff. That's what user.exe implemented. 7 Question: And the term state machine, 8 what does that refer to? 9 Answer: It's a technical term from 10 mathematics and electrical engineering that 11 refers to -- it's called finite state machine 12 is actually the correct terminology for it. 13 And it basically is -- there's a 14 concept of a state, which might be, say, 20 or 15 30 or 100 or a thousand different states and 16 inputs that change the state. 17 To put it in simple terms, think of it 18 as a machine that when you turn a dial, it 19 changes the internal gears. Very, very similar 20 sort of thing happens, although it's all 21 electronic. 22 And -- boy, I'm sorry, it's a very -- 23 it's a very specific technical term to 24 electrical engineering and computer science. 25 Question: I appreciate your effort to 11828 1 explain it to a lay person like myself. 2 Was it important to an applications 3 developer to have well-documented user.exe? 4 Answer: Clearly it was very 5 important. 6 As it pointed out in Ben's extremely 7 clear memo, in order to create an application 8 that has a good user interaction, that feels 9 normal, that feels like it, feels natural to 10 the user, you have to understand how the 11 windowing, the user interface system of Windows 12 works. 13 And you have to understand it in a way 14 that when you make a change here, that the 15 various other parts of it line up properly. 16 And so that when you pull down a menu, 17 that it looks right and that it reacts as if it 18 was effortless rather than, you know, laborious 19 or something like that. 20 So it's extremely important to 21 understand the internal workings of a windowing 22 system in order to write a window -- a 23 windowing application. 24 Question: Did you agree with 25 Mr. Slivka's statement here, that the user.exe 11829 1 was barely documented in the software 2 development kit? 3 Answer: Absolutely agree. Completely 4 right. 5 Question: And without getting too 6 technical, if you can, can you describe in 7 general terms what was not documented that 8 should have been? Are there general categories 9 of types of information? 10 Answer: Yeah. For example, in order 11 to get a window to -- the way Windows work is, 12 you would write some data into a window and 13 then you'd tell Windows to redisplay it. 14 So say if you were going to change the 15 contents of a menu, you'd have to write into 16 the menu and redisplay it. 17 In certain circumstances, those 18 redisplay messages or commands would get lost. 19 And so it became something that people learned 20 about and sort of folklore that you send -- in 21 certain circumstances, you send two redisplay 22 messages in order to get a window to actually 23 display. 24 And that's a very, you know, that's a 25 common case. There were others similar to that 11830 1 where it's just people figured out how to make 2 a Windows application work well by, in fact, 3 what does he say, experimentation. And that's 4 -- that's why it's clearly -- that's an aspect 5 of it being poorly documented. 6 Question: Mr. Slivka goes on to say, 7 any independent software vendor that wants to 8 write a great Windows application ends up 9 looking at the source code (like our 10 applications group) unassembling the DLLs or 11 writing experiment code to devine the actual 12 behavior of the system. 13 First of all, what are DLLs? 14 Answer: Dynamic link library is what 15 DLL stands for. And, effectively, it's the 16 building blocks of Windows. All the various 17 Windows pieces, device drivers, everything are 18 DLLs, dynamic link libraries. 19 Question: Now, in this sentence 20 Mr. Slivka identifies three ways that an ISV 21 can write, quote, great Windows applications. 22 They can either look at the source 23 code like the Microsoft applications group. Do 24 you see that? 25 That's the first one he identifies. 11831 1 Did the -- or to your knowledge, did 2 the Microsoft applications group have access to 3 Windows source code? 4 Answer: I believe they did. 5 Question: And can you tell me what 6 the basis for that belief is? 7 Answer: There were a number of -- 8 there were a lot of friendships between people 9 in the systems group and the applications 10 group. There was a lot of informal sharing 11 going on. 12 And application developers would show 13 up in the area where the Windows developers 14 worked and ask questions of developers, and 15 they'd sit there and look at code and answer 16 questions. 17 I believe that the apps group had 18 access to -- they did indeed have access to our 19 source code control system and they could look 20 and find out what was going on in the source 21 code. So they were only to look at the source 22 code. 23 Question: And what is the source code 24 control system? 25 Answer: It's a computer program that 11832 1 basically allows whoever the manager of the 2 source code to control access by giving people 3 passwords and also maintaining the integrity of 4 the source code. 5 People could check out the source code 6 and make changes and check it back in. If 7 somebody had a piece of source code checked 8 out, nobody else could check it out because, 9 you know, didn't want to have two people 10 writing in the same document, if you will, and 11 then having to merge the changes back. The 12 document in this case being the piece of code. 13 Question: Did anyone from the 14 Microsoft Excel spreadsheet application group 15 have access to the expertise of the Windows 16 development group? 17 Answer: Absolutely, yes. 18 Question: Can you tell me who that 19 was and what the circumstances were? 20 Answer: I believe there were more 21 than one, but in particular Chris Peters, the 22 leader of that time, and subsequently vice 23 president of Microsoft was -- frequented the 24 Windows development area and was actually quite 25 good friends with a number of people there and 11833 1 spent a lot of time talking with the Windows 2 developers and looking at the Windows code and 3 really understanding it. 4 Question: Did Mr. Peters work 5 evenings in the Windows development building? 6 Answer: He was a frequent visitor, 7 yes, in the evenings. 8 Question: Looking back at 9 Mr. Slivka's e-mail message in Exhibit 42, 10 after referring to looking at Windows source 11 code like Microsoft's application group, he 12 identifies another way that this can be done. 13 He refers to unassembling the DLLs. 14 Can you describe how that works? 15 Answer: Unassembling means to take -- 16 or disassembling actually. The more common 17 term is disassembling. Means the same thing. 18 Unassembling, in this case is 19 referring to taking a piece of -- taking a 20 program and using a special program called a 21 disassembler and running it on that program and 22 turning the machine, basically the numbers, the 23 operation codes and other data into what they 24 call pneumonics or a human readable form that's 25 describing the sequence of instructions that 11834 1 the binary code refers to. 2 And that would allow someone to look 3 at how a piece of software works. It's not 4 particularly efficient. And it's actually very 5 error fraught, but it does allow somebody to 6 understand what's going on if they allow enough 7 time and thought to it. 8 Question: And I want to understand 9 what you said. It is error fraught or it's not 10 error fraught? 11 Answer: It's highly -- it's highly 12 error fraught. It's complex and difficult to 13 do. And only a few people -- there's a small 14 number of people in the world that are good at 15 it. 16 Question: So susceptible to error, is 17 that what you're saying? 18 Answer: Yes. 19 I'm sorry. 20 Question: Then Mr. Slivka also talks 21 about an ISV looking at source -- I'm sorry, an 22 ISV writing great Windows apps by writing 23 experiment code. 24 Do you see that? 25 Answer: Yes. 11835 1 Question: To devine the actual 2 behavior of the system. What is that? 3 Answer: What that refers to is 4 writing a piece of code that does a specific 5 thing. I don't know. Calls the menu code or 6 whatever, Windows management code and tries 7 different things. Tries different inputs. 8 Allows the developer that's doing 9 this, the experimenter, if you will, to try 10 different things to see how the system behaves 11 based on different input. 12 And when different input -- by giving 13 different input to the system and seeing how it 14 behaves, you can start to form a model of how 15 the system behaves given different input. 16 That allows the developers to do a 17 better job of working with the system as it 18 actually works, not the way it's documented. 19 And this is essential when the 20 documentation does not match the actual 21 behavior of the system, which was indeed the 22 case with Windows 3.0 in many ways. 23 Question: Can you think of any 24 specifics as to how Windows 3.0 documentation 25 did not match the behavior of the system? 11836 1 Answer: Well, certainly the repaint 2 -- or the redraw thing I pointed out before, 3 where in some cases you had to send two redraw 4 messages to a window to cause it to actually 5 display its updated contents. 6 Question: That was not documented? 7 Answer: No. That was documented 8 nowhere. 9 Question: Can you think of any others 10 offhand? 11 Answer: Not off the top of my head. 12 Question: Is this system of writing 13 experiment code to devine the actual behavior 14 of this system, is that a particularly 15 efficient way of going about developing 16 applications? 17 Answer: No. It's pretty inefficient, 18 and it's also frankly -- it's inefficient for 19 another point of view in that it's not always 20 effective because it depends on how much 21 experimentation you do. 22 Question: If you look about halfway 23 down the page, Mr. Slivka says -- starts at the 24 paragraph that says, I claim that letting an 25 independent software vendor looking at the 11837 1 source code is the best way to avoid this 2 problem. 3 Then he enumerates some of his 4 concerns. He says, number one, an ISV that 5 assembles -- I'm sorry. 6 An ISV that unassembles Windows to 7 figure out its behavior is effectively looking 8 at source code, but without the benefits of 9 source comments. 10 This approach is more work for the ISV 11 and gives Microsoft no opportunity to guide the 12 ISV. With source code, there are generally 13 comments discussing rationale for the behavior 14 of the system. 15 First, I want to ask you, how much 16 more work is involved to devine how Windows 17 works by unassembling it? 18 Answer: It's -- it's a lot of work. 19 It's significantly more. 20 Question: Mr. Slivka refers to these 21 comments that are associated with the source 22 code. 23 How important is it to have access to 24 that if you're developing applications? 25 Answer: It's critical. In some ways 11838 1 the comments are more important than the actual 2 source code, the actual instructions. 3 Comments -- an instruction might say A 4 equals B plus C, which means you're adding the 5 variables B and C and putting them in A. 6 The comment might say, you know, 7 adjust the clock by some preset, or something 8 like that. 9 So, whereas, the source tells you, the 10 code tells you what specific operations are 11 being done, the source code comments tell you 12 why it's being done and what effect you're 13 trying to create. 14 Question: And why is that important 15 in the applications development? 16 Answer: It's important in order to 17 understand what the code is trying to do. Why 18 it's doing what it's doing. 19 Without those comments, you have to 20 look at what it does, understand the context in 21 which it's used, understand what the data is 22 that's being operated on. You really are 23 guessing. 24 And, typically, when someone does 25 disassembling, to understand something, they 11839 1 typically start commenting the disassembled 2 code and they actually write their own comment. 3 They put questions in and then they 4 fill in the questions as they understand what's 5 going on. So that's a very common thing. And 6 comments are far and away the most important 7 part of a good well-written piece of code. 8 Question: Item Number 2 under that 9 paragraph says, it begins, an independent 10 software vendor that writes test applications 11 to devine Windows behavior is really on thin 12 ice. 13 Either the independent software vendor 14 spends a great deal of effort writing test code 15 to be certain Windows is fully understood or 16 the independent software vendor may end up 17 making assumptions, which are not correct. 18 What would be the impact on 19 applications development if the independent 20 software vendor were to make assumptions that 21 were incorrect about the way Windows worked? 22 Answer: Well, there are two -- there 23 are probably more than that, but there are two 24 that come to mind immediately. 25 Number one, they might write code that 11840 1 doesn't work properly in certain circumstances. 2 But probably worse is that as Windows evolves, 3 as a new version of Windows comes out and some 4 of that behavior which they're relying on 5 changes, it can cause the application to fail 6 with newer versions of the operating system. 7 (Whereupon, playing of the video 8 adjourned.) 9 MR. GRALEWSKI: Your Honor, if it's 10 acceptable, this would be a good breaking 11 point. 12 THE COURT: Very well. We will be in 13 recess until 12 noon. 14 Remember the admonition previously 15 given. Leave your notebooks here. 16 All rise. 17 (A recess was taken from 10:57 a.m. 18 to 12 p.m.) 19 THE COURT: Everyone else may be 20 seated. 21 You may continue. 22 (Whereupon, the following video was 23 played to the jury.) 24 Question: Mr. Barrett, you've been 25 handed Exhibit 43 to your deposition. 11841 1 This is a letter written on Microsoft 2 stationery, which shows in handwriting at the 3 very top of the page, shows a copy to you, but 4 the typewritten portion appears to be from 5 Mr. Gates to Mr. Ballmer and others at 6 Microsoft dated December 30, 1988. 7 Answer: Yes. 8 Question: Do you recall receiving a 9 copy of this letter or I should say looks like 10 a memo, an interoffice memo? 11 Answer: Well, to be honest, I don't 12 recall getting it. 13 Question: I just wanted to ask you 14 about the substance of Item Number 3 in this 15 interoffice memo where Mr. Gates says, final 16 version source code for released products can 17 be released to qualified OEM customers on the 18 approval of the appropriate business unit 19 manager in hardware applications or systems and 20 Jeremy Butler. 21 Was this an example of what you 22 testified to before where in some cases source 23 code was released to third parties by 24 Microsoft? 25 Answer: Yes, it is precisely. 11842 1 Question: Was it your understanding 2 that some OEM customers did receive final 3 version source code for Microsoft products? 4 Answer: Yes. 5 Question: And do you know what the 6 purpose of releasing source code to qualified 7 OEM customers was? 8 Answer: I think it varied from OEM to 9 OEM. Compaq received the DOS source code, 10 MS-DOS source code in order to create a version 11 of MS-DOS called Compaq DOS. And they made a 12 number of changes, that sort of thing, to 13 better fit on their hardware. 14 Question: But that was authorized by 15 Microsoft? 16 Answer: Yes. 17 Question: Do you know of any problems 18 caused by this release of source code to a 19 qualified OEM customer? 20 Answer: Well, I'm not familiar with 21 any problems in that regard. 22 Question: Mr. Barrett, you've been 23 handed Exhibit 44 to your deposition. 24 Do you recognize this as an e-mail 25 message that you sent to a Jancl and others at 11843 1 Microsoft dated July 13, 1988? 2 Answer: Yes. 3 Question: The subject line was Compaq 4 and, quote, sim, S-i-m. 5 What does sim refer to? 6 Answer: Well, sim, as I say in the 7 first line of this e-mail, it's not sim, it's 8 CEMM, pronounced sim, which is Compaq expanded 9 memory manager. And it's a product that Compaq 10 shipped on their machines. 11 Question: And if you can, describe 12 for us in lay terms what did that product do? 13 Answer: Well, what it did was it made 14 on a 386 computer, Intel computer, there's a 15 concept of expanded memory -- well, there's 16 extended memory and expanded memory. 17 Without going into a lot of detail, 18 there's a concept of the first megabyte of 19 memory, which is real mode and accessible to 20 DOS applications. 21 Intel microprocessors had access to 16 22 megabytes of memory, but because the DOS 23 applications could only get to the first 24 megabyte, they, people created a special device 25 driver which allowed the DOS applications to 11844 1 gain access to the other 15 megabytes of 2 memory. That's what that did. 3 Question: All right. Was there any 4 friction or conflict between Microsoft and 5 Compaq with respect to this Compaq EXP memory 6 manager? 7 Answer: Well, there was a certain 8 amount of tension. They were trying to put 9 things into it that particularly us and the 10 operating system group did not like. 11 Question: All right. At the end of 12 the second paragraph of your e-mail you say, 13 the only way we have to control them on that is 14 that they'll probably have to use the knowledge 15 inherent in Windows 386 (instant data switching 16 and DOS mutual exclusion) so if we want to play 17 hardball, ellipse, close quote. 18 What did you mean by playing hardball 19 here? 20 Answer: Well, since it's our ball, we 21 could take it and go home. 22 Basically, what I'm saying here is 23 that we have some information that they don't 24 have access to and that we can use that to 25 control them, basically to get them to do what 11845 1 we want to do or not do what we don't want them 2 to do. 3 Question: All right. And was that a 4 common practice at Microsoft, using the 5 provision or withholding of information as a 6 means of controlling customers or independent 7 software vendors? 8 Answer: Well, I can't say how 9 prevalent or not prevalent it was, but it 10 didn't bother me to use it, wouldn't hesitate 11 to. 12 Question: The last sentence of your 13 e-mail says, until we get a clear indication of 14 Compaq's intent in bundling, then you say 15 (oops, copromotion of) Windows 386, I'd prefer 16 to not put much effort in. 17 First of all, what does the term 18 bundling mean as it was used here? 19 Answer: Bundling simply means, just 20 like it sounds, taking two things and bundling 21 them together. 22 In this case, it's the Compaq hardware 23 and Windows 386. Basically shipping Windows 24 386 with their computers. 25 Question: And earlier in that 11846 1 paragraph, you refer to a per system deal. Do 2 you see that? 3 Answer: Yes. 4 Question: You say, if a multi- 5 processor Win 386 means a per system deal, then 6 we're interested. If it's just a demo thing, 7 then it's plagued. 8 Was the per system deal that you 9 referred to there the means of bundling 10 Microsoft operating system with the Compaq 11 hardware? 12 Answer: Yes, absolutely. 13 Question: All right. Why did you say 14 oops, copromotion after you referred to 15 bundling? 16 Answer: Well, Compaq for some reason 17 had this aversion to the use of the word 18 bundling and so I was poking a little fun at 19 them. They call it copromotion of a product. 20 Question: Was there any aversion to 21 the use of the term bundling at Microsoft? 22 Answer: Not that I know of. I heard 23 it a lot. 24 Question: You've been handed Exhibit 25 47 to your deposition, Mr. Barrett. 11847 1 This is yet another series of e-mail 2 messages provided to us in discovery by 3 Microsoft. 4 I want to refer you to the e-mail 5 message that begins on the first page. 6 There's a -- you'll see that there's a 7 message at the very top of the page and address 8 at the very top of the page. You see what I'm 9 referring to? 10 Answer: Yeah. 11 Question: Do you recognize this as an 12 e-mail message that Brad Silverberg sent to you 13 and others at Microsoft dated April 29, 1991? 14 Answer: Yes, I do. 15 Question: And you'll see that 16 included in Mr. Silverberg's memo are a number 17 of embedded e-mail messages or portions of 18 messages that are signified by those vertical 19 lines that you referred to in the e-mail? 20 Answer: Yes, I do. 21 Question: The first embedded e-mail 22 message, do you recognize that as a message 23 from a Bobgu to Mr. Silverberg, David Cole, to 24 yourself and others at Microsoft dated April 25 29th, 1991? 11848 1 Answer: Yes, I do. 2 Question: All right. And you'll see 3 at the bottom -- first of all, who's Bobgu? 4 Answer: Bobgu, Bob Gunderson. And he 5 was a tech lead or project lead particularly on 6 the area of user.xe. The windowing and menuing 7 system of Windows. 8 Question: In that message, 9 Mr. Gunderson says at the bottom of the page 10 before the heading soap box statement, he says, 11 the question here is when are we going to 12 publish our shell APIs. 13 To date, all the juicy APIs needed to 14 write a shell have been undocumented. 15 I wanted to ask you about that. 16 First of all, what is a shell API? 17 Answer: Well, API, application 18 programming interface. And a shell is a 19 program the user uses to interact with the 20 system. 21 In DOS the command line was the thing 22 that -- the little DOS prompt and you type 23 commands in, that's called the DOS shell. 24 In Windows today there is no really 25 shell. Windows Explorer might be considered a 11849 1 shell. Basically it's a program that allows 2 you to manipulate the system, find files, run 3 other programs, that sort of thing. 4 Question: Was -- were shell APIs 5 important to independent software vendors in 6 the development of their applications? 7 Answer: Some software vendors that 8 were writing shells were writing applications 9 that had shell-like needs. 10 Question: Did you agree with 11 Mr. Gunderson's statement here that to date -- 12 this would be in April 1991 -- all the juicy 13 APIs needed to write a shell had been 14 undocumented? 15 Answer: I'm not sure I'd characterize 16 them as juicy, but, yes, I agree with that. 17 Question: Did Microsoft's 18 applications group have access to undocumented 19 APIs in Windows? 20 Answer: No. 21 Question: Now, Mr. Gunderson, Bob 22 Gunderson goes on to say under the heading soap 23 box statement, this group has been much too lax 24 when it comes to adding API in the product, 25 both external and internal. We don't do any 11850 1 sort of API policing within the group. ISV 2 usability issues are rarely considered. 3 For example, the palette support API 4 is a total mess because the developers doing 5 the work decided what the API would look like 6 without regard to how the ISV would use it. 7 For undocumented APIs, we add them at 8 will without thinking about whether or not they 9 should be documented. They are hacked in ugly 10 things that often eventually get documented. 11 PrestoChangoSelector is one of these. We need 12 to do a better job at this. 13 Let me ask you a number of questions 14 about that paragraph. 15 First of all, what is a palette 16 support API? 17 Answer: A color palette support API 18 basically in a window -- when you want to 19 display lots of colors. 20 Typically -- well, back then the 21 displays that we were using were -- had very 22 limited number of colors you could use. 23 And the way that you got richer colors 24 was through what's called a color palette, 25 which is a way of having a small number of 11851 1 colors, typically 256 color values, that you 2 could map into arbitrarily a large number of 3 colors, typically 16 million. 4 And so each entry in this color 5 palette would have a special color, and so the 6 indexing of the palette, the number from zero 7 to 255 would be that color, then you'd use that 8 as the pixel. That value would be the pixel 9 that will display a single dot of color on the 10 screen. 11 The API they're referring to is how 12 you set up that palette and how you change the 13 values for the actual colors. The one of 16 14 million or whatever it is and how you change -- 15 you might have multiple color palettes and 16 changing them and that sort of thing. That's 17 what that's talking about. 18 Question: All right. Were you 19 familiar with the problem that Mr. Gunderson 20 was talking about here with the palette support 21 API being a total mess? 22 Answer: Well, it's a general call on 23 Bob Gunderson's part, but basically he was 24 correct. 25 There were a lot of -- it was not well 11852 1 thought out. I wouldn't say it was a total 2 mess because it was usable, but it was not well 3 thought out. And it has some performance 4 problems and it created visual abnormalities, 5 visual glitches what they called it, on the 6 display when the color palette was changed or 7 whatever. 8 Question: Is that -- is that 9 something that would cause difficulty for an 10 independent software vendor trying to make use 11 of a color palette? 12 Answer: Absolutely. Absolutely. 13 Question: Did Microsoft's 14 applications group seeking to use the color 15 palette support API have any advantage? 16 Answer: Well, yeah, they went and 17 talked to the guy that created it, the guy that 18 designed it and implemented it. 19 Question: There's a reference here to 20 PrestoChangoSelector. Is that something real 21 or is that a joke? 22 Answer: No. It's real. It's -- yes, 23 it's a real function call. 24 Question: And what was that? 25 Answer: A selector in the 386 family, 11853 1 the 486, the Pentium family of microprocessors, 2 has to do with a very low-level memory 3 addressing construct that -- extremely 4 technical to some of the more advanced parts of 5 the microprocessors, but basically it's a way 6 of creating a window to look into memory 7 anywhere in all of the physical memory of the 8 computer. 9 It's not a very -- it's not a very 10 good name. It's a name that was put in by 11 David Weise, who, frankly, was being tongue in 12 cheek. 13 Question: In response to this e-mail 14 from Bob Gunderson that we were just looking 15 at, I can see at the very top of the first page 16 Mr. Silverberg says, totally agree, Bob. 17 Do you see that? 18 Answer: Yes. 19 Question: Did you agree with 20 Mr. Gunderson and Mr. Silverberg about the 21 points made in Mr. Gunderson's memorandum? 22 Answer: No, I personally did not 23 agree with everything. I agreed with his soap 24 box statement, but I did not agree with the 25 issue of publish the source code and the source 11854 1 code controls and you can no longer enhance 2 them. But anyway. 3 Question: Where is that? 4 Answer: It's call me paranoid. It's 5 the first block of text that Bob Gunderson 6 wrote under the three vertical bars. 7 Question: All right. The paragraph 8 that begins with call me paranoid? 9 Answer: Right. 10 Question: That's the portion of the 11 e-mail that you don't agree with? 12 Answer: Correct. 13 Question: You've been handed Exhibit 14 59 to your deposition. 15 This is a document entitled merged 16 DOS/Windows product and it shows distribution 17 to you among others at Microsoft. 18 Do you see that? 19 Answer: Yes. 20 Question: It's dated April 26, 1991. 21 Can you identify this document? 22 Answer: Yes. The plan for merging 23 DOS and Windows together. It's a plan for 24 merging DOS and Windows together. 25 Question: And do you recall being 11855 1 involved in this plan for merging DOS and 2 Windows at around this time frame, April of 3 1991? 4 Answer: Yes. 5 Question: And what was -- did this 6 project move forward to conclusion? 7 Answer: I believe that we built 8 something. I don't think we actually shipped 9 it, though. I'm trying -- I don't recall the 10 specifics of it, but we did not -- did not 11 receive broad distribution. 12 Question: Mr. Barrett, I have one 13 more exhibit to mark for you. 14 Marked Exhibit 60 has just been handed 15 to you. 16 This is another series of e-mails that 17 has been produced in this litigation. 18 I'd like to direct your attention to 19 the e-mail that begins at the middle of the 20 page. 21 Do you recognize that as an e-mail 22 from you to Brad Silverberg and Tom Lennon 23 dated December 3rd, 1991? 24 Answer: Yes, I do. 25 Question: And the subject line says 11856 1 re: Slick. What was Slick? 2 Answer: Slick was the code name for 3 the integrated DOS and Windows version. 4 Question: And is that the integration 5 merge, merge DOS and Windows product that was 6 referred to in Exhibit 59? 7 Answer: Yes, indeed it was. 8 Question: You say in the middle of 9 the e-mail, the objective of MS-DOS/Windows 10 merge is to combine Windows 3.1 and MS-DOS 5.0 11 (B) into a single package with a seamless 12 install. 13 Did you work on this project? 14 Answer: Yes, I did. 15 Question: And what was the outcome of 16 this project, do you recall now? 17 Answer: Well, I believe that this was 18 advanced to an actual -- it was an actual 19 product or completion of the project. And I 20 don't recall -- I believe that there was -- 21 this product was actually shipped, but I think 22 it was very quickly pulled from the market. 23 It's just a vague recollection on my part at 24 this point. 25 Question: Do you know what it was 11857 1 called -- what the product was called when it 2 was released? 3 Answer: Well, I did when I worked 4 there, but I don't recall it now. 5 Question: All right. Did you have 6 any opinions as to whether merging DOS and 7 Windows into a single integrated product made 8 sense from a technical standpoint? 9 Answer: Well, it didn't make a lot of 10 sence from a technical standpoint. It was 11 mainly a customer simplicity issue, a single 12 install. 13 Question: When you say it didn't make 14 sense from a technical standpoint, why did it 15 not make sense technically to merge DOS and 16 Windows? 17 Answer: Well, there wasn't really any 18 true benefit from the point of view of creating 19 a smaller, more efficient, whatever version of 20 the two products. It's just simply butting 21 them together. Putting them together and 22 having a single install rather than two 23 installs. Very little technical merit to it. 24 Question: Did the merging of MS-DOS 25 and Windows into a single integrated product 11858 1 cause any problems or would it cause any 2 problems for Microsoft? 3 Answer: I can't think of any off the 4 top of my head. 5 Question: Does the size of the merged 6 product cause any problems for Microsoft on the 7 technical -- 8 Answer: Yes. In fact, in that time 9 frame, because of diskettes, or the 10 distribution medium, it was larger but still it 11 didn't -- it just meant that you had two sets 12 of diskettes versus one larger set of 13 diskettes. 14 (Whereupon, playing of the video 15 concluded.) 16 MR. GRALEWSKI: Your Honor, that 17 concludes the first day of Mr. Barrett's 18 deposition. 19 And at this time Plaintiffs offer the 20 continuation of Mr. Barrett's deposition taken 21 on May 31st, 2002. 22 THE COURT: Without objection, you 23 may. 24 (Whereupon, the following video was 25 played to the jury.) 11859 1 Question: Now, I want to start by 2 making sure that we all understand your job 3 history at Microsoft. 4 You were part of the Windows team from 5 1986 through early 1992; is that right? 6 Answer: Yes. I believe that to be 7 correct, yes. 8 Question: And your job for much of 9 that period was development manager? 10 Answer: That's correct. 11 Question: And if I understood you 12 correctly, you contrasted that at your last 13 deposition with the position of program 14 manager? 15 Answer: Yes, I did. 16 Question: And your job related 17 essentially to technical development of the 18 product; is that correct? 19 Answer: A large part of it did, yes. 20 Question: Now, is it correct that 21 until 1991 there was only one development 22 manager on the Windows team? 23 Answer: That's true. 24 Question: And that was you? 25 Answer: That's correct. 11860 1 Question: And then in late 1991 or 2 early 1992 your superiors at Microsoft decided 3 that the development responsibilities for 4 Windows would be essentially split between you 5 and another employee; is that right? 6 Answer: Yes, but that employee worked 7 for me. 8 Question: That was Mack McCauley? 9 Answer: No, no. The window -- 10 well -- 11 Okay, I'll go back. If you use the 12 word Windows in a very, very broad sense, yes. 13 Question: Well, in fact -- 14 Answer: To answer the question 15 before, in a broad sense, yes. 16 Question: So in a broad sense, at 17 some point in late 1999 -- 1991 and 1992, 18 responsibilities for the Windows development, 19 in a broad sense to use your words, were then 20 divided between you and Mr. McCauley? 21 Answer: The Windows development 22 activities, Windows 3.1 development activities 23 and DOS continued under me. I believe the new 24 project was started under Mack McCauley. And I 25 believe it was related to a 32-bit version of 11861 1 DOS initially. 2 Question: Well, in fact, as it played 3 out -- this was shortly before the release of 4 Windows 3.1; is that right, a few months 5 earlier? 6 Answer: Correct. 7 Question: And as it played out, you 8 were given responsibility for finishing Windows 9 3.1 and Mr. McCauley was given responsibility 10 for the subsequent, for what was going to 11 become the subsequent version or 32-bit version 12 of Windows; is that right? 13 Answer: At that time I don't believe 14 that that was the plan. I believe, if I recall 15 correctly, it was initially looking at a 32-bit 16 version of DOS. 17 It may have played out that way after 18 I left, but I don't have any specific knowledge 19 of how it played out after I left. 20 Question: Now, within a few months of 21 Mr. McCauley being given, however you want to 22 define his responsibilities in the Windows 23 group, within a few months after he got that 24 position, you decided to leave the Windows 25 group; is that right? 11862 1 Answer: That's correct. 2 Question: And, in fact, is it true 3 that you ended up leaving the Windows group 4 before Windows 3.1 even got shipped; is that 5 right? 6 Answer: Before the actual release, 7 yes, that's correct. 8 Question: And I believe your 9 testimony was that when you returned to the 10 Windows group, you were working on issues 11 relating to Internet technology? 12 Answer: That's correct. 13 Question: And by August of 1994, 14 about four months later, you had decided that 15 you were going to leave the Windows group; is 16 that right? 17 Answer: Yes. 18 Question: Did you announce in the 19 early fall that you were going to be -- that 20 you were going to be quitting your job? 21 Answer: Yes. 22 Question: Now, in October 1994, you 23 went to work for RealNetworks; is that right? 24 Answer: Yes, that's correct. 25 Question: And I think your testimony 11863 1 was that you're now -- you're an officer of 2 that company? 3 Answer: Yes, I am. 4 Question: And Microsoft is now a 5 direct competitor of your company's? 6 Answer: Give me a minute. Let me 7 understand. 8 There are parts of RealNetworks' 9 business that are competitive with parts of 10 Microsoft's business. 11 Question: Well, in fact, sir, 12 wouldn't you -- don't you consider as an 13 officer of the company, don't you consider 14 Microsoft one of your company's major direct 15 competitors? 16 Answer: I would say there's a 17 competitive, strong competitive nature between 18 the two companies, yes. 19 Question: Now, it's true, isn't it, 20 sir, that during the period 1986 to 1992 you're 21 not aware of Microsoft ever adding code to any 22 of its applications that prevented them from 23 working on DR-DOS, do you? 24 Answer: I'm not aware of any specific 25 outcomes. 11864 1 Question: I want to be sure I 2 understand what that means. 3 Is it true, sir, that you are not 4 aware of any products that got, any application 5 products that got released to the market by 6 Microsoft during your tenure from 1986 to 1992 7 that have had code added to them that prevented 8 them from working on DR-DOS? You're not aware 9 of any products like that? 10 Answer: I am not aware of any 11 specific product releases. 12 Question: And that's for -- 13 Answer: That contained, as you 14 characterized them. 15 Question: And that's true for that 16 entire six-year period? 17 Answer: Yes. 18 Question: And it's also true, sir, 19 isn't it, that during that entire period, 20 Microsoft didn't add any code to Windows 2.0, 21 Windows 3.0 or Windows 3.1 that would prevent 22 -- prevent any of those products from running 23 on DR-DOS? 24 Answer: I'd have to answer yes. 25 Question: Yes, it's true that there 11865 1 were no such products? 2 Answer: That was your question, 3 wasn't it? 4 Question: I just want to make sure we 5 don't have a double negative here. And you're 6 under oath and I want to be clear as possible. 7 The answer is, yes, you were aware of 8 no Windows product that had code added during 9 that six-year period to prevent them from 10 running on DR-DOS? 11 Answer: Yes. 12 Question: Now, you recall testifying 13 -- you recall testifying the other day about 14 the product of a company called DRI? 15 Answer: Yes, I do. 16 Question: And DRI had a product 17 called DR-DOS, which was a clone of Microsoft's 18 DOS product? 19 Answer: Yes. 20 Question: And, in fact, DR-DOS was 21 competing directly against Microsoft's DOS; is 22 that right? 23 Answer: That's correct. 24 Question: Now, do you recall 25 testifying the other day that there was concern 11866 1 that IBM -- that IBM would switch and start 2 using DR-DOS? 3 Answer: Yes, there were concerns. 4 Question: When you were at Microsoft, 5 did you ever learn that IBM had taken steps to 6 evaluate the quality of DR-DOS? 7 Answer: Yeah, I recollect something 8 to that effect, although I don't remember the 9 specifics. 10 Question: Did you learn that IBM had 11 determined that DR-DOS had compatibility 12 problems? 13 Answer: I believe that to be the 14 case. 15 Question: Well, is it true that 16 DR-DOS itself, it wasn't a graphical system. 17 It didn't generate icons, did it, or pictures? 18 Answer: That's true. And I'll answer 19 your question based on that. DR-DOS has -- had 20 a character-based user interface. It was not a 21 graphical system. 22 Question: And is it also true that in 23 order to make the DR-DOS product, in order for 24 it to generate graphical -- generate the 25 pictures that we now associate with the 11867 1 computer, a consumer would have to buy a second 2 product? 3 Let me rephrase that. 4 Is it true that DR-DOS in order -- in 5 order to create a graphical display required 6 the operation of a second product with it? 7 Answer: Yes, that's true. 8 Question: Is it true that by the 9 early 1990s, was it your understanding in the 10 early 1990s that there was a huge movement in 11 the marketplace toward graphical user 12 interfaces? 13 Answer: Yes, that was true back then. 14 Question: In fact, by 1990 Apple 15 already had an operating system that included 16 graphical user interfaces; is that right, with 17 their operating system? 18 Answer: Yes. 19 Question: And, in fact, IBM was doing 20 the same thing with OS/2; is that right? 21 Answer: IBM and Microsoft together, 22 yes. 23 Question: Now, did you ever learn in 24 or about 1990 that DRI was exploring options 25 for developing graphical user interface 11868 1 functionality for its own products? 2 Answer: I'm sorry, it doesn't ring a 3 bell with me. 4 Question: Do you know -- let me ask 5 it this way. 6 Did DR-DOS release any kind of 7 graphical user interface functionality in any 8 of its products in the 1990 to 1992 time 9 period? Did it achieve that functionality? 10 Answer: I'm sorry, I just don't 11 recall. It's possible they did, but I don't 12 recall. 13 Question: Do you have any 14 recollection -- and I don't want you to 15 speculate. Do you have any recollection of DRI 16 or its successor Novell Corporation after it 17 purchased DRI, ever achieving that 18 functionality in their products while you were 19 still at Microsoft? 20 Answer: No, I don't. 21 Question: The other day you discussed 22 Microsoft's release of beta versions of its 23 Windows products. 24 Do you remember that? 25 Answer: Yes. 11869 1 Question: Now, to put this in context 2 a little bit, would you agree that the release 3 of beta versions is a standard practice for 4 companies that sell operating systems? 5 Answer: Yes. 6 Question: And, in fact, other 7 operating system vendors like IBM, Sun, and for 8 that matter DRI, these companies all released 9 beta versions of their products, too, didn't 10 they? 11 Answer: True. 12 Question: And is it true that a beta 13 release of a product is, in effect, an advanced 14 version of the product that's being sent to 15 some selected group of viewers or testers? 16 Answer: That's a fair 17 characterization. 18 Question: And this is all done before 19 the product is actually released into the 20 general marketplace for consumers to buy? 21 Answer: Yes. 22 Question: So it's an advanced look -- 23 in a nutshell, it's an advanced look at the 24 product? 25 Answer: Sure. 11870 1 Question: Now, you talked the other 2 day about the various products that DRI 3 released. I want to ask you about that. 4 Do you know, did DRI ever provide an 5 advanced beta of DR-DOS 5.0 to Microsoft before 6 the DR-DOS product was released to the market? 7 Answer: I'm not aware of it. 8 Question: Did DRI provide an advanced 9 beta of version 6.0 of its product to Microsoft 10 before DR-DOS version 6.0 had actually been 11 released to the market? 12 Answer: I'm not aware of DR providing 13 that, no. 14 Question: What about Novell -- after 15 Novell acquired DRI and released DOS version, 16 Novell DOS Version 7.0, did it give Microsoft 17 an advanced beta of that product? 18 Answer: I just don't recall. 19 Question: Now, is it true that during 20 the period 1989 -- you have no recollection of 21 that happening? 22 Answer: I have no recollection of DR 23 providing a copy of -- or its successors 24 providing a copy of operating system betas to 25 Microsoft. 11871 1 Question: Let me ask the question 2 more generally. 3 Are you aware of any instance in which 4 DRI ever gave one of their betas to Microsoft? 5 Answer: No, I'm not aware of DR doing 6 that. 7 Question: Now, Mr. Barrett, it's 8 true, isn't it, that DRI actually did have a 9 copy of the beta for Windows version 3.0; is 10 that right? 11 Answer: I believe that to be the 12 case. I have no direct knowledge, though. 13 Question: Did you also learn while 14 you were at Microsoft that DRI had obtained 15 access to a version of the beta for Windows 16 3.1? 17 Answer: I believe that to be the 18 case. 19 Question: Now, I believe your 20 testimony, and correct me if I'm wrong, was 21 that in the summer of 1990, Microsoft was 22 starting to provide some information about the 23 next version of DOS that it was going to try to 24 get out into the marketplace; is that right? 25 Answer: That sounds correct. 11872 1 Question: In fact, by the summer of 2 1990 Microsoft had actually released a beta 3 version of DOS version 5.0; is that right? 4 Answer: I think that's right, yes. 5 Question: And is it also true that 6 providing advanced notice about software 7 releases, i.e., providing notice before the 8 product actually gets released into the 9 marketplace, was a standard industry practice 10 during this period? 11 Answer: I don't think that it was 12 universal. I think it was one of several 13 different practices that were followed, some of 14 which included no knowledge at all, no 15 information at all, some of which included 16 advanced knowledge. 17 Question: Is it true, though, that a 18 lot of software companies, a lot of major 19 software companies during this time frame 1989 20 to 1993 would start telling the marketplace 21 about what the next version of their product 22 was going to be like well in advance of that 23 product actually being released? 24 Answer: It happened, sure. 25 Question: Let's talk about some 11873 1 specifics. 2 Do you remember Lotus Development 3 Corporation? 4 Answer: Yeah, I recall them. 5 Question: They were the leading 6 seller of spreadsheets in the late 1980s and 7 early 1990s? 8 Answer: Yes. 9 Question: Do you recall that Lotus 10 Corporation announced work on version 3 of its 11 DOS spreadsheet more than two years before that 12 product was released to the market? 13 Answer: Well, I can't -- I can't 14 testify to the specific time periods. I 15 believe that it was -- that they did discuss it 16 publicly well ahead of its release. 17 Question: Is it true that DRI also 18 provided advanced information about its product 19 releases before they actually were released to 20 the market? 21 Answer: Well, yes. Sure. 22 Question: Now, let's talk about IBM. 23 You remember the other day counsel 24 asked you a lot of questions about the 25 functionality that was going to be in IBM's 11874 1 product OS/2 version 2.0? 2 Answer: Uh-huh. 3 Question: You got to give me a yes or 4 no. 5 Answer: I'm sorry, yes. 6 Question: And, in fact, you were 7 shown a stream of e-mail about -- and you can 8 look at them if you want to. I think they're 9 Exhibits 51 through 53 in the earlier session 10 of your deposition. 11 You were shown a stream of e-mail 12 discussing that public representations IBM was 13 making about OS/2 version 2.0. 14 Do you recall that? 15 Answer: Yes. 16 Question: And, in fact, that public 17 discussion by IBM was all going on by April of 18 1991; is that right? 19 And feel free to review the exhibits, 20 if it will help you. 21 Answer: I think that's right timing, 22 sure. 23 Question: So IBM was out making 24 public statements about OS/2 version 2.0 by the 25 spring of 1991? 11875 1 Answer: Yes. 2 Question: And, in fact, IBM didn't 3 actually release OS/2 version 2.0 to the 4 general marketplace until April of 1992; is 5 that right? 6 Answer: Yes, that's correct. 7 Question: In fact, it was on April 8 1st of 1992. Does that ring a bell? 9 Answer: Oh, yes. 10 Question: So when IBM was out touting 11 the virtues of OS/2 2.0, we're basically -- 12 they're basically doing that a year in advance 13 of their actual release of that product? 14 Answer: Yeah. Yes. 15 Question: Now, let's talk about what 16 we agreed. We talked about in a short time 17 before, we referred to as the AARD code. Let 18 me ask you a few questions about this. 19 How many beta releases were there of 20 Windows 3.0, do you recall? 21 Answer: I believe there were two, two 22 major ones. There may have been a minor 23 release, a quick test of some final small 24 changes. 25 Question: And there were three more 11876 1 beta releases, at least three more beta 2 releases of Windows 3.1? 3 Answer: What was your previous 4 question, I'm sorry? 5 Question: I previously asked how many 6 beta releases there were for Windows version 7 3.0. 8 Answer: Okay, yes. I stand by that. 9 I believe a similar, in a similar vein 10 for Windows 3.1 there were two major and I 11 believe a minor, what we jokingly called a 12 gamma release. 13 Question: All right. So between 14 those two products, Windows 3.0 and 3.1, there 15 were about six beta releases? 16 Answer: That sounds correct. 17 Question: And now during the period 18 1986 to 1992, how many versions of Windows -- 19 putting aside beta releases, how many versions 20 of the Windows products were released while you 21 were part of the Windows team? 22 Answer: I'm sorry. What was the time 23 frame? 24 Question: From the time you arrived 25 at Windows in the Windows group in 1986 till 11877 1 1992. 2 Answer: Give me a minute. 3 Oh, probably five I think is the right 4 number. 5 Question: So to put this in 6 perspective a little bit, between 1986 and 1992 7 while you were on the Windows team, there were 8 in either beta form or an actual product 9 releases, there were about 11 releases, at 10 least 11 releases of the Windows product? 11 I guess, five, including five actual 12 product releases and at least six betas? 13 Answer: That's fair. There were more 14 betas obviously, but -- 15 Question: Now, and as I understand 16 it, what we've described as the AARD code, that 17 was included -- of these 11 releases, the AARD 18 code was included in one beta release? 19 Answer: The AARD code was included in 20 one beta release. 21 Question: And that was in December of 22 1991? 23 Answer: That sounds correct. 24 Question: Now, as I understand it, 25 the AARD code had -- it had some detection code 11878 1 in it and then a message associated with that; 2 is that correct? 3 Answer: Yes. 4 Question: And you testified the other 5 day that what it would do is it would detect 6 the presence of DR-DOS. And that's not quite 7 right, is it? 8 Answer: It would -- okay. I'll be 9 precise. Perhaps I misstated. 10 Question: That's fine. 11 Answer: It would test for the absence 12 of MS-DOS. 13 Question: Well, even more 14 specifically, wouldn't it -- it would test for 15 the presence of MS-DOS, and if that test 16 indicated -- if it failed to indicate that 17 MS-DOS was present, then you would infer that 18 something else was present. Is that a more 19 accurate -- 20 Answer: I believe that's a tautology 21 with respect to my previous statement. 22 Question: Yeah, I'm not trying to 23 trick you. 24 I'm just trying to be as clear as we 25 can. 11879 1 Now, and you testified a little bit in 2 response to Mr. Grossman's question about what 3 the notification, what kind of a notification 4 would be associated with this. 5 Answer: Yes. 6 Question: Now, in fact, the 7 notification that was included in this one 8 beta, all it said was, it basically said a 9 nonfatal error had been detected; is that 10 right? 11 Answer: That sounds correct. Non- 12 fatal was a word that came up. I don't recall 13 the specific -- as I testified earlier, I 14 couldn't tell you the specific final form of 15 the notification. 16 Question: And, in fact, a nonfatal 17 error message is, by definition, that's an 18 error that's not going to cause the computer 19 program to stop running; is that right? 20 Answer: That's true. 21 Question: Now, sir, would you agree 22 that the purpose of adding this AARD code was 23 not to hinder DR-DOS? 24 Answer: No, I would not agree with 25 that. 11880 1 Question: That would be an untrue 2 statement? 3 Answer: Your statement I believe to 4 be untrue, yes. 5 Question: All right. Let me, sir, if 6 you would turn, then, to your FTC deposition at 7 page 41. 8 Or, I'm sorry, I'm giving you the 9 wrong page number. I apologize. 10 If you would turn to the Caldera 11 deposition at page 196. 12 You see, sir, when you testified -- 13 let's break it down this way. 14 If you look at the cover page to your 15 deposition in the Caldera case, if you look at 16 the first page. 17 Answer: Okay. 18 Question: Do you see it's dated July 19 31st, 1997? 20 Answer: Okay. 21 Question: That was a little closer to 22 these events than your testimony today; is that 23 right? 24 Answer: Yes. 25 Question: And when you testified 11881 1 under oath in the Caldera case, you were not a 2 Microsoft employee, correct? 3 Answer: No, I wasn't. 4 Question: You were already working 5 for RealNetworks? 6 Answer: Correct. 7 Question: Now, on page 196 of your 8 deposition, you're asked -- do you see where 9 you're asked at line 5, what was the purpose of 10 saying, of having the nonfatal error message in 11 this beta release? 12 Answer, to find out if people were 13 running our beta on top of DR-DOS. 14 Question, was it to hinder DR-DOS in 15 any way? 16 Answer, I don't think so. I don't 17 think it was. 18 Answer: Okay. 19 Question: Did I quote your testimony 20 correctly, sir? 21 Answer: You read it correctly. 22 Question: Was that what you testified 23 under oath in the Caldera case? 24 Answer: I'm sure this is correct. 25 Question: Now, let me ask you, sir, 11882 1 it's true, isn't it -- I think it's what you 2 testified to the other day -- that after this 3 had been included, this language had been 4 included in the one beta release, that the idea 5 of whether it should stay in the Microsoft 6 product was then presented to Mr. Ballmer; is 7 that right? 8 Answer: Yes, that runs in my mind. 9 Question: And Mr. Ballmer basically 10 said, this is a stupid idea, we should just get 11 rid of it? 12 Answer: I don't recall his specific 13 words, but I believe that that was his 14 sentiment. 15 Question: And his sentiment was, it's 16 a bad idea, remove it from the product? 17 Answer: It's another way of saying 18 it. I think that's reasonably consistent with 19 my recollection. 20 Question: And after its inclusion in 21 that one beta release, to your knowledge, the 22 AARD code was removed from the product? 23 Answer: No, the AARD code was not 24 removed from the product. It was stubbed out 25 as I believe has been discussed in this 11883 1 deposition. 2 Question: But the effect of that was 3 that this notification, this nonfatal error 4 message never showed up in the final product, 5 did it? 6 Answer: That's true. 7 Question: Now, do you remember 8 testifying about Bambi the other day? 9 Answer: Yes, I do. 10 Question: And I want to start by 11 clearing up any confusion. 12 You initially testified that you 13 thought Bambi was a code name for DOS version 14 6. 15 Answer: It wasn't. 16 Question: And then you clarified that 17 later, right? 18 Answer: Yes. I was incorrect. I'd 19 forgotten the code name Bambi. 20 Question: All right. And to be as 21 clear -- to be as clear as we can here, the 22 testimony that Bambi was a code name for DOS 6 23 was just a mistake. 24 What Bambi was actually a code name 25 for was SmartDrive? 11884 1 Answer: Yes, that's absolutely 2 correct. 3 Question: And I believe you testified 4 that SmartDrive was a disk cache device driver? 5 Answer: SmartDrive is a disk, is and 6 was a disk cache. I'm not sure I would 7 characterize it as a driver, but -- 8 Question: All right. And Bambi was 9 the code name for the beta version of that 10 product, of that -- actually, of that function 11 amount? 12 Answer: Yes. 13 Question: And would you agree that in 14 the fall of 1989, actually at some point in 15 September of 1989 -- strike that. 16 At some point in September of 1991, 17 Microsoft discovered that DRI's product had an 18 incompatibility when it was run with Bambi? 19 Answer: Yes. There were 20 incompatibilities between the two products. 21 Question: And, in fact, the belief 22 was that if you tried to run Bambi or 23 SmartDrive on DR-DOS, it could have really 24 serious consequences for the user; is that 25 right? 11885 1 Answer: There were -- there were -- I 2 don't recall the specific problems that would 3 occur, but they were considered to be 4 significant problems. 5 Question: And could it, in fact, 6 potentially that combination, do serious damage 7 to a user as to the actual disk storage on a 8 computer? 9 Answer: I don't believe it could 10 actually physically damage the hard drive, if 11 that's what you're asking. 12 Question: No. That's not what I'm 13 asking. The information on the hard drive. 14 Answer: Yes, it's possible to corrupt 15 the data on the hard drive. 16 Question: Now, to the best of your 17 knowledge, this wasn't a problem that anyone 18 had intentionally created? 19 Answer: I believe that to be true. 20 Question: And, in fact -- but, in 21 fact, it's a problem that arose in the 22 interaction of these two products and Microsoft 23 discovered it; is that right? 24 Answer: I don't recall the genesis of 25 the discovery of the problem, but it was 11886 1 discovered. It was known at Microsoft. 2 Question: Do you remember -- I mean, 3 you don't remember anyone at DRI calling and 4 saying, gee, we have this problem with 5 SmartDrive, you need to clean it up. 6 You don't remember that either, do 7 you? 8 Answer: No, I don't. 9 Question: Mr. Barrett, I want to just 10 return with you for one moment before we talk 11 some more about Bambi about your testimony 12 earlier today. 13 You remember I asked you some 14 questions about whether it would be helpful to 15 the Microsoft people that were involved in bug 16 shooting and the like to know what kind of 17 operating system that their beta testers were 18 using? Do you remember some discussion about 19 that? 20 Answer: Yes, I recall. 21 Question: And you basically said -- 22 and tell me if I'm characterizing this 23 incorrectly -- that wasn't really information 24 that the AARD code -- it wasn't useful for that 25 purpose? 11887 1 Answer: That's correct, knowing -- 2 well, no, actually, I believe the 3 characterization was, was the nonfatal error 4 message going to be helpful. And I said no, 5 it's not going to be. 6 Question: Well, that message would 7 let your people, the Windows people who were 8 working with the beta testers, it would let 9 them know that something had been run on a 10 non-Microsoft operating system; is that right? 11 Answer: Yes, that's correct. Yes. 12 Question: Now, I want to ask you 13 about some of your testimony to the government 14 in FTC proceeding back in 1993. 15 If you would turn to page 41 of your 16 FTC testimony. 17 Answer: Okay. 18 Question: Now, if you look at page 19 21, you were asked by the government, you were 20 asking -- in this connection with the AARD 21 code, you were asked, you were trying to make 22 an assessment of what your beta testers were 23 running? 24 Answer, yes. 25 Question, why would that be useful? 11888 1 Answer, well, to know the extent if 2 there are bugs reported and other bugs and they 3 also report a nonfatal error message, it would 4 give us a clue as to the source of the problem. 5 Question, do you know if beta testers 6 were normally asked what type of operating 7 system they were running when they reported 8 bugs? 9 Yeah. 10 Question, so why would you need to 11 have the error message pop up? 12 Answer, because we inevitably got DOS, 13 we're just running DOS. They didn't say they 14 were running a version. They didn't 15 necessarily say what they were running. 16 Did I quote your testimony accurately? 17 Answer: Yes. 18 Question: And was that testimony 19 accurate? 20 Answer: Yes. 21 Question: Now, at the bottom of page 22 42 at page 21 you were asked, why would a beta 23 tester want to withhold that type of 24 information? 25 And the answer was, they weren't 11889 1 withholding. They were just being inexact. 2 Beta testers are notorious for that. They tell 3 us everything but the information needed to fix 4 the bug. 5 Did I again quote your testimony 6 accurately? 7 Answer: Yes, you did. 8 Question: And was that testimony also 9 accurate? 10 Answer: Yes. 11 Question: In fact, Mr. Grossman, 12 Plaintiff's counsel, showed you an e-mail 13 discussing whether Windows 3.1 should be made 14 incompatible with other DOS products so that it 15 would crash. 16 Do you remember that? 17 Answer: I don't recall the number, 18 but I do recall the mail, yes. 19 Question: And is it true, sir, that 20 that, in fact, was never done? 21 Answer: To my knowledge, it was never 22 done. 23 Question: And Mr. Grossman also 24 discussed with you the possibility of designing 25 Windows enhanced mode so that it would only run 11890 1 on Microsoft DOS. 2 Do you remember that? 3 Answer: Yes, I recall that. 4 Question: And is it also true that 5 that's something that was never done? 6 Answer: It was not done in 3.1 time 7 frame. 8 Question: You don't have any 9 recollection of that being done, do you? 10 Answer: No. 11 Question: Now, we talked earlier 12 briefly about the various claims that IBM was 13 making about OS/2 version 2.0. And you recall 14 that IBM started making these claims in March 15 or April of 1991? 16 Answer: Like I said earlier, yeah, 17 that sounds right. 18 Question: And is it also your 19 recollection that IBM was making those claims 20 in about as public a manner as it could? 21 Answer: That's quite true. 22 Question: And what's your basis for 23 that, for that answer? 24 Answer: Well, they touted better 25 Windows than Windows, better DOS than DOS quite 11891 1 far and wide in the press. 2 Question: And they were holding big 3 demonstrations, public demonstrations and 4 making the same claims about the 2.0 product? 5 Answer: Yeah, yes. 6 Question: And how do you know that? 7 Answer: It's hard to avoid. It was 8 in the press. 9 Question: It's all over the place? 10 Answer: Yes. 11 Question: Now, I think you said one 12 of the claims IBM was making about this product 13 it was going to be releasing in another year 14 was that OS/2 version 2.0 was a better DOS than 15 DOS? 16 Answer: Yes, that's correct. 17 Question: And was it your 18 understanding that what IBM was claiming was 19 that DOS applications would run better on OS/2 20 than they actually ran on DOS? 21 Answer: That's correct. 22 Question: And IBM was also publicly 23 making claims that this product it was going to 24 release in another year would be a better 25 Windows than Windows? 11892 1 Answer: Yes, they made that claim. 2 Question: And, again, that was a 3 claim that was being made way before OS/2 2.0 4 was generally available for customers to 5 actually purchase? 6 Answer: True. 7 Question: Do you recall being asked 8 the other day by Mr. Saveri about an 9 application that Microsoft developed called 10 Terminator? 11 Answer: Yes, I recall the question. 12 Question: And the purpose of this 13 application was to demonstrate that OS/2, or to 14 try to demonstrate that OS/2 was not a 15 crash-proof operating system; is that right? 16 Answer: Yes. 17 Question: And this is an application 18 or some code that Microsoft was using for its 19 own demonstrations as opposed to actually 20 selling on the marketplace; is that right? 21 Answer: That's correct. 22 Question: And it was being used 23 essentially for the purpose of rebutting 24 certain claims that IBM was making about this 25 product it was going to be releasing down the 11893 1 road? 2 Answer: Yes. 3 Question: Now, in fact, is it also 4 true that since March or April of 1991, Lee 5 Reiswig of IBM had been doing public 6 demonstrations to show, to demonstrate how 7 Windows could be crashed by applications? 8 Answer: Yes, he was. 9 Question: And the press was attending 10 those demonstrations? 11 Answer: Yes. 12 Question: And customers were 13 attending those demonstrations? 14 Answer: I have no idea. 15 Question: But, in any event, 16 Mr. Reiswig was doing all this again about a 17 year before IBM had actually released its 18 product to the marketplace? 19 Answer: I think I said that before, 20 but, yes. 21 Question: And is it also true that 22 Mr. Reiswig started doing those demonstrations 23 before Microsoft began doing its rebuttal 24 demonstrations with this application? 25 Answer: Yes, that's true. 11894 1 Question: Now, is it also true that 2 you informed your boss, Mr. Silverberg, that 3 the claims IBM was making about OS/2 version 4 2.0 were false? 5 Answer: I believe I said that, yes. 6 Question: And among other things, you 7 informed Mr. Silverberg that OS/2, in fact, had 8 serious performance problems running DOS 9 applications; is that right? 10 Answer: Yes. 11 Question: In fact, when you were 12 deposed before the Federal Trade Commission, 13 you indicated that it was your belief that 14 IBM's claim that OS/2 2.0 was a better DOS than 15 DOS, that that claim was false? 16 Do you see, sir, where you're asked on 17 page 19, question, do you recall that you told 18 him it's referencing Mr. Silverberg, about 19 IBM's claim about a better DOS than DOS? 20 Answer, not specifically. 21 Question, can you recall what you 22 generally told him? 23 Answer, yeah, that it was false. 24 Question, why did you believe it was 25 false? 11895 1 Answer, particularly in the 2 performance area, they had serious performance 3 problems. 4 Did I quote your testimony accurately? 5 Answer: I think that's fair. 6 Question: And is it, in fact, correct 7 that you informed your boss, Mr. Silverberg, 8 that OS/2 2.0 had serious performance problems? 9 Answer: I think I -- generally, I 10 told him, yeah. Yes. 11 Question: Now, had you also concluded 12 while you were in the Windows group that IBM's 13 public claim that OS/2 was a better Windows 14 than Windows was not true? 15 Answer: I honestly don't recall the 16 specifics. I am certain I believed that to be 17 the case when I was employed as the development 18 manager. 19 Question: And when you say you 20 believe that to be the case, what you're saying 21 you believe it to be the case, that the claim 22 IBM was making that OS/2 2.0 was a better 23 Windows than Windows was a false claim? 24 Answer: Not wanting to basically 25 nitpick on this, but I believe that at the time 11896 1 I believed that to be true. 2 Question: All right. And was it also 3 your belief while you were in the Windows group 4 that the performance that OS/2 2.0 was 5 significantly slower than Windows at running 6 Windows applications? 7 Answer: I'm sorry. I'm having 8 trouble recollecting the specifics. 9 Question: Fair enough, sir. 10 If you would turn to page 20 of your 11 FTC deposition. 12 Answer: Okay. 13 Question: Question, very bottom, 14 okay. I think you also indicated that IBM made 15 a claim that it was a better Windows than 16 Windows; is that correct? 17 Answer, yes, I said that. 18 And there's some objections. There's 19 a question asked at line 7. 20 What type of -- did you believe that 21 claim was true? 22 Answer, no. 23 Question, why didn't you believe it 24 was true? 25 Answer, well, let's see. Particularly 11897 1 in the performance area, their performance of 2 Windows applications was dramatically slower. 3 Have I quoted your testimony 4 accurately? 5 Answer: You've read it accurately. 6 Question: Was that testimony accurate 7 to the best of your ability, to the best of 8 your knowledge in 1993? 9 Answer: Boy, I'm having a hard time 10 recalling the specifics, but it must have been 11 true if it's here. 12 Question: And you then testified, 13 again, talking about IBM and OS/2, you then 14 testified -- and I'm at line 13 -- quote, they 15 claimed that they were more crash resistant, 16 which really wasn't true. 17 Again, did I quote your testimony 18 accurately? 19 Answer: Yes. 20 Question: And did you, in fact, 21 believe that when you testified to the 22 government? 23 Answer: Sure. Yes. 24 Question: Now, at page 22, and 25 there's a discussion there about IBM's claim 11898 1 that OS/2 would run seamlessly -- it's not at 2 page 22 -- at line 22 on page 21. 3 You were asked, question, you 4 indicated that you didn't believe that claim, 5 claim about being seamless, was true; is that 6 correct? 7 Answer, that claim was patently false. 8 Did I again quote your testimony 9 correctly? 10 Answer: Yes. 11 Question: And what did you mean when 12 you said that IBM's claim about OS/2 operating 13 seamlessly was patently false? 14 Answer: Well, I think this document 15 speaks for itself. 16 The printing was a big issue, I do 17 recall that. 18 Yeah, that's a very good example of 19 where it wasn't seamless because the consumer 20 sees a printer as a printer and yet OS/2 saw it 21 differently from the way Windows saw it. 22 So that's a good example of not 23 seamless, of not being seamless. 24 Question: Do you recall Plaintiff's 25 counsel asking you questions about early 11899 1 efforts to merge DOS and Windows into a single 2 product? 3 Answer: Yes, I recall. 4 Question: And that's a product that 5 you had very limited involvement in; is that 6 right? 7 Answer: I had mostly limited 8 involvement, although I did make several trips 9 to Florida. 10 Question: Did you put any significant 11 amount of time into that project? 12 Answer: Personally, no. 13 Question: And, in fact, is it true 14 that as you sit here today, you can't recall 15 whether a product was even shipped by Microsoft 16 while you were employed at Microsoft in which 17 that merger had taken place? 18 Answer: I did not recall at the time. 19 I still don't. 20 Question: You can't recall any 21 product being shipped like that? 22 Answer: Same answer before. I don't 23 recall. 24 Question: And the Windows 95 product, 25 that was shipped about a year after you left 11900 1 Microsoft? 2 Answer: Approximately. 3 Question: Now, is it true that your 4 personal view back in 1991 was that creating a 5 single product was a good idea for customers? 6 Answer: I believed it had some 7 advantages to the consumer. 8 Question: I'm sorry. Had you 9 finished? 10 Answer: I believe it had some 11 advantages to the consumer. 12 Question: And what were those 13 advantages? 14 Answer: A single installation. 15 Question: Why was that advantageous 16 to the consumer? 17 Answer: One less thing for them to 18 do. One versus two. 19 Question: Were installations also a 20 common source of -- well, strike that. 21 Now, in the early 1990s, one of the 22 alternatives for a consumer who wanted to 23 purchase a GUI, a graphical user interface, was 24 to purchase an Apple computer; is that true? 25 Answer: Yes, that's true. 11901 1 Question: And Apple's computers and 2 its operating system came with -- it came with 3 a GUI interface? 4 Answer: That's correct. 5 Question: And is it true that by the 6 early 1990s, indeed even earlier, that Apple 7 was combining its graphical user interface and 8 its operating system in a single product? 9 Was the Apple operating system and GUI 10 a single product or more than one product? 11 Answer: It was a single product. 12 Question: Would you agree that in the 13 early 1990s the most significant competitor of 14 Windows was OS/2? 15 Let me ask the question differently, 16 sir. 17 When you were in the Windows group, 18 who did you consider the most significant 19 competitor to the Windows graphical user 20 interface product? 21 Answer: I would say OS/2. 22 Question: Why? 23 Answer: Because they had the most 24 resources. Because the company had the most 25 resources behind it. 11902 1 Question: And in competing against 2 Windows, was IBM -- IBM's competition through 3 OS/2, was OS/2 being offered as a single 4 product or as separate products with the 5 separate operating system and a separate 6 graphical user interface? 7 Answer: It was being offered as a 8 single product. 9 Question: So by 1991, that was the 10 approach being taken by Windows' major 11 competitor? 12 Answer: Yes. 13 Question: And did that, to the best 14 of your knowledge, did that continue to be true 15 until the day you left Microsoft? 16 Answer: Yes. 17 Question: Okay, Mr. Barrett. 18 Let's begin by trying to put Windows 19 in a historical perspective a bit. 20 Microsoft had started releasing 21 Windows, releasing its first version of Windows 22 in 1985? 23 Answer: I believe that's correct. I 24 don't recall the specific timing. 25 Question: And between roughly 1985 11903 1 and the end of 1989, there were several 2 versions of the Windows product released? 3 Answer: That's correct. 4 Question: And is it also true that 5 those products during that time period had only 6 limited success in the marketplace? 7 Answer: Yes, that's correct. 8 Question: In fact, is it true, sir, 9 that during that period Windows, if you take 10 your figure of something less than 2 million 11 copies of Windows being sold -- was it true 12 that during the period '86 to '89, Windows was 13 being run on less than 10 percent of PCs in the 14 U.S.? 15 Answer: Significantly less than ten 16 percent. 17 Question: Significantly less than ten 18 percent of PCs? 19 Answer: Yes. 20 Question: What's the basis for that 21 answer? 22 Answer: I just know that we had 23 minuscule market share. I couldn't tell you 24 the numbers, but I know it was less than ten 25 percent. 11904 1 Question: So during that period, most 2 consumers were still opting for character-based 3 operating systems? 4 Answer: Correct. 5 Question: Is it also true that during 6 that period of time, Windows was an evolving 7 product? 8 Answer: That's true. 9 Question: Would it be fair to say 10 that as the 1980s closed out, in early 1990 11 Windows, was still something of a work in 12 progress? 13 Go ahead. 14 Answer: Sure, yes. 15 Question: Sir, isn't it true that 16 some APIs may not get documented because they 17 may not be maintained in the future? 18 Answer: That's certainly a reason for 19 not documenting. 20 Question: And in some cases they may 21 not be documented because they're not 22 sufficiently stable? 23 Answer: I guess. 24 Question: And, in fact, in those 25 circumstances, you wouldn't want applications 11905 1 to be relying on those, would you? 2 Answer: Well, certainly in the narrow 3 sense, I would agree. 4 Question: Now, is it true that as the 5 Windows product evolved during the 1980s, the 6 number of documented APIs was growing 7 substantially, was increasing substantially? 8 Answer: That's true. 9 Question: Now, by the time Windows 10 3.0 was released in May of 1990, there were 11 about 600 documented APIs in the product; is 12 that right? 13 Answer: Yeah, I think 660 something 14 is what sticks in my mind. 15 Question: That was information -- 16 would you agree that was a huge increase over 17 the number of APIs that had been documented in 18 earlier versions of Windows? 19 Answer: True. 20 Question: And, in part, this is 21 because more and more functionality was being 22 added to the product; is that right? 23 Answer: That's correct. 24 Question: And it's also -- and in 25 addition to that, you also as further versions 11906 1 came out, you also had more time to provide 2 documentation for the code that was in the 3 product; is that right? 4 Answer: I'm not 100 percent sure. 5 We're always resource constrained in 6 documentation. 7 Question: When you say resource 8 constrained, what do you mean? 9 Answer: People, not enough people to 10 do the work. Not time, the people. 11 Question: I see. In fact, that was 12 -- that kind of resource constraint was a 13 problem for the Windows team throughout the 14 1980s. Wouldn't you agree? 15 Answer: Yes, it was. 16 Question: In fact, is it true that 17 even by 1989 the systems group at Microsoft was 18 devoting about four or five times as many 19 people to OS/2 as were being devoted to the 20 Windows product? 21 Answer: I think that's -- it's 22 approximately correct. I couldn't tell you the 23 exact number. 24 Question: Let me show you what I've 25 marked, sir, as Exhibit 72, which is a 11907 1 Microsoft document, which has been Bates 2 stamped 159975. 3 In any event, Exhibit 72 is Bates 4 stamped X 159975. It's a Microsoft document 5 dated July 17th, 1989. 6 Sir, referring you to the second page 7 of the document. Do you see where it provides 8 -- 9 Answer: Systems division rollup? 10 Question: Exactly. 11 And you see in the last column where 12 it provides the head count for OS/2 and the 13 head count for Windows? 14 Answer: Okay. 15 Question: And it indicates that in 16 July of 1989, or it indicates at the time of 17 this document anyway, the average head count 18 for the OS/2 team was about 315 people and the 19 Windows team was at about 71? 20 Answer: That's what the document 21 says. 22 Question: And is that consistent with 23 your recollection of the relative resources 24 that were being devoted to these two teams in 25 1989? 11908 1 Answer: Sure. Doesn't stand out as 2 being wrong. I couldn't tell you if it was 3 right, though. 4 Question: But, in any event, is it 5 true that your recollection is that vastly 6 greater resources were being devoted to OS/2 7 than to Windows? 8 Answer: I'm not sure I'd characterize 9 it as vast, although at the time I may have 10 thought that, but it's significantly more. 11 Question: Now, Plaintiff's counsel 12 discussed the sharing of source code with you. 13 Do you recall that discussion? 14 Answer: Yes, I do. 15 Question: And do you recall 16 Mr. Grossman showed you an e-mail from 17 Mr. Slivka discussing that issue? 18 Answer: Yes, the big long one. 19 Question: But you weren't shown your 20 response to that e-mail; is that correct? 21 Answer: Don't recall. 22 Question: And, in fact, after you got 23 Mr. Slivka's e-mail, is it true that your view 24 was that from a technical perspective, making 25 source code available to ISVs was a really bad 11909 1 idea? 2 Answer: Yeah, I probably believed 3 that then. 4 Question: Let me show you, sir, what 5 I have marked as Exhibit 73. 6 Do you want me to pass these down to 7 you? I've got two extra copies. 8 Now, Exhibit 73 is an e-mail that is 9 dated X 575844. 10 You see the e-mail on the first page 11 of that document where you're responding to 12 Mr. Slivka's e-mail? 13 Answer: Yes, I do. 14 Question: And is that an e-mail that 15 you prepared while you were at Microsoft? 16 Answer: Yes. 17 Question: Now, you write in response 18 to Mr. Slivka's e-mails, you bring up some 19 valid points. 20 However, there is one other point that 21 needs to be understood. ISVs that have access 22 to our code often take advantage of internal 23 data structures, bugs or other quirks of the 24 implementation. 25 Let me ask you, sir, first, what do 11910 1 you mean when you said -- when you're referring 2 to other ISVs that had access to the code? 3 Answer: I didn't -- I don't see the 4 word other used with ISVs, but -- maybe you 5 could restate your question for me. 6 Question: All right. When you say 7 ISVs that have access to our code often take 8 advantage of internal data structures, bugs or 9 other quirks of the implementation, was it your 10 experience that if ISVs got access to the 11 underlying code, that they would, to use your 12 words, often take advantage of and then you 13 refer to internal data structures, bugs, and 14 other quirks of the implementation? 15 Answer: Yep, that's what I wrote 16 here, yep. 17 Question: What did you mean when you 18 said that ISVs with access to the code would 19 often take advantage of bugs and other quirks? 20 Answer: Well, it's human nature to 21 gain an advantage over your competition any way 22 you can. 23 And when you understand how a piece of 24 code works, you can write your application -- 25 how a piece of operating system code works, you 11911 1 can write your application to exploit that. 2 That locks the ISV in. 3 I think a good example is in DOS 4 because DOS was in real mode. And software 5 developers could look into and see what was 6 going on inside of DOS. 7 They would go and do things like -- 8 they could pull out the system, the prompt that 9 on a command line they could pull that out of 10 DOS and recognize that that's what it was, and 11 then use that in their application. That's a 12 fairly trivial example. 13 But then it causes that location of 14 that cursor, that prompt character, to be 15 somewhat locked in by the ISV. 16 Question: Is it also -- I'm sorry, 17 had you finished? 18 Answer: Yeah, sure. I'm finished. 19 Question: If you hadn't, sure, 20 please. 21 Answer: No, I'm done. 22 Question: Now, and is it true that 23 when ISVs chose to access code in that manner, 24 they might even write to bugs? 25 Answer: That's true. 11912 1 Question: Let me ask you this. 2 In the late 1980s, early 1990s, 3 roughly how many, including ISVs of any size, 4 roughly how many ISVs were there? 5 Answer: Couldn't tell you exact 6 numbers. I believe there were more than a 7 thousand. 8 Question: More than a thousand ISVs. 9 Does that include -- does that also 10 include companies, customers that might be 11 developing, customizing their own applications 12 for Windows internally? 13 Answer: No. They're probably, easily 14 multiples more of what we call enterprise 15 customers building internal applications. 16 Question: So if I understand -- and 17 you tell me if I'm wrong -- there were 18 something in excess of a thousand ISVs out 19 there, and in addition to that, there would 20 have been several thousand additional companies 21 or entities that would write their own 22 applications, they would customize their own 23 applications? 24 Answer: That's true. 25 Question: Now, when you wrote in 11913 1 response to this e-mail you were shown the 2 other day and you were referring to the 3 difficulties with giving ISVs access to this 4 code, you write this prevents us from changing 5 the code, what do you mean by that? 6 Answer: Well, a lot of this was borne 7 out of my experience with DOS and is somewhat 8 applicable to Windows, was somewhat applicable 9 to Windows at the time. 10 But by taking advantage of those 11 quirks, they make it much harder to change the 12 way a piece of code works. Much harder to 13 change the way an API works because their 14 applications, especially if they're popular 15 applications, you can't break a popular 16 application in a new release of an operating 17 system. 18 Question: And is it true that that 19 would be even a bigger problem if you had a 20 thousand different ISVs around the country all 21 engaged in that sort of exercise? 22 Answer: Well, we were never too 23 worried about the vast majority. It's the ones 24 that had popular applications. 25 Question: And the problem -- the 11914 1 problem, as I understand you, is that if they 2 decided to go into the underlying source code 3 or whatever and write to something, for 4 example, that was a bug, once it was in their 5 product, you were kind of stuck with that bug? 6 Answer: Well, it ceased being a bug. 7 It became a feature. 8 Question: Right. Even though it had 9 originally been a bug? 10 Answer: That's correct. 11 Question: And that's the problem that 12 you're talking about. That's one of the 13 problems you're talking about in this e-mail? 14 Answer: Yes, that's correct. 15 Question: Now, you also write, to 16 follow this approach that Mr. Slivka talked 17 about, you write this would add a, quote, 18 serious burden of compatibility beyond the API. 19 What do you mean by that? 20 Answer: Well, I think that's 21 precisely what we've been talking about. 22 You have to maintain a level of 23 behavior compatibility. I think we discussed 24 this in the last part of the deposition, 25 previous part of the deposition. 11915 1 Question: So is it true that your 2 view on this was rather than trying to give 3 people access to source code, that the better 4 approach was to try to do a better job, to see 5 if you could do a better job of documenting 6 APIs? 7 Answer: Yeah, that's true. Although, 8 I'll be honest with you, I didn't want to give 9 out the source code. 10 Question: Yeah, we'll get to that in 11 just a minute, sir, but I appreciate your 12 answer. 13 Now, in fact, it's true, isn't it, 14 that in 1990 Mr. Ballmer directed or instructed 15 that any APIs that were being used in Microsoft 16 applications had to be documented; is that 17 right? 18 Answer: I recall that, yes. 19 Question: And, in fact, Mark Wallson 20 and Scott Ludwig then went out and tried to 21 identify any undocumented, any undocumented 22 APIs that were in the Microsoft apps and either 23 get them documented or get them removed from 24 the applications? 25 Answer: I don't recall who 11916 1 specifically did that, but those are two likely 2 people that would have done it. 3 Question: Now, going back to source 4 code. To be clear here, during the 1988 to 5 1994 time period, is it true that almost every 6 major software company treated their source 7 code as proprietary? 8 Answer: It's probably true, yeah, I 9 believe so. 10 Question: What's your basis for that 11 answer? 12 Answer: Just a general sense of where 13 the industry was prior to open source. 14 Question: Let me ask you this, sir. 15 Since going to work for RealNetworks, 16 have you ever met -- outside of the context of 17 an actual deposition, have you ever met or 18 spoke even with Plaintiff's attorneys who are 19 suing Microsoft? 20 Answer: Beyond this? 21 Question: Any case. 22 Answer: Yes. 23 Question: And in what cases, in 24 connection with what litigation have you done 25 that? 11917 1 Answer: This one, the Caldera. I 2 think those are the only two. 3 Question: And how many times did you 4 speak with the Plaintiff's attorneys in 5 Caldera? 6 Answer: Once, I think. 7 Question: How long did you speak to 8 them for? 9 Answer: An hour maybe. 10 Question: Did you meet with them in 11 person or speak on the phone? 12 Answer: I think I met with him in 13 person. 14 Question: And did you discuss with 15 the Caldera Plaintiff's attorneys activities 16 that you undertook in the course of your work? 17 Did you discuss your employment at 18 Microsoft? 19 Answer: Yes. 20 Question: When you did that, did you 21 contact your former employer, Microsoft, and 22 notify them in any way that you were going to 23 sit down and talk to the Plaintiff's attorneys 24 about your former employment? 25 Answer: No, I did not. 11918 1 Question: Now, in this case, have you 2 had discussions with any of the Plaintiff's 3 attorneys about this litigation aside from your 4 deposition? 5 Answer: Yes. 6 Question: How did that come about? 7 Answer: Well, we've had a preparation 8 meeting, and I did discuss with them probably a 9 year ago, maybe not that long ago. 10 Question: Tell me about the first 11 meeting, if you would. 12 Answer: It's just general discussion 13 about the case. 14 Question: Now, on that occasion, did 15 the Plaintiff's attorneys contact you or did 16 you give them a call to find them? 17 Answer: I believe I called them. 18 Question: Who did you call? 19 Answer: I don't recall. 20 Question: How did you know who to 21 call? 22 Answer: I think it was Crew. 23 Question: You called Gene Crew? 24 Answer: Yes. 25 Question: And did they then come out 11919 1 -- this is about a year ago? 2 Answer: Could be. 3 Question: And did they then come up 4 and meet with you? 5 Answer: We met, yeah. 6 Question: Who did you meet with? 7 Answer: I believe it was Mr. Grossman 8 and Mr. Crew. 9 Question: Those are both attorneys 10 for the Plaintiffs? 11 Answer: I guess, yeah. 12 Question: Is it the same Mr. Grossman 13 that was questioning you at the start of your 14 deposition? 15 Answer: Yes, I believe the same one 16 that's here in this room. 17 Question: Okay. And how long did 18 that meeting last, sir? 19 Answer: Maybe an hour. 20 Question: And was that meeting, was 21 it held at RealNetworks? 22 Answer: No. 23 Question: Where was it? 24 Answer: It was in Seattle in -- I'm 25 not sure of the building. It was offices in 11920 1 Seattle. 2 Question: Was it at their law firm? 3 Answer: Yes. 4 Question: Did anyone else attend the 5 meeting? 6 Answer: I don't recall. There may 7 have been somebody else, but -- no, I don't 8 think so. I think it was just those two. 9 Question: Did you bring an attorney 10 to the meeting? 11 Answer: No. 12 Question: Did you -- did you notify 13 your former employer, Microsoft, that you were 14 going to be meeting with Plaintiff's counsel 15 and talking about your employment at Microsoft? 16 Answer: No. 17 Question: That was about a year ago, 18 you said? 19 Answer: Approximately. 20 Question: What was your next 21 communication with the Plaintiffs? 22 Answer: They sent me several pieces 23 of mail and voice mail asking to set up a 24 deposition. 25 Question: And then you said you had a 11921 1 preparation meeting? 2 Answer: Yes. 3 Question: When was that? 4 Answer: I think a day or two prior. 5 Day prior to the deposition. The first part of 6 the deposition. 7 Question: And how long did that 8 meeting last? 9 Answer: Two hours, I think. No, it 10 wasn't even two hours. It was about an hour 11 and a half. 12 Question: Did you bring an attorney 13 to that meeting? 14 Answer: Mr. Stewart was present. 15 Question: The same Mr. Stewart that's 16 here? 17 Answer: Yes. 18 Question: Did Mr. Grossman discuss 19 any documents with you during either this 20 preparation meeting or your earlier meeting? 21 Answer: Yes. 22 Question: Tell me about that. 23 Answer: He showed me several pieces 24 of e-mail. 25 Question: Do you remember what 11922 1 e-mails he showed you? 2 Answer: Not specifically. A number 3 that were -- all I believe were entered into 4 evidence. 5 Question: Do you know if he showed 6 you any e-mails that your name wasn't on? 7 Answer: I don't think so. I think my 8 name was either handwritten or on the original 9 to or cc list of the e-mail. 10 Question: Now, with respect to 11 RealNetworks, you spent the last seven and a 12 half years at RealNetworks? 13 Answer: That's correct. 14 Question: And RealNetworks licenses 15 and distributes products called, among other 16 things, called the Real Player and the Real One 17 Player? 18 Answer: Yes. 19 Question: And those are products that 20 let Internet users and online users access 21 audio and video material basically being 22 streamed over the Internet? 23 Answer: Yes. 24 Question: And what's your 25 understanding of the market those products are 11923 1 competing in? 2 Answer: I don't understand your 3 question. 4 Question: Well, I mean, as an officer 5 of RealNetworks, what's your understanding of 6 what market your products are in? What market 7 are your products competing in? 8 Answer: Online delivery of content 9 market. I'm not sure I understand your 10 question completely. 11 Question: Okay. Now -- 12 (Whereupon, playing of the video 13 adjourned.) 14 THE COURT: We will take a break at 15 this time for ten minutes. 16 Remember the admonition previously 17 given. Leave your notebooks here. 18 Thank you. 19 All rise. 20 (A recess was taken from 1:32 p.m. 21 to 1:53 p.m.) 22 THE COURT: Everyone else may be 23 seated. 24 Juror number 6 you have a child care 25 problem on Wednesday; is that right? 11924 1 If you can't find someone, we'll 2 accommodate you. Okay. Let us know next week. 3 JUROR NUMBER 1: Send the sheriff. 4 THE COURT: He said send the sheriff. 5 A housekeeping matter, Plaintiffs' 6 Exhibit 2266 is being substituted and the 7 exhibit instead will be 2266A. 8 Is that correct? 9 MR. TULCHIN: Yes, Your Honor. Thank 10 you. 11 THE COURT: Very well. So noted. 12 Please continue. 13 (Whereupon, the following video was 14 played to the jury.) 15 Question: Now, and Microsoft has a 16 competing product called Windows Media Player? 17 Answer: I believe that's a matter of 18 record, yes. 19 Question: Let me ask you, are there 20 any disadvantages to consumers to the 21 combination of DOS and Windows into a single 22 product? 23 Answer: Well, in the context of the 24 products of DOS 5 and 6 and Windows 3.X series, 25 there were disadvantages because of the fact 11925 1 that the products were evolving on different 2 paths. 3 The other aspect of it is that DOS was 4 oftentimes preinstalled on a computer at that 5 point in time and Windows was a separate 6 edition. 7 Having Windows bundled into -- DOS 8 bundled into Windows at that point would mean 9 that there was this extra piece that the 10 consumer would have to buy if that's the only 11 way that they could get Windows. 12 Question: All right. Was there 13 disadvantage to OEMs in having the DOS product 14 and Windows product bundled into the same 15 software product? 16 Answer: Well, probably the biggest 17 disadvantage to the OEMs is that they couldn't 18 pick and choose which version of DOS they 19 wanted to have on their operating system. 20 And I'll give you a good example. 21 Compaq computer had their own version of DOS 22 that was customized to their PCs. 23 A combined Windows and DOS version 24 which required that the DOS version be 25 installed -- or the MS-DOS version be installed 11926 1 over the Compaq DOS version would probably -- I 2 can't give you specific examples -- but it 3 would mean that the advantages that Compaq 4 believed they were giving to their consumer 5 through Compaq DOS would be lost. 6 Question: Were there any 7 disadvantages to Microsoft in the combination 8 of the DOS and Windows products in terms of 9 making changes to the code, costs, et cetera? 10 Answer: Well, certainly the -- 11 certainly there was -- because Windows would 12 have to be maintained separately and DOS 13 maintained separately, you're creating yet a 14 third product, so there's overhead engineering 15 cost. 16 I can't think of any technical 17 disadvantages to the company. It's more 18 engineering cost. 19 Question: I wonder if you would take 20 out from your exhibit list Exhibit 3. 21 And in that, on the third page of the 22 exhibit, Mr. Chestnut reports under the heading 23 DRI competitive response, the first Microsoft 24 product with the nontested DOS warning code, 25 QuickPascal, was released. 11927 1 Does this refresh your recollection 2 that, in fact, QuickPascal, a Microsoft 3 language product, was released with the 4 nontested DOS warning code within it? 5 Answer: Yes, it does. 6 Question: And does the heading of 7 this paragraph DRI competitive response tell 8 you anything about the purpose of the release 9 of Microsoft's language product QuickPascal 10 with nontested DOS warning code in it? 11 Answer: Well, I'm not speculating. I 12 know that even though it was called nontested 13 DOS warning code, it was really aimed at DRI. 14 It was aimed at DR-DOS, pure and simple. 15 Question: Aimed in what way? What 16 did it do to DR-DOS? 17 Answer: Well, you want -- I mean, 18 this is a weapon in a war. And the whole idea 19 is to let the customers know that this is not 20 the best combination of something other than 21 MS-DOS and other -- and Microsoft products. 22 That Microsoft products run best on MS-DOS. 23 Question: Now, tell me, if you would, 24 just briefly, what are -- what were Microsoft's 25 language products used for, specifically 11928 1 QuickPascal? 2 Answer: For developing software. 3 Question: Let me restate the 4 question. 5 Did independent software vendors make 6 use of Microsoft's QuickPascal product? 7 Answer: Yes, they did. 8 Question: And what did they use that 9 for? 10 Answer: Developing applications to 11 run on, in this case the DOS environment. 12 Question: Could language products be 13 used to develop applications for DR-DOS? 14 Answer: Certainly. 15 Question: You currently act in the 16 capacity of someone who develops applications? 17 Answer: That's true. 18 Question: And so this is an area you 19 have some expertise in? 20 Answer: Yes. 21 Question: If you received a language 22 product that had a warning code about its 23 compatibility with a particular operating 24 system, what kind of effect would that have on 25 you in terms of your assessment of that 11929 1 operating system? 2 Answer: I'd be very concerned. I'd 3 certainly want to understand what the problems 4 were or if there were problems. 5 Question: All right. Now, in Exhibit 6 12 you'll see this was another e-mail message 7 from you, that you identified being dated 8 September 30, 1991. Do you see that? 9 Embedded within there is an e-mail 10 message from Chuck Straub. 11 Answer: Yes, I see it. 12 Question: Do you see that? 13 And we talked about this last time. 14 Answer: Uh-huh. 15 Question: And in Mr. Straub's e-mail 16 message he says in the third paragraph down, 17 I've patched a version of Bambi to work with 18 DR-DOS 6 and it seems to run Windows 3.1 19 without difficulty. This same problem may have 20 caused other problems with Windows 3.1 and a 21 swap file under DR-DOS 6. 22 Then he goes on in the next paragraph 23 to say, it is possible to make Bambi work 24 assuming we can come up with a reasonably safe 25 method for detecting DR-DOS 6. 11930 1 Now, is it your recollection that at 2 this time there was, in fact, a patch to Bambi 3 that allowed it to work with DR-DOS 6? 4 Answer: Yes. 5 Question: And you responded to 6 Mr. Straub with an e-mail at the top of the 7 page dated September 30, 1991, where you said, 8 the approach we will take is to detect DR 6 and 9 refuse to load. The error message should be 10 something like, quote, invalid device driver 11 interface, close quote. 12 Now, was this approach that Microsoft 13 would detect DR-DOS 6 and refuse to load, was 14 that implemented? 15 Answer: I believe so. 16 Question: And that was implemented in 17 a released product; correct? 18 Answer: I believe it was. 19 Question: And if you look at Exhibit 20 15. 21 Answer: 15? 22 Question: 15, correct. 23 Answer: Okay. 24 Question: Do you recall you 25 previously identified this as another series of 11931 1 e-mail messages and the last one on the second 2 page being from Scott Quinn dated September 30, 3 1991? Do you see that? 4 Answer: Uh-huh. 5 Question: And in there Mr. Quinn 6 says, Bambi version 3.5 has passed developer 7 testing. The primary change fixes a major 8 problem with accessing logical units on 9 external hard disks. Also, DR-DOS is detected 10 (needs testing) and Bambi refuses to load. 11 Do you see that? 12 Answer: Yes, I do. 13 Question: And does this reinforce 14 your conclusion that, in fact, DR-DOS detection 15 and Bambi refusal to load was implemented in a 16 final released product? 17 Answer: Yeah, it does. 18 Question: All right. And what was 19 the final product that the Bambi product with 20 the detection and refusal to load was 21 implemented in? 22 Answer: It was SmartDrive, which was 23 included -- SmartDrive was the piece of 24 software which is included in both various 25 versions of DOS and Windows. 11932 1 Question: Can you tell me which 2 versions of DOS and Windows this Bambi or 3 SmartDrive with the refusal to load and 4 detection was implemented in? 5 Answer: Well, Microsoft Windows 3.1 6 and some version of DOS 6, but I'm not -- I'm 7 not familiar enough. Whatever the one that was 8 released post, you know, late 1991. 9 Question: If you would turn now to 10 Exhibit 27. 11 Exhibit 27 you'll recall you 12 previously identified is another series of 13 e-mail messages. 14 And I'd like to focus your attention 15 on the last message on the page from Brad 16 Silverberg to Brad Chase and Steve Ballmer, Tom 17 Lennon at Microsoft. 18 Do you see that? 19 Answer: Yes, I do. 20 Question: And that was dated October 21 8, 1991. 22 And in there Mr. Silverberg -- 23 Mr. Silverberg, remind me, was he your superior 24 at that time? 25 Answer: Yes, he was my direct 11933 1 superior. 2 Question: Your direct superior. 3 And in that Mr. Silverberg reports 4 under the heading weekly activity, you see that 5 he says, Windows 3.1 doesn't run with DR-DOS? 6 We sent one person MS-DOS 5.0 to use 7 and Randyn is working with another large 8 account. This may stop them from going to 9 DR-DOS 6.0. 10 Does this refresh your recollection 11 that at least at some point in time a released 12 version of Windows 3.1 did not run with DR-DOS 13 6.0? 14 Answer: Yes, it does. 15 Question: If you would turn to 16 Exhibit 12 briefly one more time. 17 Answer: Okay. 18 Question: You'll recall this is the 19 e-mail message where Chuck Straub informed you 20 and others that he had put together a patch of 21 Bambi to work with DR-DOS 6. 22 Do you see that? 23 Answer: Yes, I see it. 24 Question: Okay. And then at the top 25 you sent out an e-mail September 30th, 1991, 11934 1 saying, the approach we will take is to detect 2 DR-DOS 6 and refuse to load. 3 Answer: Yes. 4 Question: Who made the decision -- 5 was that decision to detect DR-DOS and refuse 6 to load, was that decision made by you or by 7 someone else in the Microsoft organization? 8 Answer: It was made by me. 9 Question: Were you encouraged by 10 anyone else in the Microsoft organization to 11 take that approach? 12 Answer: Well, certainly all the 13 people on the cc list, Silverberg, Cole, 14 McCauley, Lennon all bought into it and were 15 certainly very supportive of that approach. 16 Question: What was the reason for 17 detecting DR-DOS 6 and refusing to load if 18 Mr. Straub had come up with a patch of Bambi so 19 that it would work with DR-DOS 6? 20 Answer: Well, we don't have to make 21 our applications run faster on DR-DOS. I mean, 22 they're a competitor. I mean, why make the 23 competitor perform well when you can just as 24 validly disadvantage the competitor? 25 Question: If you would turn now to 11935 1 Exhibit 16. 2 You'll see on the first page of 3 Exhibit 16 is an e-mail message from Scott 4 Quinn dated October 16, 1991, concerning 5 SmartDrive. 6 Do you see that? 7 Answer: Yep. Yes, I do. 8 Question: And in that short e-mail 9 message, Mr. Quinn refers in the second 10 sentence to Big Foot packets. 11 Answer: Yes. 12 Question: Do you recall what that 13 was? 14 Answer: Yeah, Big Foot was -- in 15 order to support large -- disks larger than a 16 certain size -- at every step of the evolution 17 of operating system software disks have gotten 18 bigger and bigger and bigger and very quickly 19 outstripped the -- there were size limitations 20 in every operating system, the size of the 21 disk, hard disk that it can manage. 22 And because spindle technology, hard 23 drive technology has been advancing far more 24 rapidly than any other part of the industry, 25 there were a number of techniques for managing 11936 1 large disks. 2 And whenever a disk got over the 3 current limitation, somebody had to invent new 4 ways of doing that. 5 In this particular case, Big Foot 6 packets was Compaq's way in Compaq DOS, the way 7 Compaq chose to handle, manage large disks as 8 opposed to the way that DR chose to manage 9 large disks from the way that Microsoft managed 10 -- decided to manage large disks. 11 So there were actually three separate 12 ways of managing large disks. This was making 13 Compaq large disk support work with Bambi, with 14 SmartDrive. 15 Question: So is it accurate to say 16 that Microsoft decided to make its operating 17 system compatible with the -- let me change 18 that. 19 Is it accurate to say that Microsoft 20 made a decision to make Bambi compatible with 21 the Compaq method of handling large disks, but 22 not the DR-DOS method? 23 Answer: That's basically correct, 24 yes. 25 Question: All right. Here Mr. Straub 11937 1 says, as we noted earlier, this is the mail 2 which announced the fixing of the bug which had 3 previously prevented DR-DOS 6 from working. 4 And he's referring to the embedded 5 e-mail that he has in his message. 6 Do you see that? 7 Answer: Yes, I see it. 8 Question: Okay. But then he goes on 9 and he says, Mr. Straub says, as of this point 10 the DR-DOS 6 check could have been removed. 11 Was that period of time during which 12 the DR-DOS check and warning message remained 13 in Bambi after the bug was fixed? 14 Answer: Yes, most definitely. 15 Question: Do you know why the DR-DOS 16 6 check was kept in Bambi after the bug had 17 been fixed? 18 Answer: I don't recall any specific 19 conversations or discussions about leaving it 20 in. I think there was no incentive to take it 21 out, but I don't think it was specifically 22 discussed one way or the other. 23 Question: I wonder if you could just 24 briefly tell me what the purposes, the various 25 purposes are of a beta program for -- beta 11938 1 testing program for an upcoming release of 2 Windows operating system? 3 Answer: Okay. We went over this in 4 the previous session, but basically there are 5 multiple reasons. 6 The first is to get real world users 7 experiencing the product, feeding back bug 8 reports, that sort of thing, to get a broader 9 -- broader testing than you can do in a lab. 10 And so there are two kinds of testing, 11 bug testing and usability testing that you'll 12 get out of a beta program. 13 The second reason and the second value 14 of beta testing is to get independent software 15 vendors and specialized card manufacturers, 16 hardware manufacturers to look at the product 17 and start building support for the new version, 18 in this case an operating system, a new version 19 of the operating system. 20 And the third one is a marketing 21 program to basically let people know it's 22 coming, get the corporations prepared to help 23 roll it out, but also to forestall competition. 24 There's no question that it's 25 absolutely an invaluable tool for basically -- 11939 1 if a competitor has some feature advantages, 2 you put those features and ship them in a beta, 3 you will forestall the competitor making 4 in-roads into your market. We will hopefully 5 forestall them. Playing leap frog in some 6 ways. 7 Question: And if there are warning 8 messages within that beta release, does that 9 have a marketing effect? 10 Answer: It certainly would be one 11 reason to put those messages in. 12 Question: If you'll take a look at 13 Exhibit 77, you'll see this is another series 14 of e-mail messages. 15 I only want to refer you to the 16 message that begins at the top of the first 17 page. 18 Answer: Okay. 19 Question: And then embedded within 20 that is a message dated December 17, 1991, from 21 a Peterhey. 22 Do you know who that was? 23 Answer: Yeah, that's Peter Heyman. 24 H-e-y-m-a-n, I think. 25 Question: And in his e-mail message 11940 1 Mr. Heyman says, below is the letter created 2 for mailing to our OEM Windows licensees. The 3 first in a series to build support for the 4 Windows 3.1 launch. 5 In that draft letter he says, dear 6 Windows hardware vendor, and then about halfway 7 down there are -- well, there are a series of 8 paragraphs that are entitled fact. 9 Do you see that? 10 Answer: Yes, I do. 11 Question: And the second one says, 12 fact, the Windows 3.1 beta program is the 13 largest and most exhaustive ever in the PC 14 industry. 15 Was that your recollection? 16 Answer: Yeah, yes. 17 Question: And is this Windows 3.1 18 beta program the one that had the AARD code in 19 it that you warned about, warned about 20 potential incompatibility? 21 Answer: Yes, it was. 22 Question: The letter goes on to say, 23 feedback from our 15,000 beta testers has been 24 overwhelmingly positive. 25 Was it common to have 15,000 beta 11941 1 testers in a beta program at this time period? 2 Answer: No, that was pretty unusual. 3 That was very large, unusually large. 4 Question: And what was the reason for 5 including 15,000 end users in the Windows 3.1 6 beta program? 7 Answer: I believe we covered some of 8 this in previous testimony, but basically the 9 large majority of those were more marketing 10 related. A small number were PC OEMs. Some 11 percentage, maybe a couple thousand, I don't 12 remember the exact numbers, were for basically 13 reliability, quality assurance type testing. 14 The vast majority were a marketing. 15 Give it to the thought leaders out there. Get 16 them excited about the product. 17 Question: Thought leader is being 18 what? 19 Answer: Just people, you know -- I 20 mean, what's an example? 21 You know, Microsoft supporters in 22 corporations, that sort of thing. People that 23 had been recognized as being influential in 24 their organizations. Columnists, that sort of 25 thing. 11942 1 Question: If you'd take out Exhibit 2 49, please. 3 Answer: Okay. 4 Question: You'll recall that you 5 identified this as a document that you wrote 6 entitled Windows and DOS development August 7 status report. 8 Do you see that? 9 Answer: Yes, I do. 10 Question: And in that, in the second 11 paragraph there's a paragraph there entitled 12 Windows release 3.0. 13 Answer: Yes, I see it. 14 Question: And you talk about Windows 15 3.0 moved into high gear in August and you go 16 on to discuss the work that's being done? 17 Answer: Yes, I see. 18 Question: All right. At the end of 19 that paragraph you say, more planning has been 20 done and it appears that an end of July 1989 21 ship date is feasible. 22 Do you see that? 23 Answer: Yes, I do. 24 Question: Do you recall when Windows 25 3.0 actually did ship? 11943 1 Answer: Yeah, in May of '90. May of 2 '90. 3 Question: Do you know the reasons why 4 the original ship date of July of 1989 was 5 delayed to May of 1990? 6 Answer: Why does any software product 7 ship. 8 Bugs, additions to the code, changes 9 to the definition of the product. Lots of 10 reasons. 11 Question: But these were normal 12 development problems that you encountered? 13 Answer: Completely normal. There was 14 no desire on anybody's part to slip the 15 schedule. 16 Question: I'd like to refer you to 17 Exhibit 47 briefly. 18 Answer: Okay. 19 Question: And you previously 20 identified this as a series of e-mail messages 21 beginning with some messages at the top of the 22 first page that you participated in. 23 And what I want to do is ask you about 24 some of what appears at the top of the second 25 page of the exhibit, then ask you how that 11944 1 relates to another document that I'm going to 2 show you. 3 Answer: Okay. 4 Question: And there the e-mail talks 5 about ISV usability issues are rarely 6 considered. For example, the palette support 7 API is a total mess because the developers 8 doing the work decided what the API would look 9 like without regard to how the ISV would use 10 it. 11 Do you see that? 12 Answer: Yes, I do. 13 Question: And do you recall that you 14 testified about this palette support on the 15 first day of your deposition? 16 Answer: Yes, I recall. 17 Question: And just for context, I 18 wonder if you could just very briefly state 19 what is referred to there when you talk about 20 palette support, or when the e-mail talks about 21 palette support. 22 Answer: The palette support is about 23 how the graphics display, how the different 24 colors are managed on the graphics display. 25 Question: Could that palette support 11945 1 be the cause of or have any relationship to a 2 problem with screen flicker? 3 Answer: Yes. In fact, it would. 4 Question: And do you know of any 5 instances where that occurred? 6 Off the top of your head. 7 Answer: I can't point to specific 8 instances, but I've known a lot of applications 9 had -- it was an ongoing common problem of 10 screen flicker in application. 11 And you had to do things in a very 12 specific way, and in fact that was an 13 undocumented aspect of the product. 14 How you updated a window, the contents 15 of a window so that it didn't flicker when you 16 change data in a window. 17 Question: All right. And so this 18 problem that's referred to here in Exhibit 47 19 with palette support was the cause of some 20 screen flicker problems with applications? 21 Answer: Yeah, it was one of the 22 causes. There may have been others. I don't 23 recall off the top of my head. 24 Question: All right. Mr. Barrett, I 25 want to ask you some additional questions about 11946 1 how independent software vendors obtain the 2 technical information that they needed to 3 develop applications for MS-DOS or Windows. 4 Answer: Okay. 5 Question: First, I wonder if you 6 could just briefly describe the various types 7 of information that an applications developer 8 needs to efficiently develop applications for 9 use with the Microsoft operating system? 10 Answer: Well, at the lowest level, 11 there's a software development kit which 12 contains API documentation. Then there are 13 varying levels of access to expertise. 14 In the early days, there was basically 15 the documentation and the developers. There 16 was a support program. Support engineers were 17 not that skilled in Windows. The developer 18 support engineers were not that skilled in 19 Windows. 20 And so typically the best way to get 21 the information was to find a friend inside the 22 development team. And so there's a lot of cat 23 and mouse games of trying to keep the ISVs away 24 from my developers because they'll soak up 25 great amounts of time. Over a period of time 11947 1 that evolved in a very structured mechanism. 2 The mechanism that Microsoft has 3 today, which has multiple tiers, develop, all 4 sorts of development courses and all sorts of 5 stuff and a conference once a year, multiple 6 times a year, I think, and that sort of thing. 7 So it's gone from very -- you know, a 8 very simple ad hoc to highly structured. 9 Question: All right. 10 THE COURT: Plaintiffs, could you stop 11 the videotape, please? 12 Members of the jury, there's a matter 13 I need to take up with the attorneys. We'll 14 take a recess. I don't know how long it will 15 last. 16 Please remember the admonition 17 previously given. See you as soon as we can. 18 All rise. 19 (The following record was made out of 20 the presence of the jury at 2:27 p.m.) 21 THE COURT: You may be seated. As 22 soon as you read that, I guess we better go 23 back in there. 24 (At this time, a sealed-by-the-Court 25 record was made by Janis Lavorato.) 11948 1 (The following record was made in the 2 presence of the jury at 2:43 p.m.) 3 THE COURT: Please be seated. 4 Everyone else may be seated. 5 Members of the Jury, it's come to my 6 attention that some members of the jury have 7 seen a newspaper article today in The Register 8 concerning some issues involving this case. 9 I need to ask a couple questions, and 10 just raise your hands. 11 Has anyone visited any of the websites 12 involving this case? 13 (No response by Jury.) 14 THE COURT: Has anyone read any 15 articles other than the one today in the paper 16 about this case? 17 (No response by Jury.) 18 THE COURT: I can't emphasize to you 19 enough the admonition. 20 This case is receiving publicity. It 21 is absolutely imperative that you stay away 22 from any articles regarding this case, whether 23 they are newspaper, magazine, television, 24 radio, talk shows, and whether they're on a 25 website or on any type of Internet. 11949 1 It is absolutely imperative that you 2 abide by the admonition of this Court's order 3 to stay away from those things. So I can't 4 emphasize that enough. 5 I'm going to let you go for the day 6 unless there's anything else. 7 Counsel? 8 MR. TUGGY: No, Your Honor. 9 MS. NELLES: No, Your Honor. 10 MR. GRALEWSKI: Thank you. 11 THE COURT: Before I do, I'm going to 12 read an admonition to you, which I read every 13 week on Fridays. 14 JUROR NUMBER 1: Yeah, we know it by 15 heart. 16 THE COURT: I don't want you going 17 along with me saying it. Just listen, okay. 18 Under your oath as jurors in this 19 case, you are admonished it is your duty not to 20 permit any person to speak with you on any 21 subject connected with the trial of this case. 22 You are not to talk to any of the 23 parties, their attorneys, or witnesses during 24 the trial, even upon matters wholly unrelated 25 to this trial. 11950 1 Should anyone try to discuss this case 2 with you or in your presence, you should not 3 listen to such conversation. You should 4 immediately walk away. 5 If a person should persist in talking 6 to you, try to find out their name and report 7 it immediately to the Court. 8 You also are admonished not to 9 converse among yourselves or with anyone, 10 including family members, on any subject 11 connected with the trial of this case. 12 You should not form or express an 13 opinion on this case and you should keep an 14 open mind until you have heard all of the 15 evidence, the statements and arguments of 16 counsel, the instructions of the Court, and the 17 case is finally submitted to you and you have 18 retired to your jury room to deliberate. 19 Not only must your conduct as jurors 20 be above reproach, but you must avoid the 21 appearance of any improper conduct. 22 You must avoid reading, listening to 23 or watching news accounts of this trial or 24 visiting any websites or Internet sites 25 concerning this trial. 11951 1 Sometimes such accounts are based upon 2 incomplete information or contain matters which 3 would not be admissible in court and could 4 unduly influence your ultimate decision. 5 As a jury, you are the judge of the 6 facts, while the Court is the judge of the law. 7 During the course of this trial I will be 8 required to decide legal questions. 9 Before you leave to deliberate this 10 case, the Court will instruct you on the law 11 you are to follow in reaching your verdict. 12 You should give careful attention to 13 all of the testimony as it is presented to you, 14 for you will only hear it once and you must 15 depend upon your recollection of the testimony 16 when deliberating in your jury room. 17 As stated before, do not form an 18 opinion and keep an open mind until all of the 19 evidence has been received. 20 From time to time during the trial the 21 Court will be required to confer with the 22 attorneys upon points of law that require only 23 the consideration of the Court. These 24 conferences will be conducted outside the 25 presence of the jury. 11952 1 It is impossible to predict when these 2 conferences will be required or how long they 3 may last. However, these conferences will be 4 conducted so as to consume as little of your 5 time as possible while still being consistent 6 with the orderly progress of the trial. 7 Also, from time to time during the 8 trial, the Court will be required to rule on 9 objections or motions of the lawyers. 10 You should not infer anything by 11 reason of the objection, nor may you infer 12 anything from the rulings on the objections or 13 that the Court has any opinion one way or the 14 other concerning the merits of the case. 15 If an objection to the question of a 16 witness is made and the objection is sustained 17 and the witness is not permitted to answer, you 18 should not speculate on what the answer may 19 have been, nor may you draw any inference from 20 the question itself. 21 Additionally, in your jury room, you 22 must not refer to or give consideration to any 23 testimony which may have been given, but then 24 was stricken from the record by the Court. 25 Also, the lawyers in this case are 11953 1 under an ethical obligation not to talk with 2 you. Do not consider them to be aloof if they 3 do not greet you outside of the courtroom. 4 They are merely abiding by their own rules of 5 ethics and the rules of the court. 6 All right. Also I should mention, as 7 you know, this is a long trial. I know that 8 you are paying close attention because I look 9 over and I see that you are, and I appreciate 10 that. 11 I've talked to other judges across the 12 country and lawyers, and they envy our jurors 13 in Iowa because they do take matters seriously 14 and pay close attention. 15 So I have a great deal of respect for 16 our jury system, especially among our citizens 17 in Iowa. 18 So I want to thank you for that and I 19 want to thank you for your paying close 20 attention to everything going on. 21 I hope that you have a good weekend. 22 It's going to be extremely cold I understand, 23 so I hope everyone tries to stay warm. 24 We will commence again at 8:30 a.m. on 25 Monday. 11954 1 And Juror Number 6, you are going to 2 let us know more progress on that. We may have 3 early recess on Wednesday depending upon some 4 child care problems that Juror Number 6 has. 5 So that may or may not happen. So I will know 6 more Monday. Okay. 7 If there's nothing else, then have a 8 great weekend. Leave your notebooks here. 9 Remember the admonition. 10 Thank you. 11 (A recess was taken from 2:50 p.m. to 12 2:59 p.m.) 13 (The following record was made out of 14 the presence of the jury at 2:59 p.m. ) 15 THE COURT: All right. We are on 16 number -- where did we stop? 17 MR. GRALEWSKI: 9236A, your Lordship. 18 THE COURT: Are we just going to argue 19 that now? 20 MR. GRALEWSKI: Yes. 9236A, 9236B, 21 and 9337 are on for this afternoon. And that's 22 it. 23 THE COURT: You may begin. 24 MR. GRALEWSKI: Thank you, Your Honor. 25 9236A is another document that 11955 1 Microsoft has -- another document that 2 Microsoft argues is a posturing document and 3 should be excluded because it is in 4 anticipation of litigation. 5 And so it is not a business record, 6 according to Microsoft. 7 This is a somewhat lengthy 8 communication from Kelly Jo McCarther, the 9 general counsel of RealNetworks, to Jim 10 Allchin, senior Microsoft executive who 11 recently retired, I believe, in the last week. 12 THE COURT: Really. 13 MR. GRALEWSKI: As a side light. 14 In a nutshell, Your Honor, Plaintiffs' 15 argument, not unlike their argument on a number 16 of other documents yesterday, this document 17 would have been written despite any litigation 18 interests that Real may have had at the time. 19 Again, this document -- in this 20 document Ms. McCarther, quote, wants to figure 21 out how to wipe the slate clean and get back on 22 Microsoft's list of valued ISPs. 23 RealNetworks is a customer that needs 24 to have a positive business relationship with 25 Microsoft to succeed, and that is at a minimum 11956 1 one of the purposes of her sending this letter 2 to Mr. Allchin. 3 Microsoft in its brief opposing 4 Plaintiffs' appeal with respect to this 5 document characterizes that document as an out 6 and out attack on Microsoft. 7 And while it certainly points up some 8 of the challenges that Real has had with the 9 relationship, I would submit that it ends on a 10 fairly conciliatory note. 11 Ms. McCarther, after attempting to 12 clarify some of Real's feelings and strategies 13 on various issues, concludes not in a 14 threatening way, not issuing an ultimatum, not 15 threatening a lawsuit, but instead talks about 16 how Real is looking forward to hearing back to 17 see if there are ways we can work 18 collaboratively to promote Real's products on 19 Microsoft's products, XP and MSN, Microsoft's 20 Internet provider. 21 So, at bottom, Your Honor, to view 22 this document through the prism of the law that 23 we need to apply to these documents, to exclude 24 this document, Your Honor must conclude that 25 the primary purpose of Ms. McCarther drafting 11957 1 and sending this document to Mr. Allchin was to 2 posture for litigation, the primary purpose. 3 Under Iowa law, Your Honor also must 4 find that this document was drafted and sent 5 because of, because of litigation. 6 And I would remind Your Honor of the 7 Vaccaro case cited at page 6 of your brief. 8 There the Court held we do not understand 9 Palmer v. Hoffman -- that was the primary 10 purpose test case. 11 We do not understand Palmer v. Hoffman 12 to require the exclusion from evidence of all 13 records which were made with some contemplation 14 that might be valuable in the event of 15 litigation. 16 And at most, that is what this 17 document does and, under the Vaccaro case, the 18 Court should not exclude it. 19 Thank you. 20 THE COURT: Can I exclude it for being 21 too long? 22 MR. GRALEWSKI: No comment, Your 23 Honor. 24 THE COURT: Okay. Go ahead. 25 MR. JONES: In fact, Your Honor, if 11958 1 the Court were to conduct some surgery, this is 2 a wolf in sheep's clothing. 3 It begins with the sweetness in light 4 first paragraph, it ends in sweetness in light 5 last paragraph. In the middle are nothing but 6 litigation-type allegations leveled against 7 Microsoft. 8 This is a document that Plaintiffs ask 9 to be accepted for its truth as a business 10 record, as something done in the regular 11 course. 12 The Special Master rejected that 13 argument. And if you -- in simply going 14 through some of the paragraphs after that 15 opening paragraph and before the final one, 16 it's clear why the Special Master concluded 17 that this, like document PX 2336 and 2391, 18 which this Court found to be posturing 19 documents in Phase 6, they were letters from 20 Mr. Frankenberg to Mr. Gates leveling 21 allegations. This type of communication is of 22 apiece with that. 23 And, again, just to focus on the -- we 24 heard from the last paragraph. If you go up 25 two paragraphs from the bottom on that last 11959 1 page, you get a sentence like this. We are 2 similarly concerned that exclusive agreements 3 Microsoft has entered into with content 4 providers, such as the National Hockey League 5 leveraged Microsoft's operating system monopoly 6 in a manner that bolsters the monopoly while 7 disadvantaging competing software applications. 8 The letter is replete with 9 observations and conclusions that Microsoft is 10 exercising its power inappropriately, 11 disadvantaging RealNetworks, delaying providing 12 information, engaging in all sorts of what 13 RealNetworks characterizes as wrongful conduct. 14 I think with this document it's also, 15 again, helpful to think of context. 16 Without an understanding of the 17 context, the first and the last paragraphs 18 might make one think this was a normal business 19 communication. 20 Again, on November 15 of 2001 the 21 United States Department of Justice publicized 22 its competitive impact statement where it 23 discussed the proposed final judgment that it 24 had entered into with Microsoft resolving the 25 government case. 11960 1 RealNetworks within two months was 2 going to file a document opposing that final 3 judgment. 4 And this document that we have from 5 Ms. McCarther, the general counsel, is, in 6 fact, airing out a number of the arguments that 7 we will see in RealNetworks opposition to the 8 final judgment that was filed on January 28 of 9 2002 in which we attached to our brief as 10 Appendix B. 11 In particular, I want to note an 12 aspect of this letter that clearly shows that 13 it is a posturing document, a document trying 14 to set up Microsoft for comments that then will 15 come in in the litigation context when 16 RealNetworks opposes the settlement. 17 If you go to 5056, the letter at 5056, 18 which is about, I guess, the third page in -- 19 THE COURT: 50 -- 20 MR. JONES: Yeah, at the very bottom, 21 I'm sorry, Your Honor. 22 THE COURT: I'm sorry, okay. 23 MR. JONES: The second paragraph. In 24 the middle of the second paragraph there's a 25 reference made to the government's competitive 11961 1 impact statement. 2 And it talks about how Microsoft is no 3 longer going to withhold capabilities that are 4 inherent in the SAP -- that's secured audio 5 path features -- as they are implemented in a 6 Windows operating system. 7 And she goes on to say, we look 8 forward to getting access to SAP and Windows 9 very soon given that Microsoft is already 10 shipping products supporting SAP. 11 So what she's doing here is she's 12 interpreting the proposed final judgment and 13 saying we believe it obligates you now to 14 provide that SAP access. 15 If you go to the next document we are 16 going to talk about, 9236B, you see that's what 17 happened. 18 About two weeks after Ms. McCarther's 19 letter there is what might appear to be an 20 innocuous request from RealNetworks to 21 Microsoft for what? For the access to the SAP 22 -- to the SAP functionality. 23 So we are seeing the follow-on 24 communication. And, again, the assertion by 25 RealNetworks that Microsoft -- in a punitive 11962 1 business record, the assertion by RealNetworks 2 that Microsoft has this obligation to provide 3 immediate access. 4 And then to tie it all together, if 5 you go to the RealNetworks' opposition to the 6 proposed final judgment filed about a month 7 after the e-mail we just looked at, 9236B, and 8 if the Court doesn't have a ready copy to the 9 comments of RealNetworks, I'll provide that. 10 May I approach? 11 THE COURT: Yes. What's this? 12 MR. JONES: These are RealNetworks -- 13 I'm sorry. 14 What I've handed the Court is what was 15 attached to our brief as Exhibit B. These are 16 the comments of RealNetworks in opposition to 17 the settlement that Microsoft had reached with 18 the United States government and some settling 19 states that ended the government case. 20 And if you look at page 30 of this 21 litigation document filed about a month after 22 the e-mail, two months after Ms. McCarther's 23 letter in which she raises this SAP issue, we 24 see right before the heading F, we see this 25 passage on page 30 of RealNetworks' opposition 11963 1 to the settlement. 2 Despite repeated requests, Microsoft 3 has not provided RealNetworks with any 4 information or even confirmation that it would 5 provide access to SAP. In a January 2002 6 communication to RealNetworks, Microsoft simply 7 pointed to the 12-month time frame and claimed 8 it was in compliance. 9 So clearly what's going on is a setup. 10 9236A is the beginning. It raises a number of 11 issues, including the SAP. 12 9236B, which was originally packaged 13 with 9236A before the Special Master, 9236B 14 continues the setup by having this request to 15 Microsoft for the SAP, access to SAP. 16 And then when Microsoft provides its 17 response, RealNetworks then springs and uses 18 that as a basis to attack the proposed final 19 judgment. 20 So on a number of grounds, the fact 21 that the letter is simply a series of attacks 22 against Microsoft, that it contains 23 interpretations of matters like Microsoft's 24 agreements with other software content 25 providers, as well as the proposed final 11964 1 judgment. 2 And the fact that clearly what's being 3 done with 9236A and 9236B is simply a setup for 4 Microsoft that is then sprung a month later in 5 a litigation document, Exhibit B to our brief. 6 All these show that the documents at issue here 7 cannot be considered regular course business 8 records. 9 They are not in the regular course of 10 business. They are intended to convey a 11 dispute. They are posturing. They are with 12 litigation in anticipation. 13 THE COURT: What if they had written 14 that you owe us some money for a contract and 15 they write that and then they don't get paid 16 and then they bring a suit saying they owe us 17 money for a contract? 18 MR. JONES: That's a question that -- 19 the Plaintiffs have cited the Fortier case. 20 And it's an interesting question because it's 21 the notion of can a lawyer's communications -- 22 are they always -- can they never be business 23 records. 24 And Fortier suggests that sometimes 25 they can be. 11965 1 And if a lawyer is simply conveying 2 deeds or really acting as the business agent 3 for a party, in those instances Fortier 4 recognizes that that could -- that those types 5 of communications might be business records. 6 But in a communication where a lawyer 7 is purporting to interpret a contract and 8 assert a claim against someone, that is not a 9 regular course communication. That's not a 10 business record. 11 And you can understand and it's clear 12 why. The letter shouldn't be taken for its 13 truth. It's asserting a position on an 14 interpretation of a contract in a position 15 that's, you know, adverse to the party to whom 16 the letter is sent. 17 Why I brought Fortier into it is 18 Fortier involved two sets of documents. One 19 was a group of letters -- and this is the part 20 of Fortier that Plaintiffs will point to. 21 One is a group of letters that were 22 sort of regular course correspondence that an 23 attorney was doing for a landowner, a person 24 who had bought a parking lot. 25 But after that discussion on -- I 11966 1 believe it's on page 1333 -- the Court talks 2 about another document that was used in the 3 case. 4 And that document was a letter from 5 the lawyer asserting -- complaining basically 6 about the condition of the parking lot. 7 And the Court recognized that it was 8 -- that the document was only used for a 9 limited purpose, not as a business record, but 10 as to show that the parties at that point in 11 time had not reached an accommodation. So that 12 there was a relevant not for truth purpose for 13 that communication. 14 So communications to some -- lawyer 15 communications might in limited circumstances 16 be acceptable as business records, but the type 17 of communication we have from Ms. McCarther 18 certainly isn't. 19 And we believe the Fortier case does a 20 nice job of illustrating the two types of 21 communications that can come from lawyers. 22 THE COURT: So are you saying that if 23 a letter came from one business to another 24 that, like you said, the parking lot that you 25 sold us or did, whatever, is not in the 11967 1 condition we expected, that's a business 2 communication? 3 MR. JONES: I'm saying -- no. 4 THE COURT: It's exception to hearsay? 5 MR. JONES: What I'm saying is that if 6 you have a communication interpreting a 7 contract, which your hypothetical was, you 8 violated the contract and you owe us money, 9 that's a posturing document and that would not 10 be a business record. 11 THE COURT: Okay. 12 Anything else? 13 Mr. Jones, anything else? 14 MR. JONES: I was checking with 15 Mr. Tuggy. 16 He says keep my mouth shut. 17 THE COURT: Mr. Gralewski? 18 MR. GRALEWSKI: Yes, Your Honor. 19 With all due respect to Microsoft's 20 argument, Microsoft does have a response to the 21 Vaccaro case and the law that we believe 22 controls here. 23 Let's assume for a moment that this 24 document would be valuable to Real in future 25 litigation against Microsoft. 11968 1 Unless Microsoft can show that the 2 only reason that the letter was drafted was 3 because of that future litigation, under 4 Vaccaro, it can't meet its burden and the 5 document shouldn't be excluded. 6 It can't meet its burden under this 7 standard for the very reason that Mr. Jones 8 cited. 9 And he points to a very real issue 10 that Real was experiencing at the time, which 11 was their need for the secure audio path API. 12 And Ms. McCarther, in part, through 13 the use of this letter, attempts to acquire 14 that API so that Real can conduct its business. 15 That request is entirely consistent 16 with Mr. Barksdale requesting the phone book 17 API that Netscape needed at the time so that it 18 could compete in the marketplace. 19 As I mentioned yesterday, that was PX 20 7766. 21 Certainly Microsoft -- and they took 22 the position when we argued the Phase 6 appeal 23 that Netscape was posturing for litigation. 24 They were cooperating with the government. 25 They were getting ready to sue them. They 11969 1 wanted the government to sue them. 2 But nonetheless, in that instance, 3 Netscape needed an API so that its product 4 could compete. 5 That at a minimum, Your Honor, is one 6 reason that Ms. McCarther writes her 7 communication to Mr. Allchin here, and because 8 of that, under the Vaccaro case, even if this 9 document might be valuable in future 10 litigation, even if Ms. McCarther contemplated 11 future litigation, under the Vaccaro case it is 12 not an inadmissible document for those reasons. 13 THE COURT: Anything else? 14 MR. JONES: Your Honor, if this had 15 been a simple technical communication from one 16 technical person that had regularly 17 communicated with Microsoft or if it were -- 18 actually better, if it were internal 19 RealNetworks documents among the technical 20 folks, boy, we really need access to SAP 21 capability from Microsoft, that would help us 22 do certain things, that falls much closer to a 23 business record. 24 And, in fact, I think the argument 25 would be fairly hard for us to rebut that that 11970 1 type of communication fails to meet business 2 record status. But that's not what's being 3 done here. 4 It's not saying we need this to 5 compete. Ms. McCarther is saying we think -- 6 well, the government has just entered a 7 settlement. 8 And we think that settlement according 9 to the government obligates you to make this 10 functionality available to us. We look forward 11 to you quickly making that functionality 12 available. 13 So what's happening here is not some 14 purely technical conversation. It's an 15 interpretation of a proposed final judgment, 16 and then this bald demand, then, that Microsoft 17 getting compliance with RealNetworks' 18 interpretation of the proposed final judgment. 19 And that interpretation is what 20 Plaintiffs seek to have accepted for its truth. 21 That simply is not appropriate. It is not 22 appropriate. 23 And as the Court recognized when it 24 excluded PX 2336 and PX 2391, posturing 25 positions should not be taken for their truth. 11971 1 It's simply not an appropriate offer 2 for the business record exception. If you want 3 to make argument, make argument in closing 4 statement. But it's not the job of a letter 5 written by a lawyer raising the dispute after 6 dispute after dispute. 7 THE COURT: 9236B. 8 MR. GRALEWSKI: 9236B, Your Honor, are 9 two -- I should say a collection of Real and 10 Microsoft e-mails. 11 Plaintiffs argue in their brief that 12 the Microsoft e-mails, the statements by Steven 13 Fell are admissions. 14 The response to that argument in 15 Microsoft's resistance is that Plaintiffs did 16 not make that argument to the Special Master, 17 so it is waived. 18 In response to that, Your Honor, I 19 would submit that we did believe at the time, 20 and still do, that these are business records. 21 The Special Master did reject that. 22 As we've indicated to Your Honor 23 during the Phase 6 appeal, Special Master 24 McCormick made clear on numerous occasions that 25 evidentiary issues could be cured subsequently 11972 1 by the parties. 2 So I would ask the Court not to accept 3 Microsoft's waiver argument and to rule 4 straight up that if these are admissions, they 5 should be treated as such and admitted into 6 evidence. 7 With respect to the e-mails by 8 Mr. Banfield of RealNetworks, again, this is a 9 request by Mr. Banfield for APIs and other 10 things that RealNetworks needs to conduct their 11 business. 12 Again, not to rehash all of my 13 arguments, I will refer the Court to PX 7766. 14 This is a media player, not an 15 Internet browser, but the same exact thing is 16 going on. 17 Both the Internet company, Netscape, 18 and the media player company Real, need certain 19 information from Microsoft to do their job, and 20 that's what Mr. Banfield is attempting to 21 obtain through this series of e-mail messages 22 back and forth with Mr. Fell of Microsoft. 23 Thank you, Your Honor. 24 MR. JONES: Your Honor, I went back -- 25 well, first let me address the cure issue. 11973 1 Because while we acknowledge that 2 perhaps the Special Master indicated there 3 would be some opportunity for cure, the 4 qualifications or those things that would make 5 a communication an admission are not something 6 that could just be recently discovered by 7 Plaintiffs. 8 So this was -- there's no new 9 information that they would need or would 10 develop to determine whether the communication 11 from Microsoft was an admission. 12 It was an objection that should have 13 been raised before the Special Master. It was 14 not. Plaintiffs have in other contexts argued 15 very vigorously objections that should have 16 been raised to the Special Master are waived if 17 they were not raised before the Special Master. 18 And the cure justification simply 19 doesn't cut it. 20 Mr. Fell was clearly a Microsoft 21 employee making a statement, in fact, at the 22 time that this matter was before the Special 23 Master. And there's no new evidence that 24 Plaintiffs point to as having come to light to 25 them to help cure that. 11974 1 So really this matter could be 2 disposed quickly by just saying, look, the 3 Microsoft part of this is not an admission. 4 Even if it were, it was not raised as a 5 justification for admission when it should have 6 been. 7 But getting back to the threshold 8 issue of whether this document 9236B qualifies 9 as a business record. I touched on it when I 10 discussed 9236A. 11 It is of apiece with that 12 communication. And I think when you flip 13 through 9236B, the Court will see that there is 14 no technical discourse going on. 15 The justification that Mr. Banfield 16 states for wanting access to SAP is not we need 17 it so we can get to market sooner or we need it 18 for a particular piece of software development. 19 He's asking for it because, as he 20 states in the second communication on the first 21 page of the document, they're asking for it 22 because they say that the DOJ, in defense of 23 the proposal of the settlement, has said your 24 proposed settlement with them won't permit you, 25 Microsoft, to withhold capabilities inherent in 11975 1 the SAP features as implemented in the OS or OS 2 apps and that Microsoft will make these 3 features available. 4 So ISVs like us can build, quote, 5 products like that, products that rely on them. 6 So we see that, again, a RealNetworks' 7 employee is simply using the same 8 interpretation of the proposed final judgment 9 that Ms. McCarther used and based on that 10 interpretation of the proposed final judgment 11 makes a demand on Microsoft. 12 This is not a normal communication. 13 And we see that less than -- about a month 14 later after these communications, RealNetworks 15 then runs to the Court and touts as one of its 16 reasons for opposing the settlement the fact 17 that Microsoft didn't provide them immediately 18 with access to this SAP capability. 19 This is not a regular business course 20 communication. 21 THE COURT: Well, if you take out the 22 second sentence, isn't the document more of a 23 business record? It's asking for something? 24 The truth of the matter is it's requesting 25 something. 11976 1 MR. JONES: They are asking for 2 something, but is it -- but why they are asking 3 for it matters. 4 I mean, is there a legitimate business 5 reason for it? Have they been asking for this 6 in the past? Plaintiffs don't provide that 7 context. They don't even show that necessarily 8 this is something really that RealNetworks 9 needs. 10 We don't know why they're asking. The 11 only reason they put forward in the document is 12 that they interpret the proposed final judgment 13 as requiring Microsoft to do it. 14 So the surgery that the Court suggests 15 doesn't cure the fundamental problem, which is 16 that this document was not prepared in regular 17 course. 18 This document was prepared as a part 19 of a litigation strategy. And this document 20 doesn't just convey technical exchanges between 21 Microsoft and RealNetworks. It involves 22 interpretation of obligations that are imposed 23 by the proposed final judgment. 24 THE COURT: What if it was done for 25 both? 11977 1 MR. JONES: Well, there may -- it may 2 have been, but nothing in these documents, 3 nothing in 9236B suggests that it was done for 4 both. 5 The only justification that we can see 6 from the documents is this interpretation of 7 the proposed final judgment. 8 THE COURT: What if they need the APIs 9 for their business? 10 MR. JONES: Your Honor, if that -- if 11 Plaintiffs could have provided the context that 12 showed look at, this was a regular 13 communication, these folks have been engaged in 14 this for a while, if they had come forward with 15 that kind of showing, then they might have been 16 able to show that there was this so-called dual 17 purpose. 18 But, but that showing hasn't been 19 made. 20 Plaintiffs can't rely or can't ask the 21 Court to speculate about a legitimate reason 22 that isn't itself evident in the document. 23 THE COURT: Well, what if all that was 24 in this letter was just the last paragraph? 25 MR. JONES: If all that was in here 11978 1 was the last paragraph, I think the context 2 would still indicate that what's going on here 3 -- then there would be ambiguity. 4 There would be more ambiguity as to 5 what's the driving -- the motive force, I think 6 the context, the timing of it following right 7 on the heels of what Ms. McCarther's letter 8 suggested, but I would agree that if you 9 divorce -- let's divorce all of it though. 10 If, in fact, Ms. McCarther hadn't sent 11 her nearly identical demand to Microsoft about 12 two weeks earlier, if RealNetworks weren't to 13 have raised this same SAP access issue in about 14 a month later in opposing the proposed final 15 judgment, if those things hadn't occurred and 16 if we had all the materials surgerically 17 removed, as the Court suggests, then the case 18 for business record justification is much 19 stronger. 20 THE COURT: Is there any authority for 21 your fruit of the poisonous tree hearsay 22 evidence business exception rule? 23 MR. JONES: I don't think it's the 24 fruit of the poisonous tree. I think what we 25 are arguing is simply what Palmer teaches and 11979 1 that is -- 2 THE COURT: I mean, based on your 3 argument, they couldn't ask for any more 4 because they made demands in some stuff in 5 litigation. So even if they asked we'd like 6 3,000 copies of Windows 95, can we get those -- 7 I mean, you could say look -- Judge, look at 8 the past stuff, set them up for litigation. 9 MR. JONES: Your Honor -- 10 THE COURT: I mean, wow. When does 11 the taint end? 12 MR. JONES: Right. But that's not 13 what we're doing here. I mean, we're not 14 saying -- we're not taking an isolated 15 communication and saying -- 16 THE COURT: I am. I'm isolating just 17 the last paragraph. 18 MR. JONES: And then -- but then I 19 agreed that if you isolated just the last 20 paragraph and if you isolated what came two 21 weeks -- and if you took out what happened two 22 weeks before and then a month later, then I 23 agreed with the Court, that this looks to me 24 like a business record. 25 THE COURT: How does the -- let's say 11980 1 the other things don't come into evidence. 2 What's the problem? 3 MR. JONES: If the other things don't 4 come in -- well, part of the problem is -- 5 THE COURT: Just the letter for the 6 order. I mean, I don't know. Where does this 7 end? 8 What if you show that, hey, they had a 9 dispute with us for years, Judge, over a 10 contract and they wouldn't pay us for 3,000 11 units of something and now they want to buy 12 3,000 more units two years later. Judge, 13 they're just setting us up for litigation 14 again. Holy cow. 15 MR. JONES: Right. And that's not the 16 type of argument we are making. 17 But to use your example, if there had 18 been some contract dispute and we were claiming 19 that we were owed $3,000 under that contract 20 and then we sent them in the midst of that 21 dispute a letter that said we have long 22 interpreted the contract and you know it and 23 you have to agree that the contract obligates 24 you to pay something. 25 Next paragraph. Please pay us $3,000 11981 1 at your earliest convenience. 2 You know, I guess Plaintiffs' position 3 might be, well, you can then maybe take out 4 that first paragraph and then what's wrong with 5 that last paragraph? 6 THE COURT: All right. 7 MR. JONES: It just says pay me some 8 money. 9 But, in fact, you know, what motivates 10 it is -- 11 THE COURT: So you are telling me -- 12 let's say we just have -- forget these 13 letters -- forget this letter here. 14 Let's say I just have a letter that 15 says, dear Microsoft salespeople, we want 3,000 16 units of such and such. 17 MR. JONES: I don't see any context 18 that would suggest to me that that's posturing 19 or that that is anything but a normal regular 20 course of business business record 21 communication. 22 THE COURT: So you have no problem 23 with that? 24 MR. JONES: If what you're saying, 25 look, we got no past, no future, we've got a 11982 1 simple communication, please pay me or please 2 give me 3,000 units, that sounds to me like a 3 business record type of communication. 4 THE COURT: So do you agree, then, in 5 some business records between buyers and 6 sellers and people that deal with each other, 7 there are some communications where they say, 8 hey, we realize that, you know, you didn't pay 9 us the full amount for this product, we're 10 willing to sell you this, that wouldn't qualify 11 as a business record then? 12 MR. JONES: No, it -- 13 THE COURT: You've got a dispute in 14 there. 15 MR. JONES: There was a dispute. It 16 doesn't sound to me like a posturing document. 17 THE COURT: Doesn't that happen a lot 18 in business? 19 MR. JONES: You don't -- I haven't 20 seen quite a document like this in this case, 21 and there are millions of documents in this 22 case, but, of course, communications like that 23 occur. 24 If -- let me just -- let's play it all 25 the way out. 11983 1 If the party -- let's say that that 2 dispute then goes to court. And the other 3 side, the party who received the letter said, 4 look, you know, we oppose having that come in 5 because what Plaintiffs want to do with that 6 letter is have our silence to their letter 7 saying, you know, we know you owe us this, but 8 we want to still do business with you, as some 9 sign of acknowledgement that we, in fact, owed 10 them something. That probably wouldn't be 11 appropriate. 12 But if it's simply being admitted for, 13 you know, the truth that there was an order 14 placed or there was a request for an order to 15 be made, that's a regular business 16 communication that satisfies a business record 17 qualification. 18 And we want to make clear, Your Honor 19 -- 20 THE COURT: So -- go ahead. 21 MR. JONES: Actually, it's probably 22 better for me to answer your question than to 23 -- 24 THE COURT: Well, to determine whether 25 or not it's a business record, then you have to 11984 1 look at the -- what you're saying is the intent 2 of the person who wrote it? 3 MR. JONES: I think the intent of the 4 letter writer, as Palmer indicates, can tell us 5 whether this is a regular business 6 communication or whether, in fact, this is 7 being done for some other purpose outside 8 regular course of business. 9 And I think that we don't have to 10 necessarily -- I don't suggest that the Court 11 engage in all these sort of speculations as to 12 intent. 13 What we are asking to do is the Court 14 to look at the context. The very real, real 15 time context in which this communication is 16 taking place. 17 And then look at the content of the 18 e-mail itself. Is there a legitimate business 19 purpose that's even being hinted at or is, 20 instead, the purpose of the communication being 21 indicated by this interpretation of the 22 proposed final judgment? 23 THE COURT: Well, does the exception 24 to the hearsay rule other than saying that it 25 has to be regular course of business and 11985 1 subject to trustworthiness, does it say you 2 have to look at intent? 3 MR. JONES: I wouldn't say, Your 4 Honor, it says look at intent, but it does 5 direct that we have to look at whether, in 6 fact, this was done in the regular course of 7 business. 8 And we can tell from the context 9 here -- and we really aren't asking the Court 10 to speculate. What we're asking the Court to 11 do is look at the concrete context, look at the 12 concrete words in the exhibit itself. 13 And those tell us that this is not a 14 regular business communication. This is an 15 interpretation of a proposed final judgment and 16 a demand that eventually plays a direct role in 17 litigation, Judge. 18 THE COURT: Okay. I may agree with 19 you on that, but I just want to know if you 20 want me to look at content, context only, if 21 there's a dispute between two companies on some 22 matter in business. And during that dispute, 23 in the regular course of business, whatever you 24 want to define it, one company makes a 25 communication to the other, hey, I want to buy 11986 1 6,000 cars from you. 2 MR. JONES: Right. 3 THE COURT: According to your theory, 4 I can't -- that's not a regular course of 5 business because in the context, they're 6 setting them up for something, Judge. 7 MR. JONES: Well, I could spin out a 8 context where that might be a setup. 9 I could spin out a context where 10 perhaps there is going to be an allegation that 11 these parties never had the ability to perform. 12 And they were asking for the 6,000 knowing that 13 these guys couldn't perform and that's why. 14 But if it's just a regular course and 15 they had asked for these types of things in the 16 past, then there would be no basis to read the 17 context as taking that communication out of 18 regular business. 19 THE COURT: Let's say they wrote 20 something like that and they delivered 6,000 21 cars and they want payment and they said no. 22 Why did you interpret that letter of 23 ours as being an order? We are in dispute with 24 you. You couldn't introduce it then and try to 25 collect, could you? 11987 1 MR. JONES: No. In fact, Your Honor, 2 if it looked like a regular -- if it looked 3 like a regular course communication, unless 4 there's an acknowledgement from the party that 5 wrote it that we weren't really communicating 6 in regular course, we were doing something 7 else, but that sounds to me like a regular 8 course communication that's coming in as a 9 business record. 10 And I don't think our argument -- our 11 argument isn't to say doubt everything. It's 12 to say, though, when the context is clear and 13 the content itself of the document is clear, 14 that what's going on is a nonregular -- 15 THE COURT: And that's the point I'm 16 getting at. 17 Is the Court supposed to analyze this 18 in the context of, Judge, doubt everything? 19 Even if it's regular course, Judge, 20 you've got to look at the entire history of the 21 relationship between the parties before you can 22 say this is a regular business record? 23 MR. JONES: No, Your Honor. And we're 24 not -- 25 THE COURT: Because if that's what you 11988 1 want me to do and then out of fairness, if I 2 agree, then I got to do it for both sides. 3 MR. JONES: Right. 4 THE COURT: So I hope you are prepared 5 to defend the same defense of your exhibits. 6 MR. JONES: And I have to be before 7 the Court on Tuesday, and I understand. 8 THE COURT: As I say, that's the law. 9 And if you guys got -- both parties, you better 10 be prepared to bring a lot of evidence to show 11 why I should allow any business record. 12 MR. JONES: And, in fact, I think, 13 Your Honor, it's not the Court looks at the 14 matter with doubt. I believe the law that 15 there's an initial burden to show that this 16 hearsay document -- I mean, that's the 17 underlying -- that's where this starting place 18 is. This is an out-of-court statement. 19 THE COURT: I understand. 20 MR. JONES: And it's being put forward 21 to the Jury for the truth of the matter 22 asserted. 23 THE COURT: I understand. 24 MR. JONES: And there will be no 25 cross-examination. So that's our starting 11989 1 point. But we create an exception. 2 THE COURT: Right. Regular course of 3 business and trustworthiness. 4 MR. JONES: If you can meet it. If 5 you can meet those, then it can come in. 6 What we are saying is that you don't 7 have to show the whole context to get the 8 document admitted. 9 But if the whole context shows that 10 you are not satisfying the regular course of 11 business requirement, then that's something 12 that is appropriate for the Court to look at 13 and to then exercise judgment based on that. 14 THE COURT: Okay. That's fine. 15 What do you have to say, 16 Mr. Gralewski? 17 MR. GRALEWSKI: He's completely wrong, 18 Your Honor. 19 THE COURT: Or, as Mr. Cashman would 20 say, plain wrong. 21 MR. GRALEWSKI: He is plain wrong, 22 Your Honor. 23 THE COURT: Okay. 24 MR. GRALEWSKI: With all due respect 25 to Mr. Jones, who I admire, and always enjoy 11990 1 arguing against, I am frankly surprised, 2 shocked actually, that Microsoft would 3 seriously contest that there is a legitimate 4 purpose to requesting the secure audio path 5 API. 6 And I'm also pretty surprised that 7 Microsoft would contest that trying to get the 8 SAP API is part of one's regular course of 9 business. 10 THE COURT: Well, you've got to admit 11 that first paragraph sure sounds like they are 12 interpreting, doesn't it? 13 MR. GRALEWSKI: And all this first 14 paragraph is doing, Your Honor, all this first 15 paragraph is doing is saying, hey, this is 16 stuff we need. 17 The DOJ settlement now says you got to 18 give it to us because we need it and because 19 you apparently have to give it to us, give it 20 to us. 21 THE COURT: How do you know their 22 interpretation is true, unless you see the 23 document? Isn't that hearsay in and of itself? 24 MR. GRALEWSKI: This sentence, Your 25 Honor -- and Your Honor has suggested a 11991 1 potential solution here is to carve that 2 sentence out. That would be acceptable to 3 Plaintiffs. We don't think that it is 4 required. 5 But certainly, at a minimum, what 6 Mr. Banfield is doing here is engaging in a 7 legitimate business practice and making a 8 legitimate request for necessary technical 9 information. 10 Now, I want to -- and I don't often do 11 this -- volunteer to answer questions that 12 aren't posed to me, but I'm going to do that on 13 a Friday afternoon. 14 THE COURT: Be careful. 15 MR. GRALEWSKI: You asked Mr. Jones 16 what if this letter was written for both 17 purposes. What if it was written not only to 18 get the API, but also to set Microsoft up if 19 Real decided to sue them down the road. 20 And the answer is based on the law 21 that I cited to Your Honor that Microsoft has 22 not contested, they make argument, but they 23 don't contest the law, under the Vaccaro case 24 the document would still come in; under the 25 Wells Dairy case, the Iowa case that says it 11992 1 has to be written because of litigation. 2 If there's another purpose for the 3 document, here the other purpose is to get 4 something that they need, this technical 5 information, something that there's a finding 6 of fact on. Finding of Fact 52 talks about 7 these APIs are important. 8 Mr. Alepin testified that companies 9 like Real and Netscape need these APIs to hook 10 into the Windows system so they can function. 11 That is a legitimate business purpose. 12 And under Wells Dairy, under Vaccaro, even if 13 they are setting them up for things down the 14 road, this still comes in because of that. 15 Lastly, Your Honor, you were grappling 16 with this issue. And it is something that is 17 vexing. And the Fireman's Fund case, the 18 Eighth Circuit case, decided this issue. 19 And if intent of the writer is the 20 issue, the Fireman's Fund case has held in 21 dealing directly with this exact objection, the 22 objection was posturing in anticipation of 23 litigation, and the Eighth Circuit said if 24 that's the issue, that's the Jury's decision. 25 The Jury gets to decide whether the declarant 11993 1 is trustworthy. 2 You've suggested the problem. If we 3 start to allow this issue to be dealt with on 4 the front end, every document becomes subject 5 to this analysis. And there's no end to -- 6 there's no end to the debate. 7 THE COURT: Anything else on this? 8 MR. JONES: Nothing further, Your 9 Honor. 10 MR. GRALEWSKI: The last document, 11 Your Honor, is PX 9337. 12 It's a single-page document. This is 13 an improper opinion objection that we are 14 dealing with here. 15 These are notes that were created by a 16 Carla Schroer, S-c-h-r-o-e-r, in preparation 17 for a 30(b)(6) deposition she gave in Sun 18 Microsystems versus Microsoft. 19 She was the director of software 20 engineering at Sun. 21 And during her 30(b)(6) deposition, 22 she testified that these are notes to help 23 answer questions regarding topics in the 24 notice. And then she identified that the 25 companies listed here were under license from 11994 1 Sun and distributed a product that included 2 Java technology that failed to meet Sun's 3 compatibility requirements. 4 Plaintiffs would submit, Your Honor, 5 that there is no opinion offered in this 6 single-page document. Indeed, Ms. Schroer was 7 responsible for compatibility testing and she 8 has personal knowledge about compatibility 9 issues. 10 If the Court believes that opinion is 11 at issue here, which we don't believe it is, 12 based on her position at Sun, her opinion is 13 rationally based on her own perceptions given 14 that she was the director of software 15 engineering and she was responsible for this 16 compatibility testing. 17 MR. JONES: Response from Microsoft is 18 -- we'd like to make a couple of points. 19 First, again to set the stage for this 20 particular document, Microsoft asserted a 21 hearsay objection to it. 22 Plaintiffs never contested that 23 hearsay exception. This is hearsay. This 24 cannot be accepted for its truth. 25 And at the outset, we're a bit puzzled 11995 1 how Plaintiffs would propose to offer this 2 opinion which could only be relevant if it were 3 accepted for its truth and still -- 4 THE COURT: Well, is there two 5 objections, then, to it? 6 MR. JONES: Well, the hearsay 7 objection has never been contested, so this 8 document is not coming in for its truth. It's 9 hearsay. 10 And so what Plaintiffs have said is, 11 well -- Microsoft also asserted an improper 12 opinion. 13 THE COURT: Did the Special Master 14 rule on the hearsay? 15 MR. JONES: He did not have to because 16 Plaintiffs never raised the hearsay at 17 Microsoft's Special -- Plaintiffs never raised 18 Microsoft's hearsay objection to the Special 19 Master. They conceded that, in fact, this is 20 hearsay and on its face, that's clear. 21 This was not done for the regular 22 course of business. This was prepared to 23 assist a person in taking -- in responding to 24 deposition questions. Clearly a litigation 25 context. 11996 1 So, at the outset, we are dealing with 2 a hearsay document. And the Special Master did 3 sustain, because Plaintiffs challenged 4 Microsoft's hearsay -- I'm sorry, improper 5 opinion, but I think, again, to get the big 6 picture here, it is difficult for Microsoft to 7 even envision a manner in which this could be 8 accepted as opinion if it's not going to be 9 accepted for its truth. 10 I mean, its relevance would depend on 11 it being accepted that this is an accurate list 12 of companies whose products didn't satisfy 13 Sun's Java compliance requirements. 14 So that's a problem that Plaintiffs 15 have not addressed. How are they going to get 16 -- why are we even here in some respects? How 17 did they propose to put forward this document 18 in a manner that is not for its truth, but is 19 nevertheless asserting an opinion? 20 We have not heard -- we haven't really 21 heard from Plaintiffs on that. 22 But even more fundamentally, the 23 document simply is not proper. It does not 24 satisfy the 5.701 requirement for lay evidence. 25 This person has not been put forth as 11997 1 a technical expert. She's not a 702 or 703 2 expert. She's a 701 lay witness. 3 We accept lay testimony or lay opinion 4 testimony, but we accept it only if there's 5 some foundation. 6 The Court is, of course, familiar that 7 handwriting can be a matter of lay opinion. 8 But we don't let me opine on whether 9 some stranger to me's handwriting, in fact, 10 belongs to the stranger. 11 Before I can offer that opinion, I 12 have to show that I'm familiar with the 13 person's handwriting at issue, that I've seen a 14 number of instances of it, and that I now look 15 at the document in question and I can provide 16 based on that foundation a lay opinion. 17 THE COURT: Is this person going to 18 testify? 19 MR. JONES: No, Your Honor. My 20 understanding -- this person may -- 21 THE COURT: By depo or other 22 otherwise? 23 MR. JONES: There have been 24 designations. I don't know if the Plaintiffs 25 plan to play this person's testimony. There's 11998 1 testimony that goes to -- that talks about this 2 exhibit. 3 But there is nothing in that testimony 4 -- in the deposition record or at least the 5 record that's been cited by Plaintiffs that 6 shows me what the methodology was, that shows 7 me that this person had any foundation besides 8 a job title to make the very particularized 9 compatibility opinions that are expressed in 10 9337. 11 And it also brings -- what also makes 12 this improper opinion is it is litigation 13 conduct. The issue always with 701 testimony 14 is will it be helpful to a jury. 15 There are serious questions as to 16 whether a document prepared to assist a Sun 17 person in their litigation against Microsoft is 18 going to accurately reflect the facts, is 19 accurately going to reflect compatibility. 20 I would note that Microsoft is one of 21 the companies that this document says had 22 noncompatibility issues with the Sun Java's 23 runtime environment. 24 So because this was created in a 25 litigation context, there is reason to question 11999 1 whether it would, in fact, be helpful or 2 whether this is, in fact, an advocacy piece. 3 And, second, because there's no 4 foundation for this person having provided this 5 opinion -- I mean, you can have a title, but if 6 you haven't, in fact, done the work or been 7 associated with the compatibility work, then 8 you have no business bringing an opinion into 9 court. 10 If you're a percipient witness, then 11 you are limited to your personal knowledge and 12 you are not allowed to give opinions unless 13 you've got foundation and it's an appropriate 14 subject for lay opinion. 15 THE COURT: Anything else? 16 MR. GRALEWSKI: Your Honor, Exhibit C 17 to the Plaintiffs' memorandum in support of its 18 appeal is a copy of Ms. Schroer's deposition 19 that has been affirmatively designated from by 20 Plaintiffs in this case. And so that addresses 21 one question Your Honor has. 22 We believe that her deposition lays 23 adequate foundation for her to be able to 24 testify on issues of compatibility. 25 I will note without arguing an 12000 1 agreement that the parties' course of conduct 2 has been where you have somebody such as a 3 director like Ms. Schroer that the parties have 4 assumed for purposes of our evidentiary issues 5 that they essentially know what they are 6 talking about. 7 On the issue of the hearsay point, 8 again, without conceding that this document is 9 hearsay, I believe the explanation is that at 10 the time of the Special Master process -- and 11 I'm trying not to make excuses, but there were 12 a lot of things going on and there's a lot of 13 paper. We did not have the identification of 14 the declarant at the time the papers were 15 submitted. We do now. 16 This is a situation where we're 17 attempting to cure information that was not 18 presented in the past. 19 However, one possible solution here 20 would be to clear out the opinion objection. 21 As I said, we don't think this is opinion. We 22 hear people testify on the big screen and in 23 the chair all the time about compatibility 24 issues. This is simply the director at Sun 25 doing the same thing. 12001 1 But one way to handle this is to clear 2 out the opinion objection to grant Plaintiffs' 3 appeal and then when this document comes up 4 through the course of the trial, Plaintiffs in 5 the context of the litigation may offer it for 6 a nonhearsay purpose. 7 MR. JONES: We, of course, believe 8 that the better way to resolve the improper 9 opinion objection is to affirm what the Special 10 Master concluded, that is, that this is 11 improper opinion and should not be accepted in 12 the Court. It lacks foundation. It's prepared 13 in litigation context. It simply shouldn't 14 come in. And that would save time. 15 MR. GRALEWSKI: I guess, just quickly, 16 I would note that I believe it was in Phase 1 17 or 2 -- well, let me back up. 18 There are documents -- DX is in 19 evidence, Your Honor, based on the argument of 20 Microsoft's lawyers where lawyers themselves 21 created documents for witnesses to aid the 22 witness' testimony during the course of a 23 deposition. 24 And here we have simply the witness 25 herself creating a document to help her answer 12002 1 questions as a corporate designee. 2 And I would submit that this PX should 3 be treated at least on par with the DXs that 4 come in the context of a deposition like that. 5 Thank you, Your Honor. 6 Nothing further from Plaintiffs on 7 Phase 7 appeal with respect to PXs. 8 THE COURT: Very well. 9 Anything further? 10 MR. JONES: Nothing further. 11 THE COURT: All right. I will take a 12 look at it. 13 MR. GRALEWSKI: Thank you. 14 MR. JONES: Thank you, Your Honor. 15 THE COURT: Thank you. 16 Good job both of you. 17 MR. GRALEWSKI: Thank you, Your Honor. 18 MR. JONES: Thank you. 19 (Proceedings adjourned at 3:51 p.m.) 20 21 22 23 24 25 12003 1 CERTIFICATE TO TRANSCRIPT 2 The undersigned, Official Court 3 Reporters in and for the Fifth Judicial 4 District of Iowa, which embraces the County of 5 Polk, hereby certifies: 6 That she acted as such reporter in the 7 above-entitled cause in the District Court of 8 Iowa, for Polk County, before the Judge stated 9 in the title page attached to this transcript, 10 and took down in shorthand the proceedings had 11 at said time and place. 12 That the foregoing pages of typed 13 written matter is a full, true and complete 14 transcript of said shorthand notes so taken by 15 her in said cause, and that said transcript 16 contains all of the proceedings had at the 17 times therein shown. 18 Dated at Des Moines, Iowa, this 2nd 19 day of February, 2007. 20 21 22 ______________________________ Certified Shorthand Reporter(s) 23 24 25