Microsoft has filed its brief [PDF] in the appeal of Novell v. Microsoft, the antitrust litigation Novell brought against Microsoft over WordPerfect. I've done it as text for you. The judge in the case handed Microsoft a win on summary judgment after a jury couldn't reach a verdict. So now it's before the 10th Circuit Court of Appeals. Microsoft's airy arguments go something like this: Yes, we were a monopoly with 95% of the desktop market at the time, and yes, we reversed course after encouraging Novell to use APIs that we then decided not to support, but hey, we don't owe competitors anything. "A monopolist generally
has no duty to cooperate with or assist a competitor whether the decision is
'intentional' or otherwise." We can change our business model any time we want to, as long as we are even-handed and the effect is on everyone, not just Novell. (Novell, however, was the one that Microsoft encouraged to use the APIs, and it was the one Microsoft feared, according to Novell, writing that the decision to drop support for the APIs "involved the intentional inducement of reliance.")
Besides, when Microsoft made the change, Novell had three ways to react, and it stupidly chose the worst one. That's not our fault, Microsoft argues. Novell sealed its own fate. It could have chosen differently, and it would have been better. Anyhow, it was a change to a beta product, and everybody knows you can make sudden changes in a beta. How could Microsoft know Novell was relying on that beta stuff? Because it had an employee who was working closely with Novell after Microsoft encouraged Novell to rely on the APIs, perchance? Anyhow, Microsoft says, everybody in the industry changes beta software, Novell included, so "...Novell cannot base a claim of anticompetitive conduct on a
widespread industry practice." Um. Not everybody else had a 95% desktop monopoly. Capice? Novell has already said that it had three choices, but they were *all* awful.
None of them would have helped Novell to avoid harm. Even if it was a matter of degree, as Microsoft alleges, none of its choices led to anything but disaster for Novell. Microsoft is unable to rebut that, although if you didn't know the case and the details, its brief might make you think they have. Microsoft quotes, without seeming to realize the significance against it, a Novell employee who explained why they chose what they did: As Gibb testified, he chose this path because he wanted to "do
something cooler" and "exceed what was the default stuff." (JA-11782-83.) In
short, as Gibb recognized, Novell was "faced with the age old trade off" between
"get[ting] out a product more quickly and sacrific[ing] features," or, on the other
hand, "delay[ing] until 1996 and try[ing] to build a cooler product." (JA-11825-26.) They could have chosen another path, but it would have required sacrificing features. How is that not awful for Novell? Yes, they would have met the deadline, but with an inferior product that wouldn't have been able to match or exceed what Microsoft was offering. That's the option Microsoft offers? The judge might accept it, not being a techie guy, by his own admission, but it wasn't just the cool factor. It was also the double coding factor, in addition to the loss of features the public would be expecting. And furthermore, Microsoft not only knew it, it planned it with Novell as the target. We know that from the Bill Gates email [PDF] on his decision regarding the APIs ("We should wait until we have a way to do a high level of integration that will be harder for the likes of Notes, Wordperfect to achieve, and which will give Office a real advantage."). To which Microsoft states that even if a sudden change might be a "business tort", it's not a Sherman Act violation.
To which I say, if it's not an antitrust violation, could somebody tweak the law, please, and reboot the system so it runs better than this?
Because Microsoft's argument, as I understand it, is that it can be as bad as it wants to be, target anyone it wants anticompetitively despite being a monopoly, and as long as it carefully handles the change so it looks like it isn't targeting just one competitor and if they can plausibly claim it was just a business decision for some other invented reason, they can get away with killing off a competitor at will. Which is why I'll never buy a Microsoft phone or anything Microsoft. I don't want to reward a business that thinks that's a winning argument, and I surely don't want Microsoft to extend its desktop monopoly into the mobile space, because if they think it's OK to be awful to its competitors, they'll surely keep on doing it, carefully following legal advice on how to stay 1/8th inch inside the law (if this is inside the law). It's like Sony telling that court that promises made by its executives about the Playstation and Linux were just corporate "puffery", nothing anyone could legally rely upon. Fine. They won the litigation that way, but who wants to buy a Sony Playstation any more after hearing that? Given that insight into their values and the reliability of their promises, not me. That's for sure.
Novell is allowed to file an optional reply brief by February 22nd. I expect that they will. Boies Schiller is on the team now, and they like to file a lot of briefs, and this Microsoft brief is enough to set your hair on fire. Mine is smoking, anyway. So many details that are almost true, but not exactly. Just one example. Microsoft claims that when it made the change, it had an employee check around to see how OEMs and partners were taking the news. And Microsoft tells the court that Novell said it was "OK with it":
Before and after making the October 3
Decision, Microsoft surveyed major ISVs to find out if the decision would cause
them any difficulties, and Novell responded by saying that it was not using those
four APIs and was "OK" with the change.
But later, what actually happened is this:
Microsoft promptly notified Novell of the October 3 Decision. On
October 12, 1994, Struss reported in writing that Novell (he wrote "WP," for
WordPerfect) appears "to be OK" with the decision.
As you can see, what happened is that it was the Microsoft employee who *said* Novell "appeared to be" fine with it, not that Novell said so. It was his impression. That is totally different, and the Microsoft version is, to my eyes and sense of ethics, misleading. The whole brief is a masterpiece of selection in that way.
Microsoft argues that under prior case law, it's not an antitrust violation unless you terminate a profitable business relationship, and it didn't do that. Novell, of course, provided evidence that in fact it did terminate the previous relationship, whereby Microsoft provided a lot of handholding to help Novell get it right, and that this help suddently disappeared after the Gates memo. Again, is 1/8" inside the line enough? They didn't terminate the relationship 100%, just the part that mattered most. A too-clever monopolist could certainly plan such a maneuver, don't you think, keeping the victim hanging by a thread, just to be able to argue later in court that it never terminated the relationship?
Novell's claim of deception, Microsoft argues, is frivolous. Nobody cares, as it's not an antitrust violation to lie, it says, quoting the US Supreme Court:
In any event, Novell's "deception" argument is wrong as a matter of
law. Novell cites no case--and Microsoft is aware of none--in which deception of
a competitor gave rise to a claim under Section 2 of the Sherman Act.25 As the
Supreme Court has made clear, "[e]ven an act of pure malice by one business
competitor against another does not, without more, state a claim under the federal
antitrust laws." Brooke Group v. Brown & Williamson Tobacco, 509 U.S. 209,
225 (1993).
One hopes the court of appeals will make note of the "without more" and extrapolate that if, say, Company A Monopoly encourages a competitor, Company B, to rely on code that Company A then pulls out from under them, causing them to fail to be ready for a release date that their business is depending on, and there is an email from the CEO saying they should make the change to make sure Company B can't compete on an even playing field, it just might indicate something "more". Because if that can happen and nobody cares in the antitrust context, then antitrust law has no muscle at all and applies to nobody but companies too dumb to read antitrust case law before they make their anticompetitive moves. Microsoft mocks Novell's theory that there was any viable competition in the operating systems market or that Novell posed any such threat. "It is inconceivable that two or three applications could have impacted
competition in the PC OS market in this fashion." But Bill Gates didn't think so at the time, judging by his email, now, did he? Speaking of lawyer constructs. Which reminds me -- about that email. Microsoft offers a very clever argument about the smoking gun. It argues in footnote 14 that the email was talking about a theoretical future possible competition from Novell, not an extant one: 14 Gates announced his October 3 Decision in a long e-mail that included his
saying that "[w]e should wait until we have a way to do a high level of integration
that will be harder for [the] likes of Notes, WordPerfect to achieve, and which will
give Office a real advantage." (JA-1967 (PX1).) Even if, as Novell argues, this
shows that Microsoft was concerned with competition from Novell (see Novell Br.
21 n.3), Gates' e-mail refers only to a hoped-for advantage for Office from an
expected, later, improved version of the NSEs. Gates was not aware of any harm
to Novell flowing from the October 3 Decision (JA-13761), and he never refers to
any benefit to Windows in competition with any other operating system. Novell's
only claim here is for harm to competition in the PC OS market. That's mighty clever, maybe enough to work, but it's not true, not by my reading of the email. Gates was worried about the APIs making it easier for Novell to compete right then, so he decided to pull them. That's how I read it anyway. If he didn't know Novell was using them, he'd have no reason to worry or to make any change at all. P.S. And Microsoft's claim is that the change was really about worries over applications using the APIs crashing Windows. But the email indicates a very different motive. Not that Microsoft needs outsiders, from my experience, to crash Windows.
[ Update: We continue to work on making all the exhibits from the
Comes v. Microsoft antitrust litigation available as text. And there are a few that we've just finished that seem to contradict Microsoft's brief. Rather than add to the length of this article, I've put them here, so you can take a look if you are interested. - End update.]
Here it is as text, so you can form your own opinion. Microsoft references the Findings of Fact from the US v. Microsoft litigation, so I'm giving you the link, so you can check everything Microsoft says it says. In fact, I hope you will:
********************
No. 12-4143
IN THE
UNITED STATES COURT OF APPEALS
FOR THE TENTH CIRCUIT
________
NOVELL, INCORPORATED,
Plaintiff-Appellant,
--v.--
MICROSOFT CORPORATION,
Defendant-Appellee.
____________
ON APPEAL FROM THE UNITED STATES DISTRICT COURT
FOR THE DISTRICT OF UTAH
THE HONORABLE J. FREDERICK MOTZ
CASE NO. 2:04-CV-1045-JFM
__________________
BRIEF OF DEFENDANT-APPELLEE MICROSOFT CORPORATION
________________
STEVEN J. AESCHBACHER
MICROSOFT CORPORATION
[address, phone]
DAVID TULCHIN
STEVEN L. HOLLEY
SHARON L. NELLES
ADAM S. PARIS
SULLIVAN & CROMWELL LLP
[address, phone]
JAMES S. JARDINE RAY QUINNEY & NEBEKER
[address, phone]
Attorneys for Defendant-Appellee Microsoft Corporation
January 23, 2013
___________________
ORAL ARGUMENT REQUESTED
CORPORATE DISCLOSURE STATEMENT
Pursuant to Rule 26.1 of the Federal Rules of Appellate Procedure,
Microsoft Corporation states that it has no parent corporation and that no publicly
held corporation owns 10% or more of its stock.
Dated: January 23, 2013
SULLIVAN & CROMWELL LLP
By: /s/ David B. Tulchin
David B. Tulchin
SULLIVAN & CROMWELL LLP
[address, phone]
Attorney for Defendant-Appellee Microsoft Corporation
TABLE OF CONTENTS
Page
INTRODUCTION ............................................. 1
STATEMENT OF THE ISSUES ................................. 3
STATEMENT OF THE CASE .................................. 4
A. Novell's Claim at Trial ..................... 4
B. Trial ................................... 6
C. The Decision on Microsoft's Renewed Motion for Judgment as
a Matter of Law .............................. 8
STATEMENT OF FACTS ...................................... 8
A. Microsoft ................................... 8
B. WordPerfect Corporation .......................................... 11
C. Novell ....................................................... 12
D. Microsoft's "Evangelization" of Windows 95 ................................... 13
E. Microsoft's Beta Version of Windows 95 ......................................... 16
F. The October 3 Decision ...................................................... 18
G. The October 3 Decision Did Not Harm Novell ................................. 20
H. Microsoft's Continued Cooperation with Novell ............................... 23
I. Novell Had Three Options and Chose the Most Difficult ................. 24
J. Quattro Pro Caused the Delay ................................................ 29
SUMMARY OF ARGUMENT ......................................... 32
STANDARD OF REVIEW .......................................... 33
ARGUMENT ................................................. 34
ii
TABLE OF CONTENTS
(continued)
Page
I. THE TRIAL COURT CORRECTLY HELD THAT THE OCTOBER
3 DECISION WAS NOT ANTICOMPETITIVE CONDUCT ........ 34
A. Novell Was Aware from the Start that the Windows 95 Beta
Could Change .......................................36
B. Microsoft Never Terminated a Profitable Business Relationship
with Novell ....................................................42
C. The October 3 Decision Did Not Treat Novell Differently ...............43
D. Novell Cannot Avoid the Requirements of Aspen by Recasting Its Claim as One for Affirmative Conduct or Deception ...................44
II. MICROSOFT’S CONDUCT DID NOT HARM COMPETITION IN
THE PC OPERATING SYSTEM MARKET .............................48
A. The Franchise Applications Theory ..............................52
B. The Middleware Theory .....................................54
III. THE OCTOBER 3 DECISION DID NOT CAUSE ANY DELAY ............58
A. Quattro Pro Caused Novell’s Delay ..................................59
B. Novell’s Poor Business Choice Caused the Delay .............................61
CONCLUSION ...............................................64
iii
TABLE OF AUTHORITIES
Page(s)
CASES
A.I.B. Express, Inc. v. FedEx Corp.,
358 F. Supp. 2d 239 (S.D.N.Y. 2004) ............................................................... 43
American Professional Testing Services v. Harcourt Brace Jovanovich,
108 F.3d 1147 (9th Cir. 1997) ........................................................................... 47
Aspen Skiing Co. v. Aspen Highlands Skiing Corp.,
472 U.S. 585 (1985) ....................................................................................passim
Bankers Trust Co. v. Lee Keeling & Associates, Inc.,
20 F.3d 1092 (10th Cir. 1994) ........................................................................... 60
Berkey Photo, Inc. v. Eastman Kodak Co.,
603 F.2d 263 (2d Cir. 1979) .............................................................................. 47
Brooke Group Ltd. v. Brown & Williamson Tobacco Corp.,
509 U.S. 209 (1993) ........................................................................................... 47
Burrell v. Armijo,
603 F.3d 825 (10th Cir. 2010) ........................................................................... 33
Christy Sports, LLC v. Deer Valley Resort Co.,
555 F.3d 1188 (10th Cir. 2009) ................................................passim
Four Corners Nephrology Associates, P.C. v. Mercy
Medical Center of Durango, 582 F.3d 1216 (10th Cir. 2009) ...................passim
Herrera v. Lufkin Industries, Inc.,
474 F.3d 675 (10th Cir. 2007) .............................................. 34, 60
In re Microsoft Corp. Antitrust Litigation,
274 F. Supp. 2d 743, 746 (D. Md. 2003) ............................................... 14
Intergraph Corp. v. Intel Corp.,
195 F.3d 1346 (Fed. Cir. 1999) ................................................. 37, 47
J.T. Gibbons, Inc. v. Crawford Fitting Co.,
565 F. Supp. 167 (E.D. La. 1981) .........................................62, 63
-iv-
iv
TABLE OF AUTHORITIES
(continued)
Page(s)
CASES
(continued)
(continued)
J.T. Gibbons, Inc. v. Crawford Fitting Co., ,
704 F.2d 787 (5th Cir. 1983) ............................................................................. 62
King & King Enterprises v. Champlin Petroleum Co.,
657 F.2d 1147 (10th Cir. 1981) ......................................................................... 50
MetroNet Services Corp. v. Qwest Corp.,
383 F.3d 1124 (9th Cir. 2004) ...............................................................43, 45, 46
Mooring Capital Fund, LLC v. Knight,
388 F. App'x 814 (10th Cir. 2010) ..................................... 55
Multistate Legal Studies, Inc. v. Harcourt Brace Jovanovich Legal &
Professional Publications, Inc., 63 F.3d 1540 (10th Cir. 1995) ....................... 45
Nobody in Particular Presents, Inc. v. Clear Channel
Communications, Inc., 311 F. Supp. 2d 1048 (D. Colo. 2004) ......................... 43
Novell, Inc. v. Microsoft Corp.,
2005 U.S. Dist. LEXIS 11520 (D. Md. June 10, 2005) ....................................... 5
Novell, Inc. v. Microsoft Corp.,
505 F.3d 302 (4th Cir. 2007) ......................................... 5
Novell, Inc. v. Microsoft Corp.,
699 F. Supp. 2d 730 (D. Md. 2010) ............................................... 5
Novell, Inc. v. Microsoft Corp.,
429 F. App'x 254 (4th Cir. 2011) .................................................. 5
Olympia Equipment Leasing Co. v. Western Union Telegraph Co.,
797 F.2d 370 (7th Cir. 1986) ................................................. 41
Pacific Bell Telephone Co. v. Linkline Communications, Inc.,
555 U.S. 438 (2009) ..........................................34, 35
v
TABLE OF AUTHORITIES
(continued)
(continued)
Page(s)
CASES
(continued)
Rambus Inc. v. FTC,
522 F.3d 456 (D.C. Cir. 2008) ................................................... 47
Safeway, Inc. v. Abbott Laboratories,
761 F. Supp. 2d 874 (N.D. Cal. 2011) ........................................... 43
Specialty Beverages, L.L.C. v. Pabst Brewing Co.,
537 F.3d 1165 (10th Cir. 2008) ..................................................... 34
Telex Corp. v. International Business Machines Corp.,
510 F.2d 894 (10th Cir. 1975) ..................................................................... 40, 41
Union Cosmetic Castle, Inc. v. Amorepacific Cosmetics USA, Inc.,
454 F. Supp. 2d 62 (E.D.N.Y. 2006) .................................................... 62
United States v. Colgate & Co.,
250 U.S. 300 (1919) ...................................................32, 34, 35, 36
United States v. Microsoft Corp.,
84 F. Supp. 2d 9 (D.D.C. 1999) ...................................................... 6
United States v. Microsoft Corp.,
253 F.3d 34 (D.C. Cir. 2001) ................................................. 51, 52
United States v. Syufy Enterprises,
903 F.2d 659 (9th Cir. 1990) ............................................ 40, 41
Verizon Communications, Inc. v. Law Offices of
Curtis V. Trinko, LLP, 540 U.S. 398 (2004) ...................................36, 42, 45, 46
Webco Industries, Inc. v. Thermatool Corp.,
278 F.3d 1120 (10th Cir. 2002) ..............................................33, 34, 61
vi
TABLE OF AUTHORITIES
(continued)
Page(s)
STATUTES
Sherman Act, Section 1, 15 U.S.C. § 1 ........................................ 4
Sherman Act, Section 2, 15 U.S.C. § 2 ..........................................passim
15 U.S.C. § 15b ........................................................4, 57
15 U.S.C. § 16(i) .........................................................5, 57
RULE
Fed. R. Civ. P. 50 .......................................passim
OTHER AUTHORITY
3 PHILLIP E. AREEDA & HERBERT HOVENKAMP,
ANTITRUST LAW (3d ed. 2011) .................................................. 52
vii
PRIOR AND RELATED APPEALS
In Novell v. Microsoft, 505 F.3d 302 (4th Cir. 2007), the United States
Court of Appeals for the Fourth Circuit affirmed the dismissal of Counts II through
V of Novell's Complaint and affirmed the denial of Microsoft's motion to dismiss
Counts I and VI. In Novell v. Microsoft, 429 F. App'x 254 (4th Cir. 2011), the
Fourth Circuit reversed the entry of judgment for Microsoft on Count I of Novell's
Complaint and noted that Novell did not challenge on appeal the entry of judgment
for Microsoft on Count VI. The case was then remanded for trial on Count I.
viii
GLOSSARY
API - Application Programming Interface
DOS - Disk Operating System
DRG - Developer Relations Group
FOD - File Open Dialog
GUI - Graphical User Interface
ISV - Independent Software Vendor
JPML - Judicial Panel on Multi-District Litigation
M6 - Milestone 6
MDL - Multi-District Litigation
MS-DOS - Microsoft DOS
NSEs - Namespace Extension APIs
PC - Personal Computer
PC OS Market - PC Operating System Market
RTM - Release to Manufacturing
WPC - WordPerfect Corporation
ix
INTRODUCTION
In 1994, Microsoft's market share in the personal computer ("PC")
operating system market was above 90%, and Microsoft was then developing
Windows 95--which all recognized was a major technological advance and would
drive Microsoft's market share even higher. In June of that year, during the
development of Windows 95, Microsoft distributed a pre-release or "beta" version
of the new operating system to other software companies (sometimes called
"ISVs," for independent software vendors) so they could begin to write
applications that would be ready to come to market not long after Windows 95 was
released. Microsoft's beta license agreement, signed by Novell, said that the beta
of Windows 95 was subject to change, and Novell independently understood the
same thing--betas can and do change.
On October 3, 1994, Microsoft decided that it would no longer
support (meaning no longer provide technical documentation and assistance for)
four application programming interfaces ("APIs") in the beta of Windows 95
(the "October 3 Decision"). APIs are interfaces that can be used by ISVs to allow
their applications to take advantage of services provided by the operating system.
The APIs in question--four out of thousands of APIs in Windows 95--were called
the namespace extension APIs ("NSEs"). Before and after making the October 3
Decision, Microsoft surveyed major ISVs to find out if the decision would cause
them any difficulties, and Novell responded by saying that it was not using those
four APIs and was "OK" with the change. Windows 95 was released on
August 24, 1995 and was a huge success (as many, including Novell, had
anticipated).
Novell claims that the October 3 Decision caused a delay in Novell's
development of its PerfectOffice suite (composed primarily of WordPerfect, a
word processor, and Quattro Pro, a spreadsheet), and that this delay harmed
competition in the PC operating system market ("PC OS market"). This theory
was outlandish from the outset. It was based on the contention that Microsoft's
dominant share in the PC OS market would have declined because Novell's
applications, had they come to market sooner, somehow would have sparked
competition in the PC OS market in which Microsoft had lawfully acquired
monopoly power.
After the trial resulted in a hung jury, the court ruled that Novell had
failed to create a jury issue on any element of its claim. First, the court held that
Microsoft's decision was not anticompetitive because, among other reasons,
Novell knew that the beta of Windows 95 could change because of the license
agreement and industry practice. Second, the court held that Novell's theories as
to how its applications could have affected competition in the PC OS market
-2-
(a market in which those applications did not compete) were unsupported, and in
fact contradicted, by the evidence. Indeed, Novell's CEO admitted that
Microsoft's market share would have been higher (not lower) had Novell's
applications been released in a timely manner. Third, the court held that the
October 3 Decision did not delay Novell; instead, its inability to develop Quattro
Pro was the cause of delay. Each of these three rulings is correct and is a separate
and independent ground for affirmance. There is absolutely no merit to Novell's
appeal.
STATEMENT OF THE ISSUES
Whether the trial court correctly ruled that:
I. the October 3 Decision was not anticompetitive conduct under
Section 2 of the Sherman Act because, among other reasons, Novell
was aware that the beta of Windows 95 was subject to change, and
Microsoft continued to assist Novell in developing its applications for
Windows 95;
II. Microsoft's withdrawal of support for the NSEs caused no harm to
competition in the PC OS market because Novell's applications could
not impact Microsoft's monopoly power in that market; and
-3-
III. the October 3 Decision did not delay the release of Novell's
applications for Windows 95.
STATEMENT OF THE CASE
A. Novell's Claim at Trial
On November 10, 2004, Novell filed its Complaint in the District of
Utah, asserting six counts under Sections 1 and 2 of the Sherman Act, all based on
harm allegedly suffered by WordPerfect and Quattro Pro in the short period during
which Novell owned those applications (June 24, 1994 to March 1, 1996).1 In
April 2005, the Judicial Panel on Multi-District Litigation ("JPML") transferred
the action to the District of Maryland and assigned it to Hon. J. Frederick Motz.
In Count I, Novell alleged that Microsoft unlawfully maintained a
monopoly in the PC OS market by withdrawing support for the NSEs in 1994,
while Windows 95 was under development, thereby allegedly injuring Novell's
applications and harming competition in the adjacent PC OS market. (JA-119-25,
135-38.) This claim was not barred by the applicable four-year statute of
limitations (15 U.S.C. § 15b) because it was deemed sufficiently related to a 1998
equitable enforcement action brought by the U.S. Department of Justice (the
-4-
"Government Case") relating to the PC OS market to fit within the Clayton Act's
tolling provision (15 U.S.C. § 16(i)). Counts II through V, which alleged that the
same conduct caused harm to competition in the word processing and spreadsheet
applications markets--where WordPerfect and Quattro Pro actually competed--
were dismissed as untimely because they were not eligible for tolling. Novell v.
Microsoft, 2005 U.S. Dist. LEXIS 11520, at *9-14 (D. Md. June 10, 2005), aff'd,
505 F.3d 302 (4th Cir. 2007). Although the Complaint was filed ten years after the
allegedly wrongful conduct had occurred, Count I was not time-barred only
because it was "ingeniously designed" to recast Novell's claims as a unique cross-market claim for harm to the PC OS market. Novell v. Microsoft, 699 F. Supp. 2d
730, 736 (D. Md. 2010).2
Count I survived summary judgment, Novell v. Microsoft, 429 F.
App'x 254, 260-62 (4th Cir. 2011). On June 30, 2011, the JPML returned the
action to the District of Utah, and Judge Motz accepted an assignment to preside
over the trial. By 2011, Judge Motz had been on the federal bench for more than
25 years and, as the Microsoft MDL judge since 2000, had handled more than 100
actions against Microsoft.
-5-
Before trial, Novell sought and obtained collateral estoppel on 52
Findings of Fact ("FoF" or "Findings") entered in the Government Case, United
States v. Microsoft, 84 F. Supp. 2d 9 (D.D.C. 1999).3 Novell agrees that those
Findings are "binding in this case." (Novell Br. 6 n.2.)
B. Trial
Trial commenced on October 17, 2011 and concluded on
December 16, 2011. Novell called four live fact witnesses: Adam Harral and
Greg Richardson, former Novell programmers, Gary Gibb, former director of
PerfectOffice for Windows 95, and Robert Frankenberg, Novell's CEO during the
relevant period. Novell also played or read deposition testimony of 11 additional
witnesses, all of whom were current or former Microsoft employees, and called
three experts: Ronald Alepin on technical issues, Dr. Roger Noll on competition
issues, and Dr. Frederick Warren-Boulton on damages. Novell sought $4 billion in
damages (after trebling).
Microsoft called 13 fact witnesses (only two by deposition), including
Bill Gates, five other current or former Microsoft employees, and seven former
WordPerfect/Novell employees. Microsoft also called three experts: Dr. John
-6-
Bennett on technical issues, Dr. Kevin Murphy on competition issues, and
Dr. Glenn Hubbard on damages.
At trial, Novell's counsel read to the jury the collaterally estopped
Findings (JA-11072-86, 12616-38), which Judge Motz instructed had "binding"
effect (JA-11072).
At the close of Novell's case on November 17, Microsoft moved for
judgment as a matter of law. Judge Motz said that he was either "denying the
motion without prejudice" to its renewal "at the close of all the evidence" or was
deferring any ruling "until all the evidence is in." (JA-13881-82.) He expressed
doubt that Novell could "prove the case," but said "that as a practical matter"
the case should go to the jury. (JA-13881-82.) At the close of all the evidence,
Microsoft renewed its Rule 50 motion, and Judge Motz said that the case "is going
to the jury and then I'll focus upon [the motion] after that." (JA-16059-60.)
The jury was unable to come to a verdict. On December 16, 2011,
Judge Motz declared a mistrial.4
-7-
C. The Decision on Microsoft's Renewed Motion for Judgment as a
Matter of Law
After trial, Judge Motz granted Microsoft's renewed Rule 50 motion
on "three separate and independent" grounds (JA-211): (1) "Novell has not
created a jury question on the issue of whether Microsoft's conduct was
anticompetitive" (JA-215-18), (2) no reasonable jury could conclude that the
October 3 Decision had an adverse effect on competition in the relevant market
(JA-218-26) and (3) "the claim Novell asserts is a lawyers' construct and is not
based on an underlying business reality" (JA-226-28).
STATEMENT OF FACTS
A. Microsoft
In 1981, Microsoft released MS-DOS, an operating system with a
character-based interface that required specific user instructions to perform tasks
such as launching applications. (JA-1622 (FoF ¶6)5; JA-13649-51 (Gates).) MS-DOS became the dominant operating system for Intel-compatible PCs. (JA-1622
(FoF ¶6); see also JA-13657 (Gates).)
Bill Gates was not content to rest on the success of MS-DOS (JA-13658 (Gates)) and decided in the early 1980s to develop an operating system with
-8-
a graphical user interface ("GUI"), which would allow users to perform tasks by
selecting icons on the screen using a mouse (JA-200 (Decision); JA-1622 (FoF ¶7);
JA-13658-61 (Gates)). Microsoft tried to persuade ISVs, including WordPerfect
Corporation ("WPC"), to write applications for this new operating system, called
Windows. (JA-200 (Decision); JA-13663 (Gates); JA-15666 (Peterson).)
In May 1990, Microsoft released Windows 3.0, an operating system
that Novell's expert Noll described as a "revolutionary technological leap." (JA-12853.) Windows 3.0 became immensely popular and quickly caused a "sea
change" away from MS-DOS toward Windows. (JA-14540 (Larsen); see also JA-11977-79 (Frankenberg); JA-11755-57 (Gibb); JA-15047-48 (LeFevre); JA-15664-65 (Peterson).) In 1991, Microsoft began developing its next major operating
system, Windows 95, a project that Mr. Gates described as "one of the toughest
engineering tasks ever done." (JA-13704, 13709-10.)6
Meanwhile, Microsoft had also developed Word, a word processor,
and Excel, a spreadsheet. (JA-199 (Decision).) In 1990, Microsoft was the first
company to release a "suite" (a software package that combined a word processor,
spreadsheet and other applications). (JA-199 (Decision).)
-9-
By anticipating (indeed, creating) these two major shifts in the PC
software industry--from character-based to GUI operating systems, and from
standalone applications to suites--Microsoft's success continued to grow. From
1990 to 1995, MS-DOS and Windows together had more than 90% of the PC OS
market. (JA-1625 (FoF ¶35).) There is no dispute that by 1994 Microsoft had a
lawfully acquired monopoly in the PC OS market.7
Microsoft also became very successful in applications because, by
1994, "customers were buying suites rather than individual products" (JA-12004
(Frankenberg)) and, as a Novell memorandum conceded, Microsoft Office had
"the strongest combination [of applications] in the industry" (JA-7023 (DX7); see
also JA-9295 (DX271)). As of 1994-95, Microsoft Office was so popular that
standalone applications represented only about a quarter of total sales of
applications--suites were the rest. (JA-15401-02 (Hubbard); JA-1650.) Microsoft
had more than 80% of the suite segment in 1993-95 (JA-2893, 2895 (PX 138)),
and by April 1995, Novell pegged Microsoft's share at 86%. (JA-9295, 9298
(DX271).)
-10-
B. WordPerfect Corporation
In the 1980s, WPC developed WordPerfect, a word processing
application. (JA-199-200 (Decision).) WordPerfect was "the acknowledged 'king
of the hill' on the DOS platform" throughout the 1980s. (JA-199-200 (Decision).)
At trial, Pete Peterson, who ran WPC until 1992, testified that Bill Gates advised
him in 1989 "to write for Windows," but that WPC resisted and decided instead to
keep all its eggs in the old character-based basket. (JA-15628-29, 15666.) WPC
released its first version of WordPerfect for Windows in November 1991, about 18
months after Windows 3.0 came out. (JA-200 (Decision).)
The quality of WordPerfect for Windows was not good: Peterson
testified that the reviews were "lukewarm" and the product was "not as well
received as we had hoped." (JA-15635.) WordPerfect 6.0 for Windows 3.1, the
next version, was released in October 1993 and also was "perceived as a slow and
buggy product" (JA-9298 (DX271)) that had "too many bugs to be considered
sufficiently stable" (JA-9168 (DX259)). As a result, WordPerfect was never
highly popular on Windows8 and, by 1994, almost no one was buying MS-DOS
-11-
(where WordPerfect had been king). (JA-15047-48, 15054-55 (LeFevre); JA-11755-56 (Gibb); JA-15398-15400, 15516 (Hubbard).)
Further, WPC failed to release any suite until May 1993--almost
three years after Microsoft Office created the suite segment. (JA-11757 (Gibb);
JA-14847 (Acheson).) Frankenberg, Novell's CEO in 1994-95, conceded that
Microsoft had a "huge head start" in suites (JA-11997-99), and WPC/Novell
estimated that its own suite product (originally called Borland Office) had about a
2% share in 1993-94 (JA-9025 (DX223); JA-5345 (PX448)).
C. Novell
On March 21, 1994, Novell--which had no history of developing
office productivity applications (JA-11984-86 (Frankenberg))--announced that it
would acquire WPC and a spreadsheet called Quattro Pro developed by Borland of
Scotts Valley, California (JA-200 (Decision); JA-11973 (Frankenberg)). The deals
closed in June 1994. (JA-200 (Decision).)
In the two days following Novell's announcement, Novell's stock
price dropped by 20% (JA-11926 (Frankenberg))--a decline in Novell's market
capitalization of about $1.5 billion. (JA-12031 (Frankenberg); JA-15395
(Hubbard).) This was almost exactly equal to Novell's $1.55 billion purchase
price for WPC and Quattro Pro. (JA-12031 (Frankenberg); JA-15420 (Hubbard).)
-12-
At the time of the acquisitions, Frankenberg "recognized the need to
improve" WPC's products for the Windows platform and to develop a first-class
suite. (JA-201 (Decision); see JA-12003-04 (Frankenberg).) In December 1994,
Novell released PerfectOffice 3.0 for Windows 3.1, its first suite that received
positive reviews. (JA-11944, 11948 (Frankenberg); JA-4124-27 (PX297).)
Nevertheless, PerfectOffice 3.0 "did not achieve a sizeable market share" of suites
(JA-201 (Decision)), capturing at most 8% in early 1995 (JA-12858 (Noll)).
Novell sold WordPerfect and Quattro Pro to Corel in March 1996.
(JA-209 (Decision).) Corel released PerfectOffice for Windows 95 in May 1996,
nine months after Windows 95 came out. (JA-209 (Decision).)
D. Microsoft's "Evangelization" of Windows 95
During the development of Windows 95, Microsoft provided ISVs
with advance information about planned features in the new operating system,
code-named "Chicago." (JA-2337-52 (PX63); JA-2696-719 (PX113).) Novell's
witnesses agreed that WPC/Novell was "excited about Windows 95," which was a
"significant step forward," and that Novell wanted to "take[] advantage of the
capabilities in" Windows 95. (JA-12162-63 (Frankenberg); JA-11184-87 (Harral);
JA-11540 (Richardson); JA-11721-22, 11730 (Gibb).) Gary Gibb testified that
Novell knew that Windows 95 would become highly popular "pretty quickly" after
-13-
its release, so Novell's "number one priority" was to come out "as soon as
practicable" with a PerfectOffice suite for Windows 95. (JA-11730, 11773-76.)
Microsoft's Developer Relations Group ("DRG") encouraged WPC,
Novell and other ISVs to develop applications for Windows 95. (JA-14197-98
(Struss).) Brad Struss, who was in charge of this effort and was DRG's primary
contact with WPC/Novell, testified that his goal was to help WPC/Novell and other
ISVs "ship[] great Windows 95 applications shortly after Windows 95 shipped."
(JA-14202.) To this end, DRG provided ISVs (including WPC/Novell) with early
access to betas and development tools. (JA-14202-07 (Struss).)
8
In November 1993, Struss and others visited WPC to provide
assistance and urge WPC to be ready to release its applications "close to the time
Chicago shipped." (JA-2666 (PX105); JA-14213-14, 14216 (Struss).) Microsoft's
representatives then gave WordPerfect programmers, including Adam Harral,
information about plans for Chicago, including a new user interface that
applications could modify, using new shell extension APIs, a small subset of which
-14-
were the NSEs. (JA-2666 (PX105); JA-11218-19 (Harral); JA-14707 (Nakajima).)
The NSEs were four APIs out of "thousands" in the new operating system. (JA-12397 (Alepin); see also JA-11266 (Harral).) The NSEs were meant to provide a
means for ISVs to add their own custom namespaces to the Windows Explorer.10
(JA-11227-29 (Harral); see also JA-11596-97 (Richardson); JA-12514 (Alepin).)
11
There were other mechanisms for adding folders to Windows Explorer to display
word processing documents and spreadsheets, and those mechanisms did not rely
on the NSEs. (JA-13714-15 (Gates); JA-14704-07 (Nakajima); JA-15200
(Belfiore); JA-12518 (Alepin).)
In December 1993, Microsoft specifically warned WPC and other
ISVs in writing that the namespace extensions were "[n]ot for most applications!"
and "should NOT [be used to] edit documents with an explorer extension!" (JA-2715 (PX113).) This 1993 document, which Novell had in its files, warned Novell
not to use the NSEs to edit documents in a word processor like WordPerfect. (JA-11225 (Harral); JA-11539-40 (Richardson).)
-15-
E. Microsoft's Beta Version of Windows 95
On June 10, 1994, Microsoft provided WPC, Novell and thousands of
other ISVs with a beta version of Windows 95 called the Milestone 6 ("M6") Beta.
(JA-201 (Decision); JA-11366-37 (Harral).) The M6 Beta was the first beta
provided to Novell/WPC that contained the NSEs, and documentation for the
NSEs was provided at the same time. (JA-3385-404 (PX181); JA-11234-35
(Harral); JA-11557-58 (Richardson).)
12
The M6 Beta was provided pursuant to a Microsoft license agreement,
which both WPC and Novell signed. (JA-7177 (DX18); JA-7181 (DX19).)
The license agreement stated: (a) "This PRODUCT consists of pre-release code,
documentation and specifications and is not at the level of performance and
compatibility of the final, generally available product offering"; (b) the operating
system "may be substantially modified prior to first commercial shipment"; and
(c) the user "assumes the entire risk" of use of the beta. (JA-7175 (DX18); JA-7178 (DX19); see also JA-202 (Decision).) Likewise, the front cover of the
"Chicago Reviewer's Guide," which came with the M6 Beta, stated that the
-16-
"features and functionality present" in the beta did "not represent a commitment on
the part of Microsoft for providing or shipping [those] features and
functionality . . . in the final retail product offerings of Chicago." (JA-4696
(PX388).)
Novell's CEO, developers and other employees all testified that they
understood in 1994-95 that Microsoft might modify, and was entitled to modify,
beta versions of Windows 95. (JA-202 (Decision).) Frankenberg testified that
Novell understood that betas "could change" and "might change" prior to
commercial release of the final product. (JA-12138, JA-12146.) Nolan Larsen,
Novell's Director of Human Factors, testified that "the definition of beta" is that
"there can be and almost certainly will be changes." (JA-14557; see also JA-14608-12.) Dave LeFevre, Novell's Director of Marketing for PerfectOffice for
Windows 95, agreed that a "[b]eta by definition is an early release or a prerelease
of a product that is subject to change" and that "there is no promise that what is in
an early beta or even a late beta will be in the final product." (JA-14986-87.)
Novell's expert witnesses Noll and Alepin conceded this same point. (JA-12821;
JA-12494-95; see also JA-14075-76 (Gates); JA-15194-95 (Belfiore); JA-14209
(Struss).) Novell's "deception" argument is eviscerated by these facts; Novell was
-17-
fully aware that Microsoft made no commitment to include the NSEs in the final
product.
Novell's own practices reinforce the same point. An October 18,
1994 memorandum prepared by Novell's Corporate Development Group
recognized that, for Novell's own software, "the product and features may change
dramatically" during the beta phase, and that an "entire feature" might be
"remov[ed]" from a beta version. (JA-10347, 10349 (DX612A).) Novell's
"Software Developer's Kit" (distributed with beta versions of its software)
expressly warned ISVs that "Novell does not guarantee that Beta Products will
become generally available to the public or that associated products will be
released" and that "[t]he entire risk arising out of your use of Beta Product remains
with you." (JA-10352 (DX618).)
F. The October 3 Decision
Microsoft was concerned in 1994 about a flaw in the NSEs that made
it possible for poorly designed third-party applications calling those APIs to crash
Windows 95. (JA-13731-32 (Gates); JA-14339-40 (Muglia); JA-15225-27, JA-15235 (Belfiore).) As Judge Motz said, even "Novell's own technical expert,
Ronald Alepin, conceded[] an ISV that wrote a bad program using the namespace
extension APIs could cause Windows 95 to crash." (JA-202; JA-12540-41
(Alepin).)
-18-
As a result, on October 3, 1994, less than four months after release of
the M6 Beta and more than ten months before the commercial release of
Windows 95, Bill Gates decided to withdraw support for the NSEs, meaning that
Microsoft would no longer publish documentation for those four APIs and would
tell ISVs that Microsoft might change or remove the NSEs in future versions of
Windows.13(JA-13761-62, 13972, 14020 (Gates).)
14 The NSEs were not
removed; they remained, as Novell then understood, in Windows 95 and could be
called by Novell's applications. (JA-11557-58 (Richardson); JA-11273, 11462
(Harral).)
-19-
G. The October 3 Decision Did Not Harm Novell
As Judge Motz stated, "Microsoft had no reason [in October 1994] to
believe that Novell was relying upon" the NSEs. (JA-217.) Gates testified that
when he made his decision, he had no information that Novell was planning to use
the NSEs. (JA-13761.) There is no contrary evidence.
The contemporaneous documents further show that "shortly before
the" October 3 Decision, "Struss of Microsoft was advised that Novell was not yet
using the namespace extension APIs." (JA-217 (Decision).) In September 1994,
Struss, who was in frequent contact with Tom Creighton, Director of Novell's
shared code group (and Harral's boss) (JA-14204-05 (Struss)), surveyed major
ISVs and reported on September 22 that the "specific feedback" from
Novell/WordPerfect was that "[t]hey have not begun any work on IShellFolder,
IShellView, etc."--i.e., the NSEs. (JA-7167 (DX17); JA-14218-21 (Struss).)
15
Struss testified that Novell told him in September that it was "not using" the NSEs
and was "not dependent upon" them. (JA-14222.) As Judge Motz noted, "Novell
did not rebut this evidence in any way." (JA-217.)
-20-
Microsoft promptly notified Novell of the October 3 Decision. On
October 12, 1994, Struss reported in writing that Novell (he wrote "WP," for
WordPerfect) appears "to be OK" with the decision. (JA-6926 (DX3); JA-14223-24 (Struss); see also JA-217 (Decision).) Struss testified that he personally
informed Novell about the change. (JA-14225.) "Novell presented no witness to
rebut this testimony or to contradict what was said in the October 12 email." (JA-206 (Decision).)
Additionally, "there is no evidence that anyone at Novell made any
complaint to anyone at Microsoft who could have reversed the decision to
withdraw support for the namespace extension APIs." (JA-227 (Decision).) Struss
testified, and his contemporaneous written reports show, that Novell made no
complaint. (JA-14228-30, 14233-34 (Struss); JA-6924-25 (DX2); JA-7637-38
DX92).)
16
Likewise, Gates testified that he attended "lots of meetings with ISVs
in late 94, early 95" but that neither Novell nor any other ISV said "that they were
using" the NSEs or that the October 3 Decision was an "issue[]" for them. (JA-
-21-
13765-66.) During a January 10, 1995 meeting between the two CEOs,
Frankenberg complained vociferously to Gates about other Microsoft conduct but,
as Frankenberg testified--and the eight pages of meeting minutes (JA-10451-58
(DX636)) confirm--he never complained about the NSE decision (JA-12057-58).
Although Novell now says that Frankenberg "complained `on a
number of occasions' to Bill Gates in 1995 about the issue of undocumented APIs"
(Novell Br. 58), Novell fails to mention that, as Frankenberg conceded, these
complaints were about a different topic--"undocumented APIs as a general
matter," meaning APIs that were never documented in the first place (JA-11963-64
(Frankenberg))--and that Frankenberg never specifically raised or complained
about the NSEs or the October 3 Decision (JA-12054-55 (Frankenberg)). As
Frankenberg admitted, he never told Gates that Novell had any "problem" with the
decision that Novell's lawyers claimed at trial completely ruined Novell's
applications business. (JA-12054-55.)
17
-22-
In sum, "[t]here is no evidence to support th[e] premise" that
"Microsoft withdrew support for the namespace extension APIs because it knew
that Novell was using those APIs in the development of its applications and that,
by withdrawing support for those APIs, Microsoft knew that Novell would fall
behind schedule." (JA-216 (Decision).) This is devastating to a plaintiff whose
lawyers argue was "deceived" by Microsoft's decision to do what the license
agreement (and industry practice) specifically permitted: make a change to a beta
before releasing the final product.
H. Microsoft's Continued Cooperation with Novell
After October 1994, Microsoft continued to provide assistance to
Novell in developing applications for Windows 95. Judge Motz, citing testimony
of Frankenberg and other Novell employees, noted that even "after it withdrew
support for the namespace extension APIs, Microsoft continued to provide
assistance to Novell and never terminated their relationship." (JA-218 (citing JA-14205-06, 14211-21 (Struss), JA-6925 (DX2), and JA-12067 (Frankenberg)).) For
example, on October 21, 1994, Struss wrote that he was "[w]orking with" Novell's
"sr. management" to get them to place "more focus on their" Windows 95 product
(rather than the old PerfectOffice for Windows 3.1 product). (JA-6925 (DX2).)
Struss testified that every time he spoke to Creighton, Struss "would ask him how
could we help them" to "move forward." (JA-14229.)
-23-
Frankenberg acknowledged that Microsoft "endeavored to be helpful
to Novell" so that it could "produce a great application for Windows 95." (JA-12067, 12154.) LeFevre testified that in 1994-95 (when he worked at Novell), a
Microsoft employee was assigned full-time "to support us [Novell] in this
development effort." (JA-14984.) LeFevre and Creighton also "spent an entire
day" at Microsoft's headquarters "meeting with the development team for
Windows" who "answer[ed] some critical questions" for Novell. (JA-14984-85
(LeFevre).)
Similarly, in April 1995, Novell's Scott Nelson wrote to Novell
executives that "the cooperation between Microsoft and Novell has been very
good" and that "over the next couple of weeks our developers and testers will visit
Redmond [Microsoft's headquarters] once again to make sure that we are making
continued progress." (JA-8654 (DX172).)
I. Novell Had Three Options and Chose the Most Difficult
At trial, Novell claimed that Microsoft's October 3 Decision caused
delay because Novell's shared code group
18 (including Harral and Richardson)
spent a full year replicating the functionality of the NSEs inside Novell's custom
-24-
file open dialog ("FOD")
19 and argued that the shared code group's slowness
caused delay in the release of its applications. (See Novell Br. 25.) As Harral,
Richardson and Gibb conceded, in October 1994, the shared code group had at
least three development options: "(1) to continue to write its own programs using
the namespace extension APIs that were no longer supported by Microsoft but
were still accessible to ISVs in possession of the beta version" (Option 1); "(2) to
use Microsoft's common file open dialog . . . which remained available to ISVs
after the withdrawal of support for the namespace extension APIs" (Option 2)
20; or
"(3) to write its own customized file open dialog" (Option 3). (JA-204 (Decision)
(citing JA-11273-75 (Harral); JA-11535-37, 11608 (Richardson), and JA-11781-83
(Gibb)).) Although Novell believed in 1994 that it needed to have a version of
PerfectOffice released "within no more than a two to three months of Chicago's
ship date" (JA-8853-54 (DX202)), Novell's programmers chose Option 3--the
most difficult and time-consuming route by far (JA-204-05 (Decision)).
Novell's programmers, Harral and Richardson, both conceded that
Novell could have used Option 1--that "calling these API's was simple" because
-25-
"[w]e had the documentation" and "[w]e knew how to call them." (JA-11273; JA-11608.) Both testified that by October 1994 Novell was nearly finished doing
exactly that (writing a FOD that called the NSEs). (JA-11609-12, 11620-21; JA-11368-69.) Although Option 1 was not ideal (Novell Br. 23) because of the risk
that in later years Microsoft might remove the NSEs from Windows, it was a real
choice (JA-12069-70 (Frankenberg)).
As to Option 2, Gibb testified that it would have been "easy" for
Novell to release its applications using the Windows 95 FOD, which Microsoft
provided to all ISVs for free. (JA-11781-82.) According to Harral, Novell
considered Option 2 "many times" because it was "an easier option than the third."
(JA-11297-98.) As Judge Motz pointed out, "if, as Novell now argues, the 90-day
period after the release of Windows 95 was critical to the success of WordPerfect,
Quattro Pro, and Perfect Office, Novell could have released those products using
Microsoft's common file open dialog." (JA-228.) Harral agreed that "Novell
could have come out with a product in '95 that utilized the Windows common file
open dialog." (JA-11434.)
Others at Novell also advocated for Option 2. (JA-228 (Decision).)
Karl Ford, Lead Developer for the User Interface in WordPerfect for Windows 95,
testified that in 1995 he told Gibb and Steve Weitzel, Director of WordPerfect for
-26-
Windows, that "the safest route" was to use "the common open dialog" in
Windows 95 "if they were concerned about schedule." (JA-14664-65.) Novell
even discussed using "the common open dialog right now" and including a custom
FOD later, "in the next release." (JA-14664 (Ford).) Dave LeFevre similarly
testified that during 1995 he "became an advocate" within Novell for Option 2.
(JA-14996-97.)
Novell nevertheless decided to choose the riskiest path--the time-consuming third option of writing its own "customized" FOD. (JA-205
(Decision).) As Gibb testified, he chose this path because he wanted to "do
something cooler" and "exceed what was the default stuff." (JA-11782-83.) In
short, as Gibb recognized, Novell was "faced with the age old trade off" between
"get[ting] out a product more quickly and sacrific[ing] features," or, on the other
hand, "delay[ing] until 1996 and try[ing] to build a cooler product." (JA-11825-26.) Both Harral and Gibb conceded that Novell could have released PerfectOffice
in time using Microsoft's common FOD. (JA-11434-36; JA-11781-82.)
Further, as Judge Motz stated, "there is no evidence that any top-level
executive at Novell was involved in the decision-making process" concerning
which option to choose. (JA-206.) Frankenberg testified that any action that could
jeopardize the timely release of Novell's applications would have been referred to
-27-
some or all of four senior executives (JA-12076-78, 12116-17; see JA-206
(Decision)), but none of the four was involved in or even aware of this decision
(JA-12118-19 (Frankenberg)). "Instead, Novell assigned responsibility for
supervising the writing of code for a custom file open dialog . . . to Gibb, a middle
manager, who, in turn, assigned the responsibility for writing the code to Harral
and Richardson." (JA-206-07 (Decision).)
21
Judge Motz also noted Frankenberg's testimony "that in any business
or organization faced with an important decision, a formal memorandum would
normally be presented to the senior executives" so they could choose. (JA-206
(Decision) (citing JA-12118 (Frankenberg)).) Strikingly, no such document exists.
"Novell presented no evidence that any such memorandum was ever written, and
Frankenberg testified that he knew of no evidence that [any senior executives]
were presented with a decision about how to respond to Gates's decision to
withdraw support for the namespace extension APIs." (JA-206 (Decision).) Had
Microsoft's decision truly threatened to cause the immense injury claimed at trial,
senior executives at Novell, including Frankenberg (who knew nothing of the three
-28-
Options (JA-12063, 12068-70 (Frankenberg))) would have been involved and
would surely have protested to Microsoft. Judge Motz hit the nail on the head by
calling the case a "lawyers' construct."
J. Quattro Pro Caused the Delay
The premise for Novell's claim is that, if not for Microsoft's conduct,
it would have released PerfectOffice within 60-90 days of the release of
Windows 95. (JA-10963-64 (Novell Opening Statement); JA-13364 (Warren-Boulton); Novell Br. 22.) The overwhelming evidence at trial established that
Quattro Pro, a necessary component of PerfectOffice, was the real reason for delay.
As Judge Motz stated, Quattro Pro "was not ready for release to manufacturing
until 1996." (JA-228.)22
Even in late 1994, Novell recognized that a September 30, 1995
release date for PerfectOffice for Windows 95 (codenamed "Storm") was "barely
achievable" and that the Quattro Pro team needed "an additional three months"
(i.e., until December 30) to be ready. (JA-8893 (DX211).) On March 1, 1995, a
Novell VP wrote that the RTM date (release to manufacturing) for Storm was
being pushed "back by one month (to December 30)" because "realistic[ally]"
Quattro Pro could not be ready until then. (JA-8988 (DX221); JA-207 (Decision).)
-29-
December 30 is more than four months after August 24 (the release date of
Windows 95). Thus, as of March 1, 1995, Novell recognized that it could not
release PerfectOffice for Windows 95 within what its lawyers say was the
"critical" 60-90 day window because of Quattro Pro, not any delay attributable to
Microsoft.
In March and May 1995, Novell prepared formal memoranda that
listed "risks" to the PerfectOffice project. Those memos ranked "Quattro Pro
delivering late" as the highest "OVERALL RISK" for PerfectOffice. (JA-9059
(DX223); JA-9116 (DX226); JA-208 (Decision).) Novell's LeFevre testified that
"the product that was causing the biggest problem was Quattro Pro," and Ford said
the same thing. (JA-208 (Decision) (quoting 15000-02 (LeFevre)); JA-14653-54
(Ford).)
As Judge Motz recognized, "[t]he problem of having Quattro Pro
ready to be released to manufacturing became even more severe toward the end of
1995, when many developers left Quattro Pro for other employment in Silicon
Valley, the area where Scotts Valley--home of Quattro Pro--was located." (JA-208 (citing JA-9145 (DX230)).) On December 23, 1995, Novell's VP Brereton
wrote to Frankenberg to say that "this past Thursday/Friday, about 15 additional
people submitted their resignations," leaving the Quattro Pro development team
-30-
with "just 2 people." (JA-9145 (DX230); JA-208 (Decision).) When Frankenberg
was shown this e-mail on the witness stand, he quickly conceded that "clearly the
product wasn't complete" even on December 23, 1995. (JA-12081.)
Novell sent Nolan Larsen to Scotts Valley in January-February 1996
to help get Quattro Pro ready. (JA-14573-75 (Larsen).) Larsen testified that when
he arrived, the whole scene was "kind of a train wreck" and that he and others
desperately tried to finalize Quattro Pro but were unable even to create an interim
build. (JA-14573-78; JA-208 (Decision).) Quattro Pro was "[n]ot by any stretch
of the imagination" ready to be released even in January 1996. (JA-14578
(Larsen).) Larsen's testimony was backed up by Novell's International Marketing
Manager, Craig Bushman, who testified that the resignation of Quattro Pro
developers just before Christmas was "stunning" and "a death blow," and that
"there was still ongoing issues throughout that spring [of 1996] also." (JA-14144-45.)
Novell never released any applications for Windows 95. It sold them
to Corel in March 1996 (JA-10246 (DX382)) and Corel released PerfectOffice for
Windows 95 in May 1996 (JA-11541 (Richardson); JA-11738 (Gibb)).
-31-
SUMMARY OF ARGUMENT
The trial court was correct in granting judgment as a matter of law on
each of "three separate and independent" grounds. (JA-211.)
First, as a matter of law, the October 3 Decision was not
anticompetitive conduct under Section 2 of the Sherman Act. The only exception
to the longstanding Colgate rule that a company, even a monopolist, is free to
choose with whom it will deal, or who it will assist, is Aspen Skiing v. Aspen
Highlands Skiing, 472 U.S. 585 (1985). Novell's claim is nowhere close to
qualifying under Aspen. Indeed, this Court recently rejected two claims that were
much closer to Aspen than Novell's, holding that (1) an Aspen claim fails where
plaintiff was aware "that the relationship was temporary and subject to [the
defendant's] business judgment," Christy Sports v. Deer Valley Resort, 555 F.3d
1188, 1197 (10th Cir. 2009), and (2) Aspen is available only to a plaintiff that
proves, among other things, that the monopolist's conduct amounted to the
"termination" of a profitable, long-term relationship, Four Corners Nephrology
Associates v. Mercy Medical Center of Durango, 582 F.3d 1216, 1224-25 (10th
Cir. 2009). It is undisputed that Novell was fully aware (as the license agreement
stated) that the M6 Beta was subject to change. Further, Microsoft did not
terminate its relationship with Novell because (i) Novell could have used
Microsoft's common FOD to get its applications to market in a timely fashion, and
-32-
(ii) Microsoft continued to provide information and assistance to Novell after the
October 3 Decision. Moreover, Novell cannot assert a Section 2 claim based on
supposed "deception" because there was no deception and, in any event, deception
of a competitor cannot give rise to an antitrust claim.
Second, no reasonable jury could conclude that any delay in the
release of Novell's applications could have adversely affected competition in the
PC OS market. Both of Novell's theories of cross-market harm lacked evidentiary
support and were refuted by the binding Findings and by market realities. There
was insufficient evidence that competition in the operating system market (where
Microsoft had more than a 90% share) would have been enhanced had
PerfectOffice or WordPerfect come to market in Fall 1995 rather than 1996.
Third, Novell's claim was a "lawyers' construct" (JA-227 (Decision))
because, among other reasons, Novell had satisfactory options for the timely
release of its applications and Novell was late to market because of Quattro Pro,
not any conduct by Microsoft.
STANDARD OF REVIEW
This Court "review[s] de novo the district court's decision on a Rule
50(b) motion for judgment as a matter of law." Burrell v. Armijo, 603 F.3d 825,
832 (10th Cir. 2010). To survive a Rule 50 motion, a plaintiff must present
-33-
"substantial evidence" in support of its case. Webco Industries v. Thermatool, 278
F.3d 1120, 1128 (10th Cir. 2002). "Substantial evidence is something less than the
weight of the evidence, and is defined as such relevant evidence as a reasonable
mind might accept as adequate to support a conclusion, even if different
conclusions also might be supported by the evidence." Id. at 1128. Under this
standard, "the question is not whether there is literally no evidence supporting the
nonmoving party but whether there is evidence upon which a jury could properly
find for that party." Specialty Beverages v. Pabst Brewing, 537 F.3d 1165, 1175
(10th Cir. 2008). "For a jury to properly find for a party, the party must present
more than a scintilla of evidence supporting its claim." Herrera v. Lufkin
Industries, 474 F.3d 675, 685 (10th Cir. 2007).
ARGUMENT
I. THE TRIAL COURT CORRECTLY HELD THAT THE OCTOBER 3
DECISION WAS NOT ANTICOMPETITIVE CONDUCT.
The Sherman Act "does not restrict the long recognized right of [a]
trader or manufacturer engaged in an entirely private business, freely to exercise
his own independent discretion as to parties with whom he will deal." United
States v. Colgate, 250 U.S. 300, 307 (1919). Even a company with monopoly
power "has no antitrust duty to deal with its competitors" and "certainly has no
duty to deal under terms and conditions that the rivals find commercially
-34-
advantageous." Pacific Bell Telephone v. Linkline Communications, 555 U.S. 438,
449-50 (2009). The sole exception to the Colgate doctrine, where a duty may be
imposed on a firm to deal with its rival, is Aspen Skiing v. Aspen Highlands Skiing,
472 U.S. 585 (1985).
This Court has dealt with the proper parameters of an Aspen claim in
Christy Sports v. Deer Valley Resort, 555 F.3d 1188 (10th Cir. 2009), and Four
Corners Nephrology Associates v. Mercy Medical Center of Durango, 582 F.3d
1216 (10th Cir. 2009). These recent decisions establish that a company's refusal to
deal with a competitor may expose it to Section 2 liability only where
(a) defendant "terminated a profitable business relationship" with plaintiff, thereby
denying it terms "available to all other consumers," Four Corners, 582 F.3d at
1225 (emphasis in original), and (b) the terminated relationship was not
"temporary" or "subject to [defendant's] business judgment," Christy, 555 F.3d at
1197. Here, the evidence is undisputed that (a) even after the October 3 Decision,
Microsoft did not terminate its relationship with Novell and continued to provide
assistance to Novell in developing applications for Windows 95, (b) Novell knew
from the license agreement and otherwise that the M6 Beta was subject to change,
and (c) the October 3 Decision applied equally to all companies, not just Novell.
-35-
Remarkably, Novell fails even to cite Four Corners, and tries to avoid
the result mandated by Christy by characterizing its claim as one for deception or
inducement. This helps Novell not at all. The only allegedly anticompetitive act is
Microsoft's decision to withdraw support for the NSEs that were included in the
M6 Beta. If a claim for deceptively refusing to deal allowed an antitrust plaintiff
to evade the Aspen requirements, there would be an entirely new exception to
Colgate that has never before been recognized. On the contrary, the Supreme
Court has been "very cautious in recognizing" any exceptions to the Colgate
doctrine, and even the Aspen exception "is at or near the outer boundary of § 2
liability." Verizon Communications v. Law Offices of Curtis V. Trinko, 540 U.S.
398, 408-09 (2004). In any event, Novell was not deceived and deception of a
competitor cannot, as a matter of law, give rise to an antitrust claim.
A. Novell Was Aware from the Start that the Windows 95 Beta
Could Change.
In Christy, plaintiff for many years had operated a ski rental business
from premises it rented from defendant in the mid-mountain area of defendant's
Deer Valley resort. 555 F.3d at 1190-91. Plaintiff's lease with defendant
contained a restrictive covenant that forbade plaintiff from renting skis from the
property unless defendant gave its consent every year. Id. at 1190, 1196. For
almost 15 years, defendant allowed plaintiff to continue renting skis, but when
-36-
defendant opened its own ski rental store in the mid-mountain area, it decided that
it did not want to face competition from plaintiff--it wanted 100% of the
business--and enforced the restrictive covenant, terminating plaintiff's ability to
compete. Id. at 1190-91.
This Court rejected plaintiff's Aspen claim because the restrictive
covenant made clear to plaintiff "that the relationship could change at any time,"
and that plaintiff "should have been aware that the relationship was temporary and
subject to [defendant's] business judgment." Id. at 1197. The Court stated that "an
initial decision to adopt one business model" should not "lock" a defendant "into
that approach and preclude adoption of the other at a later time." Id. at 1196.
Even if a sudden "change" in approach (such as enforcing the covenant in Christy)
might amount to a "business tort," it is not anticompetitive conduct under the
Sherman Act. Id. at 1196-97; accord Intergraph v. Intel, 195 F.3d 1346, 1364,
1366 (Fed. Cir. 1999).
Novell's argument--that Microsoft "encourag[ed] Novell to develop
software that depended upon" the NSEs "and then withdr[ew] support" for them
(Novell Br. 32-33)--is far weaker than the claim in Christy, where defendant's
sudden decision to enforce the restrictive covenant after 15 years of non-enforcement put plaintiff out of business. 555 F.3d at 1190-91. WPC/Novell
-37-
received the M6 Beta and its associated documentation in June 1994, less than four
months before the October 3 Decision, pursuant to a license agreement which said
that the beta was "pre-release code" that could "be substantially modified prior to
first commercial shipment." (JA-7175 (DX18); JA-7178 (DX19).) Novell was
also told in writing that the M6 Beta did "not represent a commitment on the part
of Microsoft for providing or shipping the features and functionality" in the "final
retail product." (JA-4696 (PX388).) Novell offered no evidence that Microsoft
made some contrary promise.
Novell's current argument that Microsoft "induced Novell's reliance"
on the NSEs when Struss and other Microsoft representatives visited WordPerfect
in November 1993 (Novell Br. 32) also makes no sense. First, the meeting report
on which Novell relies never refers specifically to NSEs, but only to "shell
extensions" (JA-2666-67 (PX105)), which (aside from the NSEs) were unaffected
by the October 3 Decision. The shell extensions as a whole provided much more
functionality than the NSEs. (JA-7558-76 (DX72); JA-11556-57 (Richardson).)
Second, the November meeting took place a month before Microsoft warned
Novell in writing that the NSEs were "[n]ot for most applications!" and
particularly not to be used to "edit documents" by a word processor. (JA-2715
(PX113).) See p. 15, supra . Third, Harral testified that Novell could not--and did
-38-
not--begin work on the NSEs until it received the M6 Beta in June 1994. (JA-11367.) By then, Novell and WPC had executed the license agreements (JA-7177
(DX18); JA-7181 (DX19)) and had received the written disclaimer accompanying
the M6 Beta. See pp. 16-17, supra . This destroys Novell's inducement argument.
Given that the November 1993 visit to WPC was more than seven
months before Microsoft provided the M6 Beta to Novell, it could not have been
possible, let alone reasonable, for Novell to rely on this early presentation before it
had the beta with the NSEs. Microsoft often told ISVs about possible new features
even before any code was written but never committed that those features would be
included in the final product. (JA-13707-08 (Gates).) As Struss testified,
Microsoft was "careful" in its presentations "not to promise [ISVs] that something
would be part of" Windows 95. (JA-14239.) Novell has no evidence of any
promise.
Nor can Novell twist the facts into a claim for deception. As
Frankenberg testified, Novell understood that Windows 95 "might change" and
"could change" any time before final release. (JA-12138, 12146.) See pp. 17-18,
supra . Novell itself contemporaneously warned its own ISVs that "[t]he entire risk
arising out of your use of Beta Product remains with you." (JA-10352 (DX618).)
-39-
Novell argues that "the Court recognized [in Christy] the possibility
that the change in prior practice `could give rise to an antitrust claim' if 'by first
inviting an investment and then disallowing the use of the investment the resort
imposed costs on a competitor that had the effect of injuring competition in a
relevant market.'" (Novell Br. 32.) As an initial matter, Novell truncates the full
quote from Christy, which merely stated that the Court would not "preclude the
theoretical possibility" that such a claim could exist under different facts. 555 F.3d
at 1196. Moreover, the very same paragraph of the opinion makes clear that where
plaintiff "knew from the beginning" that the business relationship was subject to
change and therefore could not "claim unfair surprise," id., no valid claim exists.
Here, in light of the license agreement, the written guide accompanying the M6
Beta, and the concessions from Novell's CEO, Novell cannot possibly claim any
"unfair surprise" or deception.
In addition, Microsoft's October 3 Decision to make changes in a beta
was consistent with software industry practice and thus was not anticompetitive
conduct. In Telex v. IBM, 510 F.2d 894, 925-26, 928 (10th Cir. 1975), this Court
held that if defendant's practices are "typical of those used in a competitive
market" and "prevalent throughout the industry," then no valid antitrust claim
exists. Accord United States v. Syufy Enterprises, 903 F.2d 659, 668-69 (9th Cir.
-40-
1990); Olympia Equipment Leasing v. Western Union Telegraph, 797 F.2d 370,
375 (7th Cir. 1986).
As the trial court correctly concluded, "because of the undisputed
industry practice, Microsoft had no reason to believe that Novell was relying upon
the documentation for the namespace extension APIs contained in the beta version
Microsoft released to ISVs." (JA-217.) Indeed, the overwhelming evidence was
that the software industry--including Novell--widely recognized that a software
developer might make changes to beta versions of products under development.
Seven witnesses at trial, including two former Novell employees and two of
Novell's own experts, so testified. (JA-14557, 14561 (Larsen); JA-14985-88
(LeFevre); JA-12494-95 (Alepin); JA-12821-23 (Noll); JA-14075-76 (Gates); JA-15194-95 (Belfiore); JA-14209 (Struss).) Notably, as Frankenberg acknowledged
(JA-12139-42), Novell followed this same practice (JA-10349 (DX612A)
("features may still change dramatically during" the beta process)). See p. 18,
supra . Under Telex, Novell cannot base a claim of anticompetitive conduct on a
widespread industry practice.
-41-
B. Microsoft Never Terminated a Profitable Business Relationship
with Novell.
As this Court noted in Four Corners, the "key fact" in Aspen was that
"the monopolist was willing to jettison a profitable short-term business
relationship." 582 F.3d at 1225; accord Trinko, 540 U.S. at 409.
Here, as Judge Motz stated, "after it withdrew support for the
namespace extension APIs, Microsoft continued to provide assistance to Novell
and never terminated their relationship." (JA-218.) Novell never challenges this
point, and its CEO Frankenberg testified that he was "sure" that after October 1994
Microsoft continued to help Novell develop its applications for Windows 95. (JA-12067.) Moreover, Microsoft was "[w]orking with [Novell's] sr. management to
see about getting more focus on their [Windows 95] release" (JA-6925 (DX2)),
and in 1994-95 a Microsoft employee was assigned full-time to assist Novell in its
development efforts (JA-14984 (LeFevre)). See pp. 23-24, supra .
Novell does not dispute that it could have released its applications at
any time by choosing to use the Windows 95 common FOD, but merely argues that
this option was not commercially desirable. (Novell Br. 22-25.) If a large
software company23 makes a conscious decision to reject an available option for
-42-
getting its product to market and instead decides to risk everything on trying to
build a "cooler" product, the resulting delay cannot logically give rise to an
antitrust claim. In any event, rejecting the option to use Microsoft's technology
means that Novell chose its own fate--Microsoft did not terminate the
relationship.24
C. The October 3 Decision Did Not Treat Novell Differently.
As this Court pointed out in Four Corners, it was critical in Aspen that
"the monopolist was willing to jettison a profitable short-term business relationship
and deny to a rival the retail prices available to all other consumers." 582 F.3d at
1225 (emphasis in original). Similarly, in MetroNet Services v. Qwest, 383 F.3d
1124, 1134 (9th Cir. 2004), the Ninth Circuit held that plaintiff "d[id] not have an
actionable antitrust claim" because, among other reasons, defendant "has not
refused to deal with [plaintiff] on the same terms that it deals with direct
consumers."
-43-
Here, Novell was treated the same as all other ISVs. It is undisputed
that Microsoft (a) made the NSEs available to all third-party developers in June
1994 on the same terms as to Novell, and (b) withdrew support for the NSEs in
October 1994 for all third-party developers. (JA-6924-25 (DX2); JA-6926, 6928
(DX3).) This is dispositive under Four Corners, which requires a plaintiff to
establish that it was denied terms available to "all other[s]." 582 F.3d at 1219.
D. Novell Cannot Avoid the Requirements of Aspen by Recasting Its
Claim as One for Affirmative Conduct or Deception.
Novell attempts to avoid the requirements of an Aspen claim by
making two illogical and erroneous arguments.
First, Novell argues that it need not meet the Aspen requirements
because Microsoft's conduct was "affirmative" rather than a refusal to deal.
(Novell Br. 31-32.) This makes no sense. Withdrawal of support for the NSEs is
no different than the "affirmative" decision to enforce the restrictive covenant in
Christy or the "affirmative" decision to revoke plaintiff's hospital credentials in
Four Corners.
Novell tries to maneuver around this problem by saying that the
October 3 Decision "involved the intentional inducement of reliance." (Novell
Br. 35.) As the trial court recognized in rejecting this argument, "[a] decision not
to publish the namespace extension APIs in the first place is as 'intentional' as a
-44-
decision to withdraw support for" them later. (JA-215.) A monopolist generally
has no duty to cooperate with or assist a competitor whether the decision is
"intentional" or otherwise. Further, there is no evidence of any "intentional"
decision to hurt Novell--as Judge Motz observed, "[t]here is no evidence" that
"Microsoft . . . knew that Novell was using those APIs" or that the withdrawal of
support for the NSEs would cause "Novell [to] fall behind schedule." (JA-216.)
Moreover, Novell's reliance on Multistate Legal Studies v. Harcourt
Brace Jovanovich, 63 F.3d 1540 (10th Cir. 1995) (Novell Br. 31-32) is misplaced.
There, defendant, without any history of cooperation with plaintiff, deliberately
changed its products and created scheduling conflicts to prevent customers from
using its rival's bar review course. Id. at 1544-45, 1550-54. Defendant did not
refuse to deal with plaintiff, nor did the case involve a termination of an existing
business relationship. Id.
Importantly, where a case involves a refusal to cooperate or deal with
a rival, courts impose the Aspen requirements to protect from "forever lock[ing]
[the parties] into a business decision," Christy, 555 F.3d at 1198, and to avoid
invasive judicial review of voluntary dealings between rivals, see MetroNet, 383
F.3d at 1133. As the Supreme Court has stated, the limits of Aspen stem in part
from the "uncertain virtue of forced sharing," and courts should be "very cautious"
-45-
in imposing any duty to deal under Section 2. Trinko, 540 U.S. at 408; accord
MetroNet, 383 F.3d at 1133 ("[E]nforced sharing 'requires antitrust courts to act as
central planners, identifying the proper price, quantity, and other terms of
dealing--a role for which they are ill-suited.'" (quoting Trinko, 540 U.S. at 408)).
Second, Novell's assertion that its claim is for deception (not for a
refusal to deal) because Microsoft allegedly misled the "ISV community" when
explaining its rationale for the October 3 Decision (Novell Br. 19-21, 35-36) is
frivolous. Novell does not--and cannot--articulate how any purported deception
in describing (after the fact) the reasons for the October 3 Decision could possibly
have delayed the release of Novell's applications. In any event, Microsoft's
communications with ISVs after the October 3 Decision refute Novell's argument;
they show that Microsoft reached out to ISVs and assisted those ISVs who had
planned to use the NSEs to adjust without any major difficulty. (See JA-13765-67
(Gates); JA-7608 (DX82).) There was no deception of the "ISV community," as
shown by (1) the contemporaneous documentary evidence that Novell and others
were "OK" with the decision (JA-6926 (DX3)) and "found ways not to use" the
NSEs (JA-7608 (DX82)); (2) the fact that no ISV complained to Gates (JA-13765-66, 13771-73 (Gates)); and (3) the overwhelming evidence, see pp. 16-18, supra ,
-46-
that Novell and others in the software industry understood that betas can and do
change.
In any event, Novell's "deception" argument is wrong as a matter of
law. Novell cites no case--and Microsoft is aware of none--in which deception of
a competitor gave rise to a claim under Section 2 of the Sherman Act.25 As the
Supreme Court has made clear, "[e]ven an act of pure malice by one business
competitor against another does not, without more, state a claim under the federal
antitrust laws." Brooke Group v. Brown & Williamson Tobacco, 509 U.S. 209,
225 (1993). Indeed, "the Sherman Act does not convert all harsh commercial
actions into antitrust violations," Intergraph, 195 F.3d at 1354, and "[t]he Sherman
Act is not a panacea for all evils that may infect business life," Berkey Photo, Inc.
v. Eastman Kodak, 603 F.2d 263, 288 n. 41 (2d Cir. 1979). Deception of a
competitor (as opposed to consumers) has never been sufficient to give rise to a
Section 2 claim.26
-47-
II. MICROSOFT'S CONDUCT DID NOT HARM COMPETITION IN
THE PC OPERATING SYSTEM MARKET.
Novell offered the unusual theory that injury to products in one
market (here, applications such as word processors) had the effect of harming
competition in a different market (i.e., PC operating systems). No private treble
damages plaintiff has ever obtained a damages award on such a cross-market
theory. Beyond that, the fundamental and undisputed facts about the market
disprove the claim.
From 1990 to 1995, Microsoft's share of the PC OS market was above
90%. (JA-1625 (FoF ¶35).) The release of Windows 95 in August 1995 only
enhanced Microsoft's market position. All agree that Windows 95 was a
"'significant step' forward" from its predecessors. (JA-220-21 (Decision) (quoting
JA-12162-63 (Frankenberg)); JA-11184-85 (Harral).) Windows 95 was the
preeminent software platform of its time, exposing thousands of APIs that ISVs
used to write tens of thousands of full-featured applications. (See JA-11387-88
(Harral); JA-12054 (Frankenberg); JA-12397-98, 12497 (Alepin); JA-14364
(Muglia).)
In other words, in 1994-95, Microsoft had a monopoly in PC
operating systems, and its market share was about to go even higher because it was
introducing Windows 95, which all agree was then the best product. Novell's
-48-
claim--that had its Windows 95 applications been released earlier, this would
somehow have influenced consumers to buy other operating systems in much
greater numbers--has no evidentiary support.27
First, Novell's CEO squarely admitted just the opposite--that
Microsoft's market share would have been higher, not lower, if Novell had not
been delayed. Frankenberg testified that if Novell's applications had been timely
released, that "would have made Windows 95 more desirable in the marketplace"
(JA-12163), would have "increase[d] the sales of Windows 95" (JA-12164) and
thereby "would have made Windows 95's market share even higher than what it
turned out to be" (JA-12164-65). This admission affirmatively disproves Novell's
assertion that the allegedly wrongful act adversely affected competition in the PC
OS market.28
-49-
Second, by the time Windows 95 was released (and even before),
there were no effective PC operating system competitors.29 (JA-15693-94, 15721
(Murphy).) Even Novell's expert Noll acknowledged that no other operating
system was "really a competitor" to Windows 95 during the relevant period. (JA-12846, 12903-04; see also JA-14065-66 (Gates).)30
Third, Microsoft's PC operating system monopoly was protected by
an "applications barrier to entry" created by a "positive feedback loop" of
Windows users and the many thousands of Windows developers. (JA-1624
-50-
(FoF ¶31); JA-1626 (FoF ¶39).) This feedback loop arises because consumers
choose only PC operating systems "for which there already exists a large and
varied set of high-quality, full-featured applications" (JA-1624 (FoF ¶30); see also
JA-1625 (FoF ¶37)), and "developers generally write applications first, and often
exclusively, for the operating system that is already used by a dominant share of all
PC users," i.e., Windows. (JA-1624 (FoF ¶30); see also JA-1626 (FoF ¶38).) The
resulting "large body of applications" further "reinforces demand for Windows,
augmenting Microsoft's dominant position and thereby perpetuating ISV
incentives to write applications principally for Windows." (JA-1626 (FoF ¶39).)
There was no other operating system--not a single one--to which ISVs had
written so many applications during the relevant period. (JA-1625 (FoF ¶37.) As
a result, the notion that if two additional Novell applications had been available in
the market in 1995, some unidentified operating system would have taken market
share away from Windows 95, is completely implausible.31
-51-
A. The Franchise Applications Theory
Novell's first theory of cross-market harm is the "franchise
applications" theory (Novell Br. 45-49), which posits that WordPerfect and
Quattro Pro were so popular that had they been available on rival operating
systems, they would have threatened Microsoft's PC operating system monopoly
by "offer[ing] competing operating systems the prospect of surmounting the
applications barrier to entry and breaking the Windows monopoly."32
It is inconceivable that two or three applications could have impacted
competition in the PC OS market in this fashion. Findings 36-39 establish that a
"vastly larger number of applications are written for Windows than for other PC
operating systems" and the "positive feedback loop" cemented Microsoft's market
position. (JA-1625-26.) Even ignoring those binding Findings and assuming that
-52-
a few highly popular applications could somehow make a difference, Novell's
applications enjoyed nothing close to the level of popularity required to have any
such effect. As Judge Motz found--and Novell does not dispute--WordPerfect
had about 15% of the Windows-compatible word processing market, and
PerfectOffice had about 5% of the suite segment. (JA-222; see also JA-6835
(PX599A); JA-12858 (Noll).)33 Quattro Pro had an even smaller share of the
spreadsheet market--about 2% in 1994-95. (JA-15713-14 (Murphy); JA-1656.) It
is thus easy to conclude that a delay in the release of these products could not have
affected competition in the PC OS market. (JA-221 (Decision).)
Novell's franchise applications theory is also contradicted by the fact
that years earlier (late 1980s to early 1990s), there were versions of WordPerfect
that ran on many non-Microsoft operating systems, yet the availability of
WordPerfect on those other operating systems in no way diminished Microsoft's
share of the PC OS market. (JA-1625 (FoF ¶35).) Novell argues about
WordPerfect's "history" of being available on non-Microsoft operating systems
(Novell Br. 9), but the evidence was that Novell's applications were not available
-53-
on any competing platform during 1994-96. Frankenberg and Gibb conceded that
during the period from December 1994 (when Novell released PerfectOffice 3.0)
until March 1996 (when Novell sold to Corel), Novell was never even working on
a version of PerfectOffice for any operating system other than Windows 95. (JA-12105-06 (Frankenberg); JA-11721 (Gibb).)
B. The Middleware Theory
Novell's second theory was that its applications were "middleware"
that, as stated in the Findings, could have posed a threat to Microsoft's PC
operating system monopoly because they "had the potential to weaken the
applications barrier severely without the assistance of any other middleware." (JA-1627-28 (FoF ¶68); JA-123.) To pose such a threat, such applications had to
support "full-featured applications that will run well on multiple operating systems
without the need for porting," meaning without having to be adapted or re-written
to run on multiple operating systems. (JA-1629 (FoF ¶74).) Strikingly, as of 1999,
when the Findings were entered--let alone in 1995-96--no such cross-platform
middleware existed. (JA-1623-24 (FoF ¶¶28, 29).)
Applying the binding Findings, the trial court said that "[f]or a
middleware product to have an impact on competition in the PC operating system
market, the product (1) must be cross-platformed to various operating systems;
(2) must be ubiquitous on the 'dominant operating system'; and (3) must expose a
-54-
sufficient number of APIs of its own to entice ISVs to write applications to it rather
than to the operating system on which it sits." (JA-221 (citing JA-12866-69 (Noll),
and JA-1623 (FoF ¶28)).) Novell agrees that the first requirement applies (Novell
Br. 7, 51), and the Findings and testimony of Novell's expert establish that Judge
Motz was correct to apply the second and third as well. Novell came nowhere
close to proving any of the three.34
First, Novell's applications were not cross-platform. (JA-221-22
(Decision).) As stated above, Novell designed its PerfectOffice suite (purportedly
containing its middleware) exclusively for Windows, and Novell never wrote
PerfectOffice for any other platform. (JA-12105-06 (Frankenberg); JA-11721
(Gibb).) As a result, any applications written to Novell's middleware would still
need Windows because those applications could "not function" without an
underlying "operating system to perform tasks such as managing hardware
-55-
resources and controlling peripheral devices." (JA-1623 (FoF ¶29).) For this
reason alone, Novell's purported middleware could not have reduced the
dominance of Windows.
Further, as Judge Motz said, Novell's use of the NSEs would have
made PerfectOffice less cross-platform, because the NSEs were "'platform
specific' to Windows." (JA-222 (quoting JA-12420-21 (Alepin), and citing JA-12471-72 (Alepin), JA-15741-42 (Murphy), and JA-15983 (Bennett)).) As
Novell's expert conceded, not a single competing operating system implemented
the functionality of the NSEs (JA-12471-72), and thus writing Novell's
applications to the NSEs would have made it much more difficult for Novell to
create a version of its purported middleware for non-Microsoft operating systems.
Second, as the trial court held, Novell's applications were not
ubiquitous on Windows. (JA-222.) Novell's expert Noll agreed that ubiquity was
a requirement for the "middleware" theory to make any sense. (JA-12866, 16868-69.) Notwithstanding Novell's disagreement as to the meaning of ubiquity, Novell
does not challenge that its share of the Windows-compatible word processing
market was 15% (and its share of the suite segment was 5%). (JA-222 (Decision).)
This falls far short of ubiquity under any definition. Indeed, given its modest
market share, Novell's purported middleware could not offer developers the
-56-
prospect of reaching the "dominant share of all PC users" (JA-1624 (FoF ¶30)),
leaving developers no incentive to write for Novell's "middleware" rather than for
Windows (which offered ISVs access to more than 90% of consumers).
Third, "Microsoft is [also] entitled to judgment as a matter of law
because Novell did not present evidence to show that its software exposed
sufficient APIs of its own to allow ISVs to write full-featured personal
product[ivity] applications to it." (JA-224 (Decision).) Indeed, Novell conceded
at trial that it could not meet this third element of the middleware theory. (JA-16399-400, 16402 (Novell's chief lawyer saying that the court "would be directing
a verdict" for Microsoft if it applied the third requirement).)
Novell's argument that the third element should be altered to say that
middleware need only "expose[] a wide range of APIs and sophisticated
functionality to developers" (see JA-223-24 (Decision); see also Novell Br. 52) is
contrary to the binding Findings.35 Finding 74 explains that Sun's Java, to which
Novell analogizes its purported middleware (Novell Br. 12, 49-50; JA-123), posed
a threat only because it would allow "[a] program written in Java and relying only
-57-
on APIs exposed by the Java class libraries" to "run on any PC" on which Java
technology could run.36(JA-1629 (FoF ¶74); see also JA-1627-28 (FoF ¶68).)
In sum, competition in the PC OS market was not harmed by the delay
in the release of Novell's applications. As Frankenberg admitted, if Novell's
products had been timely released, Microsoft's market share would have been
"even higher than what it turned out to be." (JA-12164-65.)
III. THE OCTOBER 3 DECISION DID NOT CAUSE ANY DELAY.
PerfectOffice was delayed beyond the time that Novell says was
critical to its success (60-90 days after August 24, 1995) because Novell could not
finish Quattro Pro for reasons having nothing to do with Microsoft.
Novell's internal documents and the testimony of five Novell
employees--including its CEO, who testified that "clearly" Quattro Pro "wasn't
-58-
complete" as of December 23, 1995 (JA-12081)--wholly disprove Novell's claim
about delay. Further, the facts that (a) Novell never complained to Microsoft about
the October 3 Decision and (b) no senior Novell executive was contemporaneously
told that any delay was caused by Microsoft, are added proof that the reason for
delay lies elsewhere. As Judge Motz said, the entire case was a "lawyers'
construct." (JA-227.)
A. Quattro Pro Caused Novell's Delay.
Novell's claim is that absent the October 3 Decision, Novell would
have released PerfectOffice within "the critical window of opportunity that would
close 90 days after" August 24, 1995. (Novell Br. 22.) It is undisputed that
PerfectOffice could not be released without Quattro Pro. (See JA-12079
(Frankenberg); JA-8988 (DX221).) The overwhelming evidence shows, as Judge
Motz stated, that "Quattro Pro was not ready for release to manufacturing until
1996." (JA-228.) This simple fact kills Novell's case.
Five former Novell employees, LeFevre (Product Marketing Director
for Windows Product Management), Ford (Team Leader for WordPerfect for
Windows 95), Larsen (Software Developer for Quattro Pro), Bushman
(International Product Marketing Manager for PerfectOffice) and Frankenberg
(CEO) all testified that Quattro Pro was not ready to be released by
-59-
Christmas 1995. The documentary evidence also eviscerates any contrary
contention. See pp. 29-31, supra .
The standard on a Rule 50 motion is "not whether there is literally no
evidence supporting the nonmoving party, but whether there is evidence upon
which a jury could properly find for that party," which requires that the nonmoving
party "present more than a scintilla of evidence supporting its claim." Herrera,
474 F.3d at 685; accord Bankers Trust v. Lee Keeling & Associates, 20 F.3d 1092,
1100-01 (10th Cir. 1994).
In the face of the mass of evidence that PerfectOffice was delayed
because of Quattro Pro, Novell has only the vague testimony of Gary Gibb.37 He
testified that Quattro Pro was "basically code completed" as of December 23,
1995, when DX230 shows that almost all the Quattro Pro developers resigned.
(JA-11742.) Even this fails to help Novell because Christmas 1995 was 120 days
after the release of Windows 95, and Novell insisted at trial that its product was
dead on arrival unless it could be released within 60-90 days of August 24, 1995.
Indeed, all of Novell's damages (about $1.3 billion before trebling) were
-60-
specifically conditioned by Novell's damages expert on a release of PerfectOffice
"within 30 or 60 days" of the August 24, 1995 release of Windows 95. (JA-13364
(Warren-Boulton).)
Moreover, there was no evidence that Gibb ever went to Scotts Valley
(where Quattro Pro for Windows 95 was being developed). Nolan Larsen did; he
was there every week for two months beginning in January 1996 and testified that
Quattro Pro was not then ready "by any stretch of the imagination." (JA-14575-78,
14618-19.) Gibb's testimony falls far short of the "substantial evidence" that a
"reasonable mind might accept as adequate to support a conclusion" under
Rule 50. Webco Industries, 278 F.3d at 1128. As Judge Motz said during trial,
"there comes a point where somebody comes in and they say it's like saying the
world is flat. There . . . could not be clearer evidence that Defendant's Exhibit 230
[JA-9145] says that as of January 1996 Quattro Pro is not ready yet." (JA-13875.)
B. Novell's Poor Business Choice Caused the Delay.
Even if, contrary to the evidence, Novell's work on a custom file open
dialog caused any delay, Microsoft cannot be held liable for Novell's poor business
choice. At trial, Novell's employees unanimously testified that Novell's shared
code group could have completed its development of PerfectOffice for
Windows 95 had it chosen Options 1 or 2 instead of writing its own custom FOD.
See pp. 25-27, supra . As Judge Motz said, if "the 90-day period after the release
-61-
of Windows 95 was critical to the success of WordPerfect, Quattro Pro and
PerfectOffice, Novell could have released PerfectOffice using Microsoft's
common file open dialog." (JA-228.) Novell does not disagree, arguing instead
that Options 1 and 2 were far from ideal. Yet--according to Novell--Option 3
destroyed its whole applications business.
When an antitrust plaintiff "bypasses an obviously adequate
alternative," it "should not recover." J.T. Gibbons v. Crawford Fitting, 565 F.
Supp. 167, 180 (E.D. La. 1981). In Gibbons, the district court directed a verdict
for defendants in a Sherman Act action where plaintiff could have obtained from
an alternative source the product that defendants refused to sell to it. Id. at 171,
181. The Fifth Circuit affirmed, noting that "[t]he sine qua non of the injury
caused by a refusal to deal would be inability to obtain the product." 704 F.2d 787,
792 (5th Cir. 1983); accord Union Cosmetic Castle v. Amorepacific Cosmetics
USA, 454 F. Supp. 2d 62, 71 (E.D.N.Y. 2006) (holding in a Sherman Act case that
plaintiffs' "free choice to reject" an alternative solution "acts as an intervening
cause of" the injury).
As Gibb acknowledged, Novell was "faced with the age old trade off"
between "get[ting] out a product more quickly and sacrific[ing] features" or
"delay[ing] until 1996 and try[ing] to build a cooler product." (JA-11825-26.)
-62-
Karl Ford testified that people at Novell discussed even yet another alternative, that
Novell should "use the common open dialog right now" and include a custom
dialog later, "in the next release." (JA-14664.) Novell's argument that these
available options would have sacrificed "cooler" functionality does not give rise to
an antitrust claim. Gibbons, 565 F. Supp. at 180.
Novell's own witnesses admitted that "Novell could have come out
with a product in '95 that utilized the Windows common file open dialog" (JA-11434 (Harral)) and that it would have been "quite easy" for Novell to release its
applications using Microsoft's FOD. (JA-11781-82 (Gibb).) This is another
independent ground for affirmance.
-63-
CONCLUSION
This Court should affirm the Judgment.
Dated: January 23, 2013
Respectfully submitted,
/s/ David B. Tulchin
David B. Tulchin
Steven L. Holley Sharon L. Nelles Adam S. Paris
SULLIVAN & CROMWELL
Steven L. Holley
[address, phone]
Steven J. Aeschbacher
MICROSOFT CORPORATION
[address, phone]
James S. Jardine RAY QUINNEY & NEBEKER
[address, phone]
Attorneys for Microsoft Corporation
-64-
CERTIFICATE OF COMPLIANCE
I certify that this brief's type size and typeface comply with Fed. R.
App. P. 32(a)(5) and (6) because this brief has been prepared in a proportionally
spaced typeface using Microsoft Word 2010 in 14-point Times New Roman font. I
certify that this brief complies with the type-volume limitation of Fed. R. App. P.
32(a)(7)(B) because this brief contains 13,993 words, excluding the parts of the
brief exempted by Fed. R. App. P. 32(a)(7)(B)(iii) and 10th Cir. R. 32(b).
Dated: January 23, 2012
/s/ John Quinn
John Quinn
SULLIVAN & CROMWELL LLP
[address, phone]
John Quinn
SULLIVAN & CROMWELL LLP
-65-
STATEMENT REGARDING ORAL ARGUMENT
Microsoft Corporation respectfully requests that this Court hold oral
argument on this appeal. Microsoft believes that the decisional process would be
aided by oral argument.
Dated: January 23, 2013
/s/ David B. Tulchin
David B. Tulchin SULLIVAN & CROMWELL LLP
[address, phone]
Attorney for Microsoft Corporation
David B. Tulchin
SULLIVAN & CROMWELL LLP
-66-
CERTIFICATE OF SERVICE
I hereby certify that on January 23, 2013, I caused the foregoing Brief
of Appellee Microsoft Corporation to be electronically filed with Clerk of the
Court for the United States Court of Appeals for the Tenth Circuit using the
appellate CM/ECF system. Participants in the case who are registered CM/ECF
users will be served by the appellate CM/ECF system.
I further certify that on January 23, 2013, I caused two hard copies of
the foregoing Brief of Appellee Microsoft Corporation to be mailed by Federal
Express for next day delivery to each of:
David Boies
BOIES SCHILLER & FLEXNER LLP
[address]
Jeffrey M. Johnson DICKSTEIN SHAPIRO LLP
[address]
/s/ John Quinn
John Quinn SULLIVAN & CROMWELL LLP
[address, phone]
-67-
CERTIFICATE OF DIGITAL SUBMISSION
I hereby certify that (i) the document electronically filed with Clerk of
the Court for the United States Court of Appeals for the Tenth Circuit in Digital
Form via the Court's CM/ECF system is an exact copy of the written document
filed with the Clerk; and (ii) the digital submission has been scanned for viruses
with Microsoft Forefront, and according to that program, is free of viruses.
/s/ John Quinn
John Quinn SULLIVAN & CROMWELL LLP [address, phone]
John Quinn
SULLIVAN & CROMWELL LLP
-68-
______________
1
Novell purchased WordPerfect Corporation and bought the Quattro Pro
spreadsheet application from Borland International on June 24, 1994. (JA-118.) In
March 1996, Novell sold both to Corel Corporation. (JA-163.)
2
Count VI--which alleged that Microsoft entered into exclusionary
agreements to inhibit distribution of Novell's applications (JA-167-68)--was later
abandoned by Novell. Novell v. Microsoft, 429 F. App'x 254, 258 n. 7 (4th Cir.
2011).
3
In the Government Case, the DOJ alleged that Microsoft unlawfully
maintained a monopoly in the PC OS market, but did not allege that Microsoft had
acquired its monopoly unlawfully. Novell also made no such allegation at trial.
4
With the court's permission, Microsoft contacted members of the jury and
seven were willing to be interviewed. Five of the seven said that they would not
have awarded Novell even one dollar in damages, and four or five would have
found no liability if the outcome included an award of damages.
5
All Findings of Fact cited in this brief were given collateral estoppel effect
in this action.
6
To state the obvious, Windows 95 was Microsoft's creation, and Microsoft
(not Novell or other ISVs) was entitled to make decisions about the design of the
new operating system.
7
As shown below, see p. 49, infra, Novell's CEO testified that Microsoft's
already huge market share would have gone up (not down) if Novell's products
had been released in a timely fashion. This alone destroys Novell's assertion that
the October 3 Decision adversely affected competition in the relevant market.
According to Novell's CEO, Microsoft's market share would have been "even
higher" had it not engaged in the challenged conduct. (JA-12164-65.)
8
WordPerfect's share of the word processing market on Windows peaked at
about 30% in 1992, and declined to "the low 20s" by 1994. (JA-15709 (Murphy);
JA-9396 (DX294); see also JA-7040 (DX9).)
9
Indeed, Novell's entire case is illogical because it is in Microsoft's business
interest to encourage ISVs to write their applications to Windows. Denying ISVs
access to necessary APIs would make Windows less appealing. In re Microsoft
Antitrust Litigation, 274 F. Supp. 2d 743, 746 (D. Md. 2003) (Microsoft's
monopoly position "depends upon Microsoft encouraging ISVs to choose the
Windows operating system" and it is not "logical" to allege that Microsoft would
deprive ISVs of "access to the APIs" they need).
10
Windows Explorer, not to be confused with Internet Explorer, is the
operating system's integrated resource viewer, which allowed users to access a
variety of resources such as files, devices and networked computers.
11
It was undisputed at trial that Novell's applications did not need to use the
NSEs to be compatible with Windows 95, and that without using the NSEs, Novell
could launch its applications from the Start menu or from icons on the desktop.
(See, e.g., JA-12516-17 (Alepin).)
12
Novell's assertion that the documentation was merely "written guidance"
that was insufficient to allow Novell to use the NSEs (Novell Br. 17) is refuted by
its own programmer, Richardson. He testified that Novell was "able to make use
of" the NSEs based on what Microsoft provided in June 1994. (JA-11667; see also
JA-11673.)
13 Once Gates made the decision, Microsoft's applications also did not utilize
the NSEs. (JA-13776 (Gates); JA-12580-82 (Alepin); JA-15236-37 (Belfiore); JA-
15950-51 (Bennett).) Novell's brief (at 20-21) misleadingly refers to Marvel (an
online system utility) that shipped as part of Windows and was not a separate
product (JA-13765, 13767 (Gates)).
14 Gates announced his October 3 Decision in a long e-mail that included his
saying that "[w]e should wait until we have a way to do a high level of integration
that will be harder for [the] likes of Notes, WordPerfect to achieve, and which will
give Office a real advantage." (JA-1967 (PX1).) Even if, as Novell argues, this
shows that Microsoft was concerned with competition from Novell (see Novell Br.
21 n.3), Gates' e-mail refers only to a hoped-for advantage for Office from an
expected, later, improved version of the NSEs. Gates was not aware of any harm
to Novell flowing from the October 3 Decision (JA-13761), and he never refers to
any benefit to Windows in competition with any other operating system. Novell's
only claim here is for harm to competition in the PC OS market.
15 Novell's brief (at 37) misleadingly refers to an attachment to Struss' report
(written before September 22 and by a different Microsoft employee) that says that
"based on speculation," it was "[l]ikely" that WPC/Novell planned to use the NSEs
(JA-3658 (PX215); JA-7170 (DX17)). Struss testified at trial that his
September 22 report saying that Novell was "OK" was written later, and was based
on "specific feedback" from Novell. (JA-14219-20.)
16 The only documents that Novell cites for the proposition that it "complained
repeatedly about the de-documentation" are PX236 and PX238, e-mails written by
Kelly Sonderegger of Novell in November 1994. (Novell Br. 37.) Neither
document contains a word of complaint about the NSEs or the October 3 Decision.
(JA-3767-68; JA-3772.)
17 Novell points out (Novell Br. 37) that programmer Harral testified that he
spoke with Microsoft's Premier Support telephone hotline three times about the
NSEs. Harral was unable to provide even a month in which any such call took
place (JA-11260-62), or the names of anyone with whom he spoke (JA-11329,
11331, 11346). There is no e-mail or other document that reflects that any such
thing ever occurred. Moreover, Harral's boss Creighton was in frequent contact
with Struss and never made any complaint. (JA-14211-12, 14214-15, 14225
(Struss).)
18 The shared code team was responsible for writing the code for "file
handling," which included open and save functions, in the PerfectOffice suite.
(JA-11140 (Harral).)
19
A FOD is a window that opens on the screen and allows users to select and
open a file from the file system. (JA-11140, 11178-80 (Harral).)
20
Microsoft provided all ISVs--free of charge--with a standard FOD
(sometimes called the "common" FOD) in Windows 95 for opening and saving
files. (JA-11202, 11566 (Harral).)
21 Novell criticizes the Court's use of "middle manager" to describe Gibb, but
Gibb was one of ten people who reported to Bruce Brereton, VP of Business
Applications Group (JA-4448-49 (PX372)); Brereton reported to a Senior VP (JA-10217 (DX380)), who in turn reported to an Executive VP (JA-10216 (DX380)),
who reported to CEO Frankenberg (JA-10215 (DX380)).
22 It is undisputed that Microsoft did nothing to cause any delay in Novell's
development of Quattro Pro.
23
In 1994, Novell's annual revenues were nearly $2 billion. (See JA-10230
(DX 381).)
24
Novell is wrong that Aspen does not require "complete termination of a
relationship." (Novell Br. 41-42.) Three of the four cases Novell cites recognized
that in Aspen, defendant completely terminated the relationship. MetroNet
Services v. Qwest, 383 F.3d 1124, 1132 (9th Cir. 2004); Safeway v. Abbott
Laboratories, 761 F. Supp. 2d 874, 893 (N.D. Cal. 2011); A.I.B. Express v. FedEx,
358 F. Supp. 2d 239, 250-51 (S.D.N.Y. 2004). The fourth case, Nobody in
Particular Presents v. Clear Channel Communications, 311 F. Supp. 2d 1048 (D.
Colo. 2004), is inapposite. Plaintiff there did not allege that defendant altered or
terminated its business relationship with plaintiff, but rather that defendant's
conduct was unlawful from the start. Id. at 1105-06.
25
Novell cites only one case on this subject--Rambus v. FTC, 522 F.3d 456,
463 (D.C. Cir. 2008). There, plaintiff alleged that defendant deceived a trade
organization, and the court held that plaintiff failed to establish liability under
Section 2 of the Sherman Act.
26
The cases on which Novell relied in the district court all involved deception
of customers and/or deceptive advertising. Even there, other courts have been
highly reluctant to find liability. See, e.g., American Professional Testing Service
v. Harcourt Brace Jovanovich, 108 F.3d 1147, 1152 (9th Cir. 1997).
27
The availability of more Windows applications obviously increases demand
for Windows; it does not decrease it.
28
Of course, when Gates made the October 3 Decision, he did not know that
Novell was planning to use the NSEs and thus could not have made the same
calculation. See p. 20, supra . Moreover, as Judge Motz stated, "there is no
evidence that Microsoft sacrificed any short-term profits." (JA-219 (Decision).)
Novell misleadingly attempts to convert Judge Motz's hypothetical following this
statement into an affirmative judicial finding that short-term losses in the operating
system market were offset by increased application sales. (Novell Br. 41 n. 6
(citing JA-219 n.18).) There was no such finding, and, as the court said, "no
evidence" to support Novell's argument.
29
Novell does not dispute this, and instead argues that "in violation of the
hypothetical marketplace principle, the District Court improperly required Novell
to speculate about alternative operating systems." (Novell Br. 47.) To the
contrary, no "speculation" is necessary. Novell's theories of harm to competition
depend on the contemporaneous existence of a competing operating system that
could gain market share. The evidence was that no such operating system existed.
Moreover, King & King Enterprises v. Champlin Petroleum, 657 F.2d 1147 (10th
Cir. 1981), stands for the unremarkable proposition that a plaintiff need not create
a hypothetical market for the purposes of calculating "the amount of damages"
suffered by plaintiff. Id. at 1162 n. 1 (emphasis added). It does not give Novell
license to dispense with evidence establishing that Microsoft's conduct actually
harmed competition in the relevant market.
30
Noll testified that by 1995, IBM's OS/2 was "not an effective competitor" to
Windows (JA-12846) and that Linux also "wasn't really a competitor" because it
only "became a full-fledged, commercial product" in 1996 (JA-12846, 12904).
Indeed, Finding 60 (entered in November 1999) states that even by 1999
consumers had never turned to Linux as an alternative to Windows and were not
"likely to do so in appreciable numbers any time in the next few years." (JA-1627.)
31 There was no evidence at trial of any plausible manner in which
WordPerfect or PerfectOffice could have altered the competitive landscape for
operating systems. Instead of addressing this point, Novell argues (Novell Br. 29-31) that the standard for causation should not be the "contributes significantly"
standard applicable to treble damages cases, but instead the "reasonably capable of
contributing significantly" standard that was applied in United States v. Microsoft.
As the D.C. Circuit expressly stated in the Government Case, it was applying an
"edentulous test for causation" and thus, the DOJ was not required to show that "Java or Navigator would actually have developed into viable platform
substitutes." 253 F.3d at 79. In an equitable enforcement case, the DOJ is entitled
to stamp out "nascent threats" to competition. Id. In contrast, in private treble
damages cases, two-thirds of the amount sought is entirely "punitive," and it is
therefore "critical" that remedies "be strictly limited to those aspects of a plaintiff's
injury that were in fact caused by an unlawful exploitation of market power or
unlawful quest for such power in attempt cases." 3 PHILLIP E. AREEDA & HERBERT
HOVENKAMP, ANTITRUST LAW ¶657a (2012). In any event, under either standard,
Novell did not submit enough evidence to get to a jury.
32 This quoted language is how Novell characterized the theory in its brief in
support of proposed jury instructions. (Dkt. 336 at 4.)
33 There is no dispute that even considering Novell's installed base (the
percentage of PCs with WordPerfect installed), which included earlier versions of
WordPerfect for the DOS platform, WordPerfect's share did not exceed 36%. (JA-222-23 (Decision).)
34
Novell now argues that its PerfectOffice suite, in conjunction with Java and
Netscape--products that Novell never owned or developed--"posed a
considerable threat to the applications barrier to entry and Microsoft's monopoly
power." (Novell Br. 49; see also id. 53-54.) Before trial, Microsoft moved in
limine to exclude evidence of any theory that PerfectOffice could be a middleware
threat "in combination with" Netscape Navigator and Sun's Java (Dkt. 112) and
Judge Motz granted Microsoft's motion. (JA-10689.) Novell does not challenge
that ruling on appeal, and accordingly has waived any argument that PerfectOffice
posed a threat to Microsoft's monopoly in combination with these other products.
See Mooring Capital Fund v. Knight, 388 F. App'x 814, 818-19 (10th Cir. 2010)
(challenges to evidentiary rulings not made in opening brief are waived).
35
To the extent that Novell seeks to stray from the Findings and the definition
of middleware used in the Government Case, its claim is barred by the applicable
four-year statute of limitations (15 U.S.C. § 15b) because rather than being "based
in whole or in part" on the Government Case, 15 U.S.C. § 16(i), it would instead
run contrary to the Government Case. See pp. 4-5, supra .
36
Grasping at straws, Novell argues that it need not meet this requirement because PerfectOffice was "simultaneously middleware and a set of personal
productivity applications." (Novell Br. 52.) This makes no sense; for middleware
to threaten Windows, it must expose sufficient APIs so that other ISVs can write
"full-featured personal productivity applications" to the middleware, instead of to
Windows. (JA-1623-25 (FoF ¶¶28, 32).) Indeed, Novell's technical expert agreed
that "there are literally thousands and thousands of software products that expose
some APIs" but conceded that "[t]here's got to be more than" that for a product to
be middleware of the sort that eliminates reliance on the underlying operating
system. (JA-12399-400.) Novell fails to argue--and there is no evidence--that
PerfectOffice exposed sufficient APIs to support full-featured personal
productivity applications. (See JA-1623 (FoF ¶28).)
37
Novell also refers to DX231, which it argues shows that Quattro Pro was
code complete in August 1995. (Novell Br. 61-62.) Novell failed to ask any
witness about this document. In fact, DX231 shows that the date Quattro Pro was
scheduled to be released to manufacturing was March 31, 1996 (JA-9147), thus
showing again that Judge Motz got it right.
|