decoration decoration
Stories

GROKLAW
When you want to know more...
decoration
For layout only
Home
Archives
Site Map
Search
About Groklaw
Awards
Legal Research
Timelines
ApplevSamsung
ApplevSamsung p.2
ArchiveExplorer
Autozone
Bilski
Cases
Cast: Lawyers
Comes v. MS
Contracts/Documents
Courts
DRM
Gordon v MS
GPL
Grokdoc
HTML How To
IPI v RH
IV v. Google
Legal Docs
Lodsys
MS Litigations
MSvB&N
News Picks
Novell v. MS
Novell-MS Deal
ODF/OOXML
OOXML Appeals
OraclevGoogle
Patents
ProjectMonterey
Psystar
Quote Database
Red Hat v SCO
Salus Book
SCEA v Hotz
SCO Appeals
SCO Bankruptcy
SCO Financials
SCO Overview
SCO v IBM
SCO v Novell
SCO:Soup2Nuts
SCOsource
Sean Daly
Software Patents
Switch to Linux
Transcripts
Unix Books
Your contributions keep Groklaw going.
To donate to Groklaw 2.0:

Groklaw Gear

Click here to send an email to the editor of this weblog.


To read comments to this article, go here
Novell v. Microsoft - The June 7, 2012 Hearing on MS's Renewed Motion for JMOL ~ pj
Monday, July 09 2012 @ 01:11 PM EDT

I've now read through all 261 pages of the transcript of the June 7, 2012 hearing that the judge forgot to tell the public was scheduled for that day in Novell v. Microsoft. This is the antitrust case Microsoft is trying to get tossed out on a motion for judgment as a matter of law, after an 8-week jury trial that ended with mistrial. It was reported by the jurors to be 11 to 1 for Novell, with one holdout juror. That was as close as Microsoft would like to get to a jury trial on this WordPerfect litigation ever again, so it filed its renewed motion, asking the judge to let it win without having to face a second jury.

The judge seems inclined to do exactly that, from all we've seen. For Microsoft, we have David B. Tulchin, Esq., and for Novell Jeffrey M. Johnson, Esq. and they couldn't be less enamored of one another. But to understand this litigation, you need to understand the current state of antitrust law in the US.

Jump To Comments

To follow along with precision, you need the documents First, here are the filings and transcripts regarding Microsoft's initial motion for judgment as a matter of law, brought at the end of Novell's presentation of evidence at trial but prior to the jury verdict, its Rule 50(a) motion:
After the denial of the Rule 50(a) motion, the trial continued, the jury deliberated, there was an inability to arrive at a unanimous verdict, and after that Microsoft renewed its motion for judgment as a matter of law, which is now a Rule 50(b) motion:
  • Microsoft's renewed motion
  • Memorandum in support
  • Novell's opposition
  • Microsoft's reply
  • You'll see the lawyers and the judge in the transcript reference some of these documents, claiming one or the other side said this or that, and this list makes it possible for you to test their claims for yourself.

    It seems when regulators were more or less ignoring the excesses on Wall Street, the same thing was in essence happening in the antitrust world. To understand the sorry condition of antitrust law, and what an uphill struggle Novell has, you could read this paper that Microsoft recommended to the court, on page 22 of the PDF, The “No Economic Sense” Test for Exclusionary Conduct, by Gregory J. Werden. It includes a collection of judges' pondering how to tell the difference between antitrust actions that kill off a competitor and good old apple-pie American competition that just happens to kill off a competitor.

    Microsoft's Arguments:

    Microsoft's theory of the case is that while it might be a monopolist, it didn't single Novell out for death. It was just busy making a great new innovative operating system at the time, Windows 95, and Novell just couldn't keep up. Monopolists are allowed to improve their products. Only middle management at Novell were even involved in the API issue, and if this was such a huge issue at the time, why didn't highest management get involved? Besides, you have to be really, really bad and provably so to lose a private antitrust case these days, and Novell has no real evidence that anyone at the time thought there was a problem, so that's where Microsoft hangs its hat. There is no antitrust case here. Maybe Novell could have sued under some tort theory, but not antitrust, and it's too late now to sue under any tort theory. Microsoft had no duty to provide Novell anything, which Novell has conceded, Microsoft argues. Deception, Microsoft states, is not an antitrust claim, even if we did deceive Novell, because it signed that release Novell signed in 2004, which released all claims except those in the complaint in this case, and the complaint didn't list a deception claim. And anyway, Microsoft says, antitrust law isn't designed to enforce "gentlemanly conduct over tea in Greenwich". There are no cases, it argues, that deceiving a competitor or other such bad conduct is actionable.

    If you were Microsoft's lawyers, you would do this too. You'd argue the same way. The deeper question, though, aside from whether or not you'd agree to be a Microsoft lawyer, is a moral, not a legal, one: what if you were Microsoft? Should you defend yourself like this?

    I know. Businesses don't seem to pay much attention to morals these days, and with the state of antitrust law, why would they?

    That article Microsoft relies on begins with an Introduction:

    Section 2 of the Sherman Act prohibits the acquisition or maintenance of monopoly power through the use of “predatory” or “exclusionary” conduct.1 In Trinko,2 the Solicitor General argued that when “the plaintiff asserts that the defendant was under a duty to assist a rival . . . conduct is not exclusionary or predatory unless it would make no economic sense for the defendant but for its tendency to eliminate or lessen competition.”3 Although the Solicitor General only advocated this “no economic sense” test for a narrow class of conduct, the Department of Justice (DOJ) has consistently advocated this test in all of its section 2 cases.4 This Article briefly discusses Trinko and explains how the “no economic sense” test is motivated by policies the Supreme Court articulated in Trinko and prior cases. This Article also explains the role the test can usefully play in section 2 cases and why it does not have the flaws of the “profit-sacrifice” test with which the “no economic sense” test has been confused.
    Four footnotes in the first paragraph. That gives you the idea, and you can read them all, if you are curious, but here are a few choice quotes from various courts, minus the footnotes, to show you what Novell is up against, beginning with the US Supreme Court in Trinko:
    The Court then explained why the Sherman Act should not be understood to impose a general “duty to aid competitors”:
    Compelling . . . firms to share the source of their advantage is in some tension with the underlying purpose of antitrust law, since it may lessen the incentive for the monopolist, the rival, or both to invest in those economically beneficial facilities. Enforced sharing also 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. Moreover, compelling negotiation between competitors may facilitate the supreme evil of antitrust: collusion.
    The Court held that “as a general matter, 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 the parties with whom he will deal.’”

    Acknowledging there were “exceptions” to this general rule, the Court reviewed a particularly notable example—its Aspen decision -- making clear that the facts of Aspen were “at or near the outer boundary” of circumstances in which “a refusal to cooperate with rivals can constitute” a violation of section 2. In explaining why it made sense to impose antitrust liability in Aspen, even if not in most other cases, the Court stressed that the defendant had refused to sell lift tickets to the plaintiff even at the retail price charged to individual skiers. Hence, the Court concluded that the facts of Aspen “suggested a willingness to forsake short-term profits to achieve an anticompetitive end.” The Court thereby hinted at something akin to the “no economic sense” test.

    In short, this is a paper pushing the "no economic sense" test, as you may have gathered:
    Although Trinko does not adopt the “no economic sense” test, both it and other Supreme Court decisions from the past twenty years articulate policies motivating that test. First and foremost, the Court has made clear that section 2 of the Sherman Act, which applies to single firm conduct, sweeps far less broadly than section 1, which prohibits all concerted conduct that “imposes an unreasonable restraint on competition.” The Court has described the Sherman Act as a “consumer welfare prescription,” but only section 1 makes consumer welfare the touchstone for legality.

    It is not enough that a single firm appears to “restrain trade” unreasonably, for even a vigorous competitor may leave that impression. For instance, an efficient competitor may capture unsatisfied customers from an inefficient rival, whose own ability to compete may suffer as a result. This is the rule of the marketplace and is precisely the sort of competition that promotes the consumer interests that the Sherman Act aims to foster.

    “Judging unilateral conduct” less strictly than concerted conduct, the Court reasoned, “reduces the risk that the antitrust laws will dampen the competitive zeal of a single aggressive entrepreneur.” Put another way: “Subjecting a single firm’s every action to judicial scrutiny for reasonableness would threaten to discourage the competitive enthusiasm that the antitrust laws seek to promote.”

    A significant part of the Court’s rationale for treating single firm conduct less strictly than concerted conduct is that single firm conduct presents a greater danger that procompetitive conduct will be erroneously condemned. That danger is especially severe with exclusionary practices that attract customers away from rivals by offering a better bargain. Offering customers a better bargain is characteristic of the conduct deemed “competition on the merits,” yet it is also precisely how certain conduct, like predatory pricing, may exclude competition. Consequently, “[i]t is sometimes difficult to distinguish robust competition from conduct with long-run anticompetitive effects.” The potential to erroneously condemn procompetitive conduct is obvious with allegedly predatory pricing, and the Court’s Matsushita and Brooke Group decisions emphasized that “mistaken inferences” of predatory pricing “are especially costly, because they chill the very conduct the antitrust laws are designed to protect.”

    In short, what the paper is arguing is that if a few companies get together and do anticompetitive things, it's easier to nail them to the antitrust wall, but if a single company does anticompetitive things against just one competitor, it's hard, really hard, to prove it wasn't just normal business dealings. And so Microsoft is saying it was just minding its own business, not trying to kill Novell off, and Novell can't prove otherwise.

    Novell, of course, keeps trying to show evidence, arguing that Microsoft was willing to forgo profits in the short term in order to kill off Novell, that it was deliberately dropping documentation on the APIs Novell had been encouraged to use to throw tacks in Novell's road, evidence like the notorious email [PDF] where Bill Gates wrote that he'd decided to undocument the APIs. But, say Microsoft and the judge, the company had good business reasons not to want people using those APIs, due to stability issues, or so went the Microsoft testimony from Bill Gates, and even a monopolist gets to do that, no matter how it happens to affect competitors. And how can we know what Gates was thinking?

    Um, says Novell, because the same email tells us? Gates said he had decided to do it so they could arrange things so it would be harder for Novell to use successfully:

    I have decided that we should not publish these extensions. 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. This means that Capone and Marvel can still live in the top level of the Explorer namespace, but will run separately. We can continue to use the iShellBrowser APIs for MS provided views such as control panel, and can use them for other MS-provided views that don't create a large compatibility or ISV issue.
    Duh. There's a funny exchange in one hearing where Novell is trying to remember the name of the retreat that inspired this email, something Canal, the Novell lawyer says, root canal? And Microsoft's lawyer says, I certainly hope it wasn't called root canal. It wasn't. But that's how it felt to Novell at the time, the outworking of this scheme. Maybe Love Canal, actually. Nah. It was the Hood Canal email, because the retreat was held there. You can read that email in plain text on this page, from the exhibits in the Comes v. Microsoft antitrust trial by searching for #2151. There's a remnant of a prior email at the very top, so ignore it. The subject line is "Subject: Shellplans - iShellBrowser" and the date is October 3, 1994.

    But, says Microsoft, if it was so bad, why didn't Novell complain immediately? Um, says Novell, because we hadn't seen that internal email. And we did complain. We complained to them to the degree we dared and we complained to the Department of Justice, which successfully sued Microsoft.

    Well, too late now, Microsoft says. That was about something else, anyway. You're just creating a case out of thin air. At the time, no one minded, and you have no proof otherwise.

    Oh, and all the exhibits Novell wants to add to the pile before the court that would prove it? Microsoft wants them to not be allowed in.

    That's Microsoft's position. We weren't that bad and if we were, you can't prove it, due to technicalities of antitrust law and rules of the road.

    It's Microsoft's Rule 50(b) motion, so it gets to go first (and last), and the hearing -- which the docket indicates lasted all day until 6 PM at night -- opens with Mr. Tulchin handing up two binders of materials to the judge. I can't show you the transcript until September, but I can tell you what happened. Hearings are supposed to be open to the public, after all, unless there is a good reason to close the doors, and this was an oversight. So I'll be your reporter for the day, you could say.

    But the judge, the Hon. Frederick Motz of Baltimore interrupts and asks him his first question, namely why take the motion seriously? Why not just have the second trial and see what the jury does? Microsoft says Rule 50 motions are there for a reason. It's the judge's responsibility, after a trial, to assess whether a reasonable jury had a legally sufficient basis to render the verdict that it did or, in this situation, whether a jury has anything to decide. It cites a case, Gibson v. Old Town Trolley, a 4th Circuit case, which it spins as showing that even if the jury accepts the facts as presented, the judge has to still look, on a 50(b) motion, whether the jury had a legally sufficient basis for finding for the plaintiff.

    The judge knows of the case, and he points out that it was tied to the peculiarities of prima facie law employment discrimination cases, not antitrust. He can't judge credibility of witnesses on a Rule 50 motion, he states firmly.

    So it looks like strike one for Microsoft, although Tulchin tries to argue the point further, but then suddenly the judge says, without explanation, that he's persuaded by Tulchin that he ought to take this motion seriously.

    Tulchin then argues even more, saying that it's not enough to have a "scintilla" of evidence (which obviously Novell has, that email alone would be at least that). Novell has to have more than that, Tullchin argues. And then he turns the page on his binder to page two. And his theme becomes this: that there has never in the history of antitrust been a case like this, that no private plaintiff has ever recovered damages on a "cross-market" theory of harm to competition. He distinguishes US v. Microsoft, as of course it was all about that very thing, saying it was a *government* enforcement action, so to him, it doesn't count, I guess.

    His second point is that Novell conceded at trial that if Novell were forced to adopt the third prong of the middleware theory, about middleware exposing sufficient API so that a full productivity application could be written, it has no case.

    Yes, says the judge, he's been thinking about that too and he wants to ask Novell's Mr. Johnson when it's his turn to speak to answer his questions about that. He realizes that Dr. Noll testified to the contrary, but it makes no sense to the judge. To him, the whole theory of middleware must be that you have to expose as many APIs as the operating system.

    Lordy. Following tech trials is so frustrating sometimes.

    Mr. Tulchin is thrilled, of course, and says that the judge is exactly on target, that because Novell's case was tolled by the government's case, it was because the complaint adopted the same theory. And, the judge jumps in, Novell presented a theory to the government that it didn't pursue, so how can it be the same theory of the government case, if the government didn't pursue it? And now Novell wants to introduce some exhibits, apparently tied to that unpursued theory, and that "almost" makes the judge "angry."

    This is almost like watching a puppet show, where two puppets keep hitting a third puppet over the head with a baseball bat every time he puts his head up. Mr. Johnson is standing right there while these two describe his case in negative terms, like two buddies at a bar, who, the more they drink their beers and shoot the breeze find they like each other more and more.

    Tulchin then suggests that Novell is stuck with the government case, and it can't deviate from it, because that's how it got the case tolled. Plus, in 2004, Novell provided a release to Microsoft, one that was "very broad", Tulchin says, with three exceptions, only one relevant here, for claims set forth in the draft WordPerfect complaint. So Novell, Microsoft argues, is stuck with it. It can't add new theories now, new allegations.

    Why argue so hard about this? Because Novell wants to accuse Microsoft of deception, and Microsoft says there's no such claim in the complaint. What could Microsoft possibly say in defense to an allegation of deception?

    Wait, the judge breaks in. Novell says you didn't raise this in your original Rule 50 motion, so you can't raise it now in the renewed, post-trial motion, but you, Microsoft, say you did raise it. And Mr. Tulchin says they did raise it. It is preserved. He doesn't say what page you'll find it on.

    And Microsoft's point 6 is that one thing in an antitrust analysis is harm to the relevant market, and it seems to Microsoft the judge should look at what the market looked like before the conduct complained about by Novell occurred. Microsoft had a monopoly long before 1995, Tulchin argues, and it had it still long after the events in this litigation. Novell didn't argue at trial that Microsoft got its monopoly unlawfully, only that it maintained it unlawfully. And then all the witnesses, even Novell's, told the court that Windows 95 was a great innovation, a breakthrough. And Novell wanted to be on this great product. Novell's witness, Robert Frankenberg, and Dr. Noll, both said Microsoft's market share would have been much higher if Novell had had its products ready for the launch of Windows 95. The whole point of antitrust law is to encourage companies to come out with superior products, and if you maintain your monopoly by having great products, better products, that's lawful.

    I think you can see how Microsoft is following the line in that article about Trinko and the rest, can't you? But I have a question: Does Microsoft believe that when it's Google being examined over search that if they are doing things for business reasons to benefit users that it's not antitrust? Google Search is incredibly superior to all those complaining about the matter. Just thinking out loud, as the good judge says from time to time.

    Anyway, after listening to Microsoft's arguments for some time, the judge says that he thinks deception can be an antitrust claim, but there's some kind of line as to how much deception counts, and he's thinking about that. The sticky wicket for him is, since the APIs were known to be beta, and the industry knows what beta means, why couldn't Microsoft make any changes it pleased, without crossing the "sufficient deception" line? This is about Microsoft not sharing its IP, the judge says, which it alone developed, the product of its own "ingenuity and capital investment".

    Uh oh, Novell. That's a window into the judge's mind. Of course, Microsoft "couldn't agree more".

    Novell couldn't get its suite out in time because QuattroPro wasn't ready. Wait, says the judge, it was "code complete", the testimony brought out. That's the "scintilla" of evidence, Microsoft responds. No, says the judge. It would speak to evidence, facts that a jury would need to rule on, no? Microsoft cites a 4th Circuit case, and the judge asks for one in this circuit, the 10th, and Microsoft says it's where the need for more than a scintilla of evidence came from. Anyway, the judge says, there's no evidence as to what is the difference between code-complete and ready for distribution. Could it have been ready if the APIs were not withdrawn, and Novell could have finished the rest of the suite?

    It wasn't ready, Microsoft stresses for pages and pages, every which way, and then it argues that anyway, the APIs were still there, and Novell could have used them, accepting the risk of future changes for the time being. Maybe that would have been a lesser risk than not getting its product out in time. This rings the judge's bell, for sure. He sees that as a solution Novell could have used, which is why covering tech trials is so frustrating. The judge has an excuse, as he's not a programmer. But does Microsoft really think that would have been a solution? Really?

    The judge says that the testimony of Gary Gibb is so important that without his testimony, in the judge's opinion, this case would be over. Here's his testimony, so you can follow, but the essence of it was that Novell missed the launch deadline for Windows 95 because of the lack of cooperation from Microsoft, not because of Quattro Pro, that at first Novell did try to stick with the APIs, taking that risk, but Microsoft completely stopped helping them, so they gave up after a while and tried Plan B. Mr. Gibb is a nice man, Microsoft concedes, but his testimony can't outweigh all the other evidence contradicting him.

    Then, on page 80, the judge reveals that he's been thinking about the reversal of his original ruling, reversed by the Fourth Circuit Court of Appeals, and he's thinking that the evidence at the trial supports him, not them, so maybe the law of the case doesn't apply.

    Wow.

    He hopes it's not just his pride talking. Maybe he's wondering if he does it again, will he be reveresed again. Here's his issue: He ruled that when Novell got Caldera to sue Microsoft, it was about MS DOS, about operating systems. This is supposedly about middleware, but the middleware theory "depends upon WordPerfect and PerfectOffice having been popular products". But his question is, if they were popular in DOS, and now up comes Windows 95, so if Novell was ready at launch, Novell could leverage its position in the DOS market to a higher percentage in the Windows 95 market. How is that not "indirectly" related to DOS, and if so, why wasn't this claim released by the agreement with Caldera?

    Oh, brother. Novell just can't catch a break with this judge, who exalts minitia of his own invention to find against Novell, over and over.

    On page 89, Microsoft sums up its five reasons the judge can say yes to Microsoft's motion. I mention pagination so that when September gets here, and you can read the transcript for yourself, you can find things. The five are:

    • No duty to deal or cooperate. Microsoft continued to help Novell in any case. Microsoft had business justifications.
    • Novell argues deception, but there's never been a case that deception of a competitor forms the basis of a Section 2 claim.
    • No harm to the relevant market. The middleware theory isn't supported by evidence.
    • No evidence the decision to withdraw the APIs is the reason why Novell experienced the delay to market. Well, there's a scintilla. But Novell had other options.
    • There can't be damages when the evidence doesn't support that Novell ever planned to get the product out before December, not August, when Microsoft released Windows 95.

    That ends Microsoft's three hours of oral argument. And they break for lunch.

    Novell's Turn at Oral Argument:

    Novell has a binder of materials also. Jeff Johnson opens by reminding the judge that they are there to decide a Rule 50(b) motion. "We're not here to decide who's going to win the case. We're here to decide whether a jury should decide who's going to win the case," he states.

    Oh, and that Fourth Circuit case, Old Town Trolley? Microsoft filed over 200 pages of brief on this motion, and it never once mentioned that case until this day, in oral argument. But in any case, this is the 10th Circuit, but even if one accepted a 4th Circuit case, it can't trump the US Supreme Court on the matter. He'll send a letter to the judge about the case, once he has an opportunity to research it, if he has anything else to say about it, but it disturbs him that Microsoft made it "a centerpiece of Mr. Tulchin's argument."

    It's only on TV that lawyers surprise one another with cases or evidence they didn't know about in advance, if everyone is playing by the rules, in other words.

    In all those 200 pages, and at oral argument, Microsoft neglected to mention the Jeff Raikes email [PDF] to Warren Buffett, where he admitted that Microsoft widened the moat protecting its monopoly by "controlling the Office productivity applications sitting on top." That document, which you can find as text by searching for exhibit #2758 on this page, he argues, is a direct admission from a high Microsoft executive "that killing WordPerfect served to increase the barriers to entry in the operating systems market," and the court of appeals highlighted that document, yet Microsoft never mentions it, thus failing to address Novell's evidence, which on a Rule 50(b) motion, the judge, at least, has to do. In fact, it has to view all evidence in the light most favorable to Novell, not Microsoft, and make all reasonable inferences in favor of Novell.

    We'll see if he does, when he rules on this motion.

    If there is conflicting evidence, it's for the jury, not the judge, unless no reasonable jury could find for Novell. Microsoft in its arguments did the exact opposite track, resolving all conflicts in favor of Microsoft, but the US Supreme Court has ruled that you can't do that on a Rule 50 motion.

    There are only two questions to resolve here: 1) did Microsoft's conduct violate Section 2 of the Sherman Act? and 2) did that conduct cause antitrust injury to Novell? The first focuses on the effects of the conduct on competition in the relevant market, and the second on the effects on Novell, and those two questions can't be conflated. Novell has presented evidence that a reasonable jury could conclude showed that Microsoft used its power to tighten its monopoly power.

    He cites a case, Reazin v. Blue Cross/Blue Shield, where the 10th Circuit stated that a monopolist can only maintain its market power by maintaining the barriers to entry, and that harm to competition question was addressed by "substantial evidence" that what Microsoft did it did to do precisely that by increasing the barriers to entry to the operating systems market. Raikes, in his deposition, admitted that in that email to Buffett, he was talking about Microsoft Office, that suite. Back in 1994, WordPerfect's installed base was virtually identical to Microsoft's. By 1997, WordPerfect was more or less wiped off the map, with Microsoft now having a market share of 95%. So by doing what it did, it did "widen the moat" and protect its monopoly, even increasing it. Professor Noll concluded that this increased the applications barrier to entry in the operating system market, and a jury, a reasonable jury, could accept that. That's the jury's job.

    Microsoft argues, Johnson continues, that it would have had an even higher market share if it had fully documented and published the APIs in question. But when a monopolist does something that makes no economic sense, apart from its harmful effect on the competition, that's strong evidence of anticompetitive conduct.

    Where is the evidence, the judge asks, that Microsoft had any loss of short-term profits? Both Adam Harral's and Robert Frankenberg's [part 2] testimony, Novell says. The judge say no, that the testimony was that Windows 95 would have had greater sales, but that "says nothing about the application side".

    Because no one buys Microsoft Word to run on Windows, presumably, I say with a curled lip.

    The judge's thinking is that even if it lost on the Windows 95 side, it could pick up on the Office side.

    Uh huh.

    Johnson can't answer with a curled lip, so he says that at the time, WordPerfect has millions and millions of users, both on DOS and on Windows 3.1, and it was a very loyal customer base. Microsoft's own witnesses admitted that operating systems don't sell themselves.

    I just don't get it, the judge says. There's no proof of this connection between Windows 95 and the Office suite.

    Except real life, I would say.

    Let me address that, Johnson more tactfully says, although it makes him feel like it is Groundhog Day. Professor Noll answered your question, away from the jury, he says. He said the simple answer is that you can't look to whatever profits Microsoft made in applications to offset the loss to Windows 95. I remember now, the judge says, but it doesn't make any sense. "I know he's a Stanford guy and very smart, but it doesn't make sense," the judge argues. This is about alleged harm to the *operating systems market*, so the court has to focus on harm *in that market*, Johnson helpfully points out.

    What about common sense, the judge replies? Why would you sacrifice short-term profits unless you are doing it for some other incentive? If it can make more selling Office to offset any loss in selling Windows 95, why isn't that relevant? Not that there is any evidence that Microsoft lost a penny, the judge insists. Novell's theory, he scornfully says, is that Microsoft made lots of money, "because it let Jay Leno, but not Gary Gibb, into the tent." That's not our argument, Novell tries to interject, but the judge insists. "You said this was an academic question," Johnson says. "It's not an academic question," the judge now says. "It's a question of inference." Lordy. I guess the only real response to that would be, it depends on who is doing the inferring and whether they are struggling to infer in favor of Microsoft. But I digress.

    Novell shows him on his screen the answer Professor Noll gave to this exact question, an answer from "an expert in antitrust economics in this case" and he told the judge no, you can't offset like that in this situation. It's about anticompetitive actions, and if it had more sales of Office, did it result from anticompetitive conduct, "and of course," Johnson says, "it did." If WordPerfect had been available at that Jay Leno-attended event, in the next tent over, shall we say, a lot of the Office sales would have been WordPerfect sales and Microsoft would not have gained as much. They didn't get those extra sales by producing a better product. They did it by wiping out WordPerfect, so it couldn't be there on launch day. It's all anticompetitive conduct, in both Windows 95 and in applications, and you can't ignore the anticompetitive conduct in either. A reasonable jury could conclude that what Gates did made no business sense apart from the harm to competitors.

    But is that actionable as an antitrust claim, the judge asks. If Gates was worried about WordPerfect being better than Word, and it withdrew the APIs to give Word time to catch up, is that actionable? And Novell says, yes, if it caused harm to competition in the operating systems market. It goes right to the heart of the widening the moat issue. The Sherman Act doesn't require Novell to prove intent, even though this judge "is fond of talking about Mr. Gates and his intent", Johnson pointedly says.

    At this point, we're on page 113 of the 261-page PDF, and all I can think about is, this just shouldn't be this hard.

    Novell next addresses the question of whether deception can be an antitrust claim, and Johnson points out that there is all kinds of case law that says that there is no definition of what anticompetitive conduct is. You look at the facts, situation by situation. And for Microsoft to claim there's never been a deception case, when one of the headings in the DC Court of Appeals opinion in the US v. Microsoft case is "Deception of Java Developers", Johnson says, is a bit much. "I believe Microsoft paid out quite a bit to Sun to get rid of their Section 2 claim," Johnson says, and that claim is "virtually identical" to Novell's Section 2 claim.

    Business torts, it's true, Johnson continues, standing alone don't create an antitrust violation, but when they result in harm to competition in the applicable market, they indeed do.

    What about the fact that your complaint doesn't mention the word "deception" or Microsoft's argument that everything not in the complaint was released in 2004, the judge asks. That release, Johnson says, specifically said that nothing in the release shall prevent Novell from raising any facts in support of its cause of action that it wishes to raise. We are not required to list all our evidence in a complaint.

    Microsoft, Johnson says, urged Novell to make an investment in the APIs to create a product for Windows 95. Then Gates withdrew the technology for the express purpose of harming Novell. Did Gates know Novell was using the APIs, the judge asks. Two days after that email, Brad Silverberg sent Gates a memo, Johnson answers, that informed him if he didn't know it before. Which he obviously did, judging from the Gates email, speaking of inferences. In any case, that is an issue for the jury. It's a fact question, Johnson says. And since intent isn't the right issue, what difference does it make anyway? But addressing it, Johnson reminds the judge that a Microsoft employee said there would be "hell to pay" if Microsoft did it. Other Microsoft people said WordPerfect was using the extensions. There was a meeting with Microsoft and Novell, where Novell's employee, a Mr. Cole, said how excited Novell was about the extensions. The judge says there's conflicting evidence, about an email saying Novell was "OK with it", the withdrawal. And Novell patiently says, once again, that conflicting evidence is for the jury, and we're here on a Rule 50 motion.

    The judge interrupts and say he knows exactly where he is. He wishes he wasn't here. If he were a fact-finder, as opposed to the judge, Novell would lose on delay, so don't remind him he's here on a Rule 50 motion.

    That's clear without me saying one word, is it not?

    Johnson then changes the subject. Let's look on the middleware side of this, he suggests, since there's enough evidence we've discussed for a reasonable jury to find harm to competition, just from the Raikes email and the Dr. Noll testimony.

    In the US v. Microsoft case, middleware was defined as software that relies on the interfaces provided by the underlying operating system while simultaneously exposing its own APIs to developers. The DC Circuit appeals court agreed, writing that middleware refers to software products that expose their own APIs, Johnson continues. Microsoft is arguing for a different definition, one that requires middleware to be a complete operating system, based solely on their own APIs, in order to impact competition in the operating systems market, then Java wouldn't have had a Section 2 case. Or Netscape Navigator either. Neither could do that. It's absurd, Johnson says. Microsoft paid millions, hundreds of millions of dollars to settle the Netscape claims. Is that in the record, the judge asks? Incredibly asks. How do you know, he inquires?

    I can't believe what I'm reading.

    It's publicly available, Johnson tersely replies. Is it in the record, the judge then asks. I don't know, Johnson says. Well, the judge triumphs, it isn't, meaning the jury didn't hear about it, thanks to his ruling that they couldn't hear about it, and we're here on a Rule 50 motion, "as you've just reminded me." So he can't consider it on this motion.

    For Novell, this judge has set up a Catch 22. Unless the payment is mentioned in the Findings of Fact, the judge won't consider it. He has another question: if we're talking about barriers to entry into the operating systems market, if middleware doesn't expose sufficient APIs to allow ISV's to write "full-featured personal productivity applications", how is it relevant? The judge says he's confused, and I'd heartily agree. Johnson says if middleware could do that, you wouldn't need an operating system under the middleware. Duh. This is about reducing the applications barrier to entry, not destroying it. Novell's middleware was cross-platform. It did run on multiple operating systems. And it had sufficient APIs to encapsulate meaningful functionality, Johnson explains, which is why it was a threat to Microsoft's operating system monopoly power. Novell had many partners writing applications based on WordPerfect's APIs. WordPerfect had a real market opportunity at the time.

    Why doesn't this relate to DOS, then, the judge again asks? Because the court of appeals, Johnson answers, ruled that the contract with Caldera only applied to the 13 identified products in the contract itself.

    The judge asks more loopy questions, what about this possibility or that one, and Johnson patiently reminds him of trial testimony or why it doesn't matter, and we're only up to page 132. It's frankly unbearable to me to read. The judge is looking under every stone and pebble and grain of sand for something contrary to Novell. Maybe it's his method, but you don't see him doing that with Microsoft in the same hearing on the same day. So to me, it's now abundantly clear that this judge does want to rule for Microsoft, but he's not sure how to make it happen.

    As for Microsoft's argument that Windows 95 was this great innovative breakthrough, Johnson reminds the court of all the testimony that Apple had already moved to a GUI, that Microsoft was just bringing Windows up to speed with what Apple had already done. Gibb testified to that, and Mr. Sinofsky. It was a paradigm shift, but not Microsoft's. And WordPerfect had to be successful on Windows to do all the rest of what it intended to do, because that's where the action was. But those loyal WordPerfect users, down the road, would be asking, since WordPerfect runs on Linux and Microsoft, and Linux is free, why am I paying for Windows? That was the threat.

    But, the judge asks, so what if WordPerfect was cross-platform in the past. Windows 95 was changing the world, so if WordPerfect wasn't cross-platform at launch, would it have been in 1996? I think that's the question. His questions are so strange, I'm never sure I really understand them, because they make so little sense. He is stuck on the idea that there is no evidence Novell would have invested in WordPerfect to make it cross platform going forward.

    Huh?

    An expert in the case, a Mr. Murphy, testified that by 1998, Linux was comparable in size, capability and complexity to Microsoft's Windows 98 and NT. It had by then five to ten million users. Note to IBM: SCO claimed otherwise, that it was just a bicycle, not a race car, until IBM got involved in 2001. Just saying.

    Johnson shows the judge that in June of 1994, in a filing with the SEC, WordPerfect was available in 23 languages on all of the most widely used platforms, DOS, MS Windows, UNIX, Apple, VAX, VMS, etc. A reasonable jury could conclude that WordPerfect was, in fact, cross-platform.

    Ah, but we are dealing with what I'd call an unreasonable judge, not the jury. They already were 11 to 1 for Novell. It's the judge Novell can't seem to convince.

    I'd give Novell's Mr. Johnson a gold star for never giving up. But when you get the transcript, I think you'll agree with me that this is all stupidness. Having to prove that WordPerfect was cross-platform and would naturally want to continue to be is just stupid. I'll stop here, actually, because it's affecting my view of the things enough that I can't write about it clearly and fairly without a curved lip *and* a snort. And I insist on trying to show respect to judges, so I have stop before I find that impossible. The rule of law depends on respect, and that applies to me too. But I've shown you enough that you won't be utterly amazed and caught off-guard if this judge rules for Microsoft.

    Why, though, I can't help but ask, if the judge knows so little about Microsoft's antitrust activities -- Microsoft is no antitrust virgin, shall we say -- and knows so little about the tech, as he himself admits, then why did this judge from the East insist on commuting to Utah to handle the trial when the court of appeals sent it there? Surely there are other judges who do understand tech issues better than this, as the Hon. Judge William Alsup so ably demonstrated in Oracle v. Google. Why insist on staying with a case you don't understand?

    I'll show you the transcript as soon as the court allows it. And as we continue to go through the trial transcripts as text, if there is evidence that relates, I'll try to highlight it.


      View Printable Version


    Groklaw © Copyright 2003-2013 Pamela Jones.
    All trademarks and copyrights on this page are owned by their respective owners.
    Comments are owned by the individual posters.

    PJ's articles are licensed under a Creative Commons License. ( Details )