Slashdot Mirror
Interview: Antitrust Experts Respond re MS
1) by Robotech_Master, adding to c+era's original question...
What are the chances of Microsoft being forced to open the
entire source code to Windows, not just its APIs? Some
articles I've read/listened to bandied that about as a
possible solution.
What sort of software license would an "opened Windows" be most likely to employ?
What would the short- and long-term effects of such a release be? (Other than a sudden dreadful attrition of open source programmers as about half of those who take a look at the Windows source code die laughing. :)
Don Weightman responds:
As the questioner says, the details are critical. Rather than pure open
source, think of a scenario where the source code is published on the Net,
but without permission to re-publish or use it. Instead the idea would be
to have the global community of programmers do a massively distributed peer
review to find the bugs and traps and propose fixes. (Also see my answer to question # 12.)
David Niemi responds:
The chances of Microsoft being forced to open the source code to MS
Windows are slim to none -- because it is a novel and drastic approach,
and because it is already being trumpeted by Microsoft's political
allies as "government seizure of Microsoft's intellectual property".
Code licensed from other companies which has been incorporated into MS
Windows would also be very problematic and would be hard to solve
without Microsoft's cooperation (and Microsoft can be expected to fight
such a remedy tooth and nail).
Richard Hawkins responds:
This would amount to a "taking" by the government, which is not
possible without compensating Microsoft for the value of the source
code. While some level of punishment will occur, this is capital
punishment, and I don't see any court upholding such an action. In the
past when monopolists have been forced to give up entire divisions,
they've still been paid for them.
John Lederer responds:
I think it unlikely that the source code would be opened, if by that you
mean free use of the cource code. I could conceive of it being required
to be public in the sense that one could read but not copy, as part of a
larger remedy.
I think several of these questions reach the most difficult part of the Microsoft case -- what's the remedy?
Splitting the company into say, three companies, "The Micromicrettes" or "The Baby Bills", ends the monopoly. However, it invites the development of incompatibilites and does little to prevent any of the Baby Bills from using some of Microsft's nasty techniques to make sure that no other company beyond the Bills can compete.
Making the source code true open source is both drastic and a seeming taking of property. However, splitting the company vertically and requiring the baby Bills to publish source code and/or API's might be a solution. It dioes not "take" the property which is still copyrighted, but it does provide a means for the Baby Bills to provide compatibility and for other operating systems to develop that can use the established base of applications.
Something along that line, along with a judicially imposed requirement of same terms and prices to OEMs and a prohibition on concerted action by the Bills, would be my suggestion. Frankly, I don't have the slightest idea what the court will do.
The Court cannot throw anybody in jail or impose any fines -- this is a civil action for injunctive relief not a criminal action.
2) by Greyfox
I'm curious about the potential for legal action outside
the DOJ's case now that it's been found that Microsoft is in
fact a monopoly and has used their position to stamp out
competition.
If I spent my time and money to become certified in a product competing with Microsoft (Say, A Novell or OS/2 Certified Engineer) and that product was subsequently stomped into the ground by Microsoft, could I justifiably sue Microsoft for making that investment of time and money worthless?
On a slightly related note, if I owned stock in one of their competitors who was eliminated through their dirty tricks, could I sue them for that?
Richard Hawkins responds:
While I won't absolutely rule it out, I would expect that the harm you
suffer is both too remotely linked to microsoft, and too difficult to
quantify, to maintain an action.
John Lederer responds:
Generally federal antitrust law disfavors actions by those who are not
direct customers or are not competitively injured. However, states have
their own antitrust laws, and some states, notably farm states, have
broader provisions.
3) by Myddrin
I've read in many places that the FoF is hard to overturn.
Reading the FoF it seems to me that if you accept it as
fact, then you have to conclude that MS broke the law and
should be punished (harshly).
Assuming the FoF is unappealable then, where is the opening for Microsoft to appeal the decision? I.E. Assuming the FoF is untouchable how could MS "win on appeal" as we keep hearing from various tech analysts?
Don Weightman responds:
Two possible courts here. If the trial judge and the Supreme Court agree,
then either party can
go straight to the top, rather than intermediate appeals court. If I were
DOJ, I'd rather be in front of the Supreme Court, I think - if only because
MS won in the lower appeals court, last visit, and so you think MS would
rather be _there_.
Except the odds and strategy change if the remedy starts looking radical - this Supreme Court is pretty protective of property rights. For the most part, though, in contrast to, say, abortion rights or something, Capital P Politics count less in antitrust - it's seldom a matter of a straightforward read of a case result from the political affiliation of the judges.
Richard Hawkins responds:
Conclusions of law are easily appealable; the higher court
automatically substitutes its own judgment on the law. Factual
findings, however, whether by judge or jury, are almost impossible to
appeal. The trier of fact (here, the judge) actually sees the evidence
and hears the witnesses, observes body language, nervousness, etc. To
disturb a factual finding, the higher court pretty much has to find
that no reasonable person could have reached that conclusion from the
evidence presented. If it could have gone either way, the finding
won't be disturbed, even if the appellate court disagrees.
The findings are tailored to the "elements" that must be proven. In a murder case, you might have findings that the defendant took the life of the victim, that the defendant acted with criminal negligence. These meet the requirements for manslaughter. Change the intent from criminal negligence to a reckless disregard for human life, and these findings would support a legal conclusion of murder. Add premeditation and intent to kill to the factual findings, and you have capital murder.
Judge Jackson's findings are set up similarly, but instead of the elements of murder they represent elements for antitrust law. And they satisfy the requirements for finding monopoly power, abuse of the power, etc. For that matter, of the laundry list of illegal acts by monopolists, I can't think of any that aren't covered in the findings. I could teach most of a course in antitrust law from these findings.
The bottom line is that the rest of the case is really about *what* remedies will be implemented, rather than *whether* they will be implemented--and this includes appellate review.
John Lederer responds:
I think the findings of fact will hold up on appeal -- the barrier to
overturning a finding of fact is quite broad. I also suspect that the
conclusions of law will be upheld unless the appellate court is bound
and determined to reverse and willing to indulge in some pretty dodgy
reasoning. However, I think the potential for an appellate court
messing around with the remedy is quite strong.
4) by chromatic
From a legal standpoint, can Microsoft point to such
developments as the Sun/AOL/Netscape deal and the mainstream
attention devoted to Linux and the BSDs and say, "Things
move too fast and we're struggling to catch up" and avoid or
beg down punishment?
David Niemi responds:
Microsoft has tried like crazy in the courtroom and in the political
arena to claim that MS Windows is beset by competitors, but their
credibility is hurt by the fact that they tell business and stock
analysts just the opposite.
What really matters is whether anyone else has enough market power in the relevant market (Intel-compatible personal computer operating systems) to deny Microsoft monopoly power; mere existence of a potential competitor is not enough. Competitors in related markets like servers and browsers (or in specialized niches) clearly haven't prevented Microsoft from keeping the price of MS Windows high, nor have they prevented it from imposing increasingly less favorable licensing terms on its customers.
Microsoft keeps saying that real competition in the relevant market is "right around the corner", but this is only idle, self-serving speculation, considering that they have dominated this market since the mid-1980s. It will take Linux several years to introduce meaningful competition in the desktop operating system market even if Microsoft is forced to abandon its anticompetitive tactics. No other operating system even aspires to being a direct competitor at present.
Richard Hawkins responds:
Avoid, no. Beg down, yes. The remedies are supposed to punish as
appropriate, but are also supposed to solve the problem. If the market is now
competitive, there is no need to now break up MS. However, this issue
was addressed in the findings; the time for this argument (which
Microsoft made) has passed, and they lost the argument.
Nonetheless, it's still possible (though highly unlikely) that the judge fines Microsoft $1 and lets them go on their way. The USFL won its antitrust suit against the NFL, and was awarded one dollar in damages--but it was tripled. :)
5) by jabbo
Let's say that George W. Bush gets elected, and as I have
been told he is wont to do, embraces a laissez-faire,
hands-off style of conservatism which places a more
conservative judge on the Supreme Court (assuming one or
more of the present justices were to die or step down).
Let's assume further that these judges feel that Microsoft
has done nothing so wrong as to merit real punitive action.
What is to stop such a bench from favoring Microsoft and their beyond-hardball tactics? Are Judge Jackson's Findings of Fact so unassailable that, by the rules of the game (as it were), their content cannot be ignored or overruled even by a higher court? (from the paucity of cases that I have read in classes, it seems that the Supreme Court justices can do just about anything short of striking down an existing law) Moreover, if justice is actually meted out, is it possible for Microsoft to simply buy enough senators/congressmen that new laws protecting their investments in the role of Digital Media Broker To The World will favor MS or the Baby Bills?
In short, Can this finding be made to stick, even with all of Microsoft's money aligned against it, and even in light of a potentially conservative government coming to the fore?
Richard Hawkins responds:
There's a common misconception about the current makeup of the US
Supreme Court, namely that it has two wings, conservative and liberal.
At the moment, it has three: conservative, (modern) liberal, and
classic liberal, with the third wing consisting of Scalia, Thomas, and
Kennedy (on his good days). Kennedy bolts the the conservative block
quite frequently, and Scalia's knee tends to jerk there when police
safety is involved. Thomas remains there consistently. This block
votes most often with the conservatives, but at rather predictable
times votes with the liberal block. These tend to be cases involving
the reach of state power---the liberals and conservatives disagree
over *what* to do with the government authority, while the classic
liberals tend to distrust *all* government actions.
The type of judge you describe would fall into the third (classic liberal) block. Shifting my emphasis from lawyer to professor of economics, laissez-faire approaches do *not* tolerate monopolies, but instead find them repugnant. Monopolists interfere with the cherished operation of the market, and need to be slapped down. I don't see such appointments as helping microsft; they'd be better with *either* conservative or liberal judges.
Anyway, assuming some kind of warm disposition towards Microsoft, it would still be difficult for the court to overturn the findings of fact--see the answers to 3). However, the court could come to a legal conclusion calling for a far milder penalty.
John Lederer responds:
I would be less concerned about judges appointed than I would about who
gets appointed in the Department of Justice. However, 19 state attorney
generals will presumably not change at the same time, and they will have
a significant effect on the plaintiffs' positions.
6) by dieMSdie
Many people seem to be considering the breakup of Microsoft
into 2 or more separate companies as the most likely outcome
of all this. However, would that really do us any good?
Let's say you broke them up into an OS division and an APPS
division. What is to stop the OS folks from sharing their
secret API's with the APPS company? Would we not just be
right back where we are now? What would all of you recommend
as a solution that would allow MS to survive, but not as a
predator that destroys all competition and stifles true
innovation?
David Niemi responds:
Breaking apart operating systems, apps, and online services (vertical
divestiture) would do *a lot* of good. It is much easier to detect and
prosecute price-fixing and collusion between separate companies than it
is to detect internal conflicts of interest inside a single company.
The different companies would then be required to separately reporting
profit and loss to their shareholders, and it would be next to
impossible for them to justify helping the other parts of the former
Microsoft at their own expense (whereas today, that happens all the
time).
Richard Hawkins responds:
No argument here, but a bit more.
"There is no honor among thieves." Why would these companies help each other? That means giving up their own advantage, and after a split there would be no reason to do so. The OS division ends up with just as much reason to help netscape as the former microsoft division.
David Niemi adds...
Vertical divestiture would be immediately painful, and would not
immediately create competition. But in the long run it would be one of
the most effective ways of leveling the playing field without the use of
intrusive restrictions on the new parts of Microsoft and their "freedom
to innovate".
And Richard Hawkins says...
I don't agree with this part. Vertical divestiture would leave the
underlying windows monopoly intact, and not solve the problems. The
case was *not* about including IE with windows; it was about the
actions taken to protect the windows monopoly from other competition.
I agree that intrusive restrictions are unlikely to work. I would prefer some type of horizontal split, but will leave that for 12)
John Lederer responds:
I strongly feel that the operating systems and applications need to be
broken apart. However the government(s) in this case deliberately
decided to drop the charges that Microsoft was monopolizing
applications. Regrettably, I think that weakens a bit the rationale for
separating the two. One is left with the application contributing to
the operating system monopoly rather than the stronger, and unbrought,
case that the operating system and its sales methods cause a monopoly
in applications.
However, the court may suggest as a remedy most anything. There is case
law that where a monopoly is "structural and pervasive", the remedy
ought to also be structural.
Don Weightman responds:
This remedy, standing alone, leaves the "OS market power" intact - what you
can do when you control Windows. There's a danger of bad habits moving into
the successor organization, and the temptation to exploit that good old
Windows monopoly would be strong. So maybe you need disinfectant.
One trick would be to require disclosure of current API's, and predisclosure of new ones doesn't really address Microsoft's talent for strategic conduct, and would still require some kind of ongoing supervision from the legal system.
What I like better is "condemning the asset" and having a forced judicial sale. The successor - "Washed Windows"? - would then publish the source code, as I discussed responding to #1, and sell "clean" Windows, with a must-carry-customer-choice condition for browsers. You'd maximize interoperability and end-user freedom, ant the expense of a design freeze and loss of innovation at the OS level - except for competitors. To be blunt about it, I'm beginning to think that "Windows innovation" is an oxymoron - or else code for "try and stop us from bundling another application into Windows and killing another market". Which is what a remedy is supposed to prevent.
So you divest and disinfect Windows, and you probably lose OS innovation. The hope would be that innovation would happen, on the Net (and in application layers we can't even imagine) without entrepreneurs having to look over their shoulders at where and how Microsoft might be coming after them using Windows as an invasion beachhead. Or those who want to, can innovate on top of what would be - on my assumption - a stable platform. Stable not just for the PC, but for the PC-as-gateway-to-the-Net. Not such a bad result, as I see it.
As to business applications (Office & its competitors): assuming -- and it's a very big assumption -- that you can deal with enduring market power back in the OS market, then wouldn't vertical divestiture solve the problem? I'm assuming here that, even though the Microsoft applications have a big installed base, once separated from the bottleneck, Word and company would have to compete on the merits.
7) by Otter
One of the most startling things about this trial was how
utterly inept Microsoft's defense seemed. Who do you think
is at fault there -- Sullivan & Cromwell, MS leadership or
both? I find it hard to believe that such a prestigious firm
could keep botching things so badly, especially when the
media
were full of stories ridiculing them. My guess is that MS
wanted the defense to go a certain way (give no ground on
any front) and the lawyers felt obliged to go along. What do
you think?
Don Weightman responds:
Probably this was mostly a case of the lawyers being losing control of
their case to the client. I haven't checked the credentials of the counsel
who actually went into the trial, but, it turns out, perhaps surprisingly,
that even senior attorneys in large firms like the one representing
Microsoft tend not to have a lot of experience in actually trying cases
with witnesses and stuff, and maybe inexperience was part of the problem here
John Lederer responds:
I don't know what actually happened between Microsft and its lawyers. I
assume that Microsoft's lawyers advised their client that the case was a loser at the
trial level because the evidence against them was so strong. They
may have well decided that it was still worth fighting as part of a
larger picture.
The trial may represent the effort at a best defense in the face of damning facts. However, they certainly seemed to muff a lot of details -- the video presentation, Gates' deposition, their economic expert, etc.
8) by Effugas
Is it a) Legally possible and b) Remotely conceivable that
Microsoft and the U.S. Government will come to a settlement
that will insulate Microsoft from any and all lawsuits based
upon their abuses over the last ten years?
David Niemi responds:
The DOJ does not have the authority to grant Microsoft immunity from
other prosecutions or private lawsuits. Congress could conceivably
craft
legislation to accomplish something along these lines -- quite a scary
thought!
Richard Hawkins responds:
Congress couldn't do that without paying everyone -- there is a property
right in the right to litigate, and this would be a taking in violation
of the Fifth Amendment. However, it's conceivable that Congress might
set damages by statute (the prohibition on ex post facto laws applies
to criminal cases, not determinations of what civil law means), but
this would guarantee years of constitutional litigation.
John Lederer responds:
Seems pretty unlikely to me that Congress would do anything. They
generally are pretty resistant to picking up hot potatoes.
9) by heretic
My understanding is that will next be a verdict followed by
the remedy phase.
What are the applicable statutes that apply to Microsoft's behavior? Is it limited to the Sherman Act and the Consent Decree, or is there a boatload of other law that has to be taken into account?
Regarding the remedy phase, I presume that that the DOJ and the state attorneys general will have some recommmendations. Is the judge limited to the scope of these recommendations, or is he free to devise whatever remedy he sees fit? If the judge does have a free hand, what are the limitations to his remedies?
Richard Hawkins responds:
It's largely the Sherman & Clayton acts here. I don't think that the
prior consent decree is really at issue; the judge wasn't bothered by
the inclusion of windows, but by the acts taken to prevent disabling IE
as the default choice by OEM's, to enable netscape, etc.
There are very few limits on the remedies available. Most of the damaged parties aren't involved, and there are no findings as to the amount of damages other than that the $89 upgrade price was at least $40 greater than the competitive price (Windows98), and that damages to Netscape exceeded $100M, meaning that we can't expect large damage awards. A fine is certainly possible, but it's the other remedies that are most important.
The DOJ will certainly have recommendations, but the judge can come to his own conclusions, or ask for input elsewhere.
What limitations there are will come down to the constitutional prohibitions on taking (he can't simply seize microsoft assets), and whether the remedy solves the underlying monopoly problem.
10) by bhurt
What are the chances that the Supreme Court will decline to
hear the appeal? And what justifications are they likely use
should they decline the appeal? I remember hearing that the
Supreme Court prefers to only hear cases with constitutional
implications. Could the Supreme Court refuse the appeal
based simply on that (and a full schedule)?
David Niemi responds:
Market analysts and pundits have already concluded that the appeals
process will take many years without having any idea what ruling would
actually be appealed or why. They also forget to mention that an appeal
has to have some legal basis other than merely that Microsoft didn't
like the outcome. The Supreme Court will usually refuse to hear a case
if it believes that existing case law clearly covered all of the key
issues -- imposing a heavy burden on Microsoft to come up with very
strong legal arguments indeed.
Richard Hawkins responds:
I would expect to see the Supreme Court take this directly. If Judge
Jackson rules by mid-year (briefs are all due by January), the case
could conceivably be heard in the October term (recall that this case
can bypass the appellate courts).
I think that the mixing of intellectual property law and antitrust law will be enough to get this case heard; it will significantly affect the economy, and there don't seem to be any cases with similar issues on the horizon.
Finally, they don't need any justification to refuse to hear cases. Cases are *not* heard unless four of the judges vote to hear them.
John Lederer responds:
The Supreme Court need not give any reason for refusing to hear an
appeal. The Court of Appeals which would hear the appeal if the Supreme
Court decides not to hear it, or if the parties don't seek the route
directly to the Supreme Court, must hear the appeal.
The Supreme Court dislikes the statutes that gives one a direct right of appeal to the Supreme Court, and generally are predisposed not to hear those cases if there is an intermediate appellate possibility. However, it seems to me that Microsft is such a big player in the economy that they are likely to hear it. A summary affirmance would not surpise me as a result -- despite all the screaming and yelling about novelty, new government intrusion, etc. etc., this really is not a legally groundbreaking case, save possibly on the remedies.
11) by JordanH
In your experience in various courtrooms, did Judge Jackson
exercise unusual restraint in not sanctioning or making a
finding of contempt for the apparently faked videotaped
demonstration?
There are a number of clear misrepresentations made in the video, including a Microsoft executive saying "We have not made any other changes to this computer or Windows 98, except to run Dr. Felten's program." Microsoft later admitted that this was not true.
As Judge Jackson did not mention this apparent falsification of evidence in the Finding of Fact, is it unlikely that this incident will be used to prejudge Microsoft in appeals?
Don Weightman responds:
When you
prepare a witness you're supposed to leave no question that Extremely Bad
Things Will Happen If the Judge Thinks You're Getting Cute. Either this
wasn't made clear enough to the witnesses or the advice was ignored. In
either case both lawyers and client got what they deserved: a very angry
judge.
Richard Hawkins responds:
The first rule of litigation is not to get the
judge angry at you. Perjury and otherwise falsifying evidence tends to
have this effect.
On the other hand, I generally find courts too slow to find contempt and issue sanctions for frivolous cases; this is the single biggest issue I'd focus on in reforming litigation.
John Lederer responds:
Judges very rarely impose sanctions during the course of a trial. the
damage to Microsft was huge -- not on that single piece of evidence, but
because they lost credibility. Credibility in the case is the single
largest asset a trial lawyer can have. My impression, from afar, is
that this is the point where the judge decided that the Microsft case was BS
and that led directly to the findings of fact reiterating the
government's case.
Microsoft's lawyers' response on the [Gates] tape seemed to reflect Microsoft's usual tactics when tough questions are asked of it about things like the NSA key, the DRDOS error message in the Windows beta,or the use of DOS in Windows 95. They give a flurry of beta explanations, searching for one that will hold up, They never give a a clear "here is what happened, here is how we screwed up, we apologize". The result is a loss of credibility whether in the computer community or at trial.
12) by ClarkEvans
This is the first time where breakup of a monopoly based
soley on intellectual property may occur. For monopolies
past, breakup involved splitting the properties along
physical boundaries. It is clear that the definition of this
breakup will be along intellectual boundries instead.
However, this leads to one question.
In the past, the assets were exclusive; thus only one group could control each asset after the breakup. Could the non-exclusive nature of information change the method of breakup?
In particular, it has been argued that the public of the United States (and the world at large) has already paid for Windows 98 far more than they would have otherwise; all in told billions of dollars more than they would have. So, due to the non-exclusive nature of the operating system; is putting the operating system code in the public domain a possible solution? After all, it is the abuse of the very intent of copyright law (to promote the arts and sciences by providing, for a limited time, exclusive rights) by interfering with the advancement of the arts and sciences which has been proven. Would it not make sense to simply revoke the copyright?
David Niemi responds:
I think involuntary breakup of Microsoft into parallel entities, each
with access to the same source code base, is technically unworkable and
is the result of too many literal comparisons with the 1980s AT&T/Baby
Bell consent decree. Any breakup of Microsoft needs to be along
existing internal boundaries for it to be practical and sustainable.
For example, online services (MSN, Hotmail, etc.) have absolutely
nothing legitimate to do with Microsoft's software business and would be
relatively easy to separate from it.
Richard Hawkins responds:
I really have to disagree here. My preference would be both the
horizontal and vertical splits, but I see the horizontal as more
important.
There are certainly issues to solve in creating multiple companies out of the OS division, but I don't see them as being insurmountable. Even if they are, forcing three or four unlimited licenses of the source code could provide the needed competition. That is, leave the MS OS division in place, but auction off (with MS keeping the proceeds) rights to use the code. The winners could then do whatever they want with it, most notably modify it and sell a competing OS that can run programs written for the Windows API. While it would be possible for one of the licensees to put it under some sort of open source license, this would require handling license issues with licensors of included technology, and does not seem to be a likely reason for someone to spend hundreds of millions of dollars.
Another possibility is absolute and full disclosure of all API functions. This would make it possible (at least conceivably) for another vendor to make a product that would run the applications. However, to be effective, the API's would have to be disclosed well before new versions would be released, and would be difficult to enforce--this would require *proving* that the API doesn't do what it says it does, which would be difficult without an entire working alternative OS that runs the API.
David Niemi continues...
Making MS Windows public domain would be a very drastic step and would
provide a lot of ammunition for politicians to step in and interfere
with the case.
Richard Hawkins replies...
It would also be illegal :) See my other answers. No matter how bad
they may be, it's illegal to take the property from them.
David Niemi continues...
It would be more prudent to make a legal conclusion and a more modest
remedy which can withstand appeal, and let consumers file class action
lawsuits afterwards if they believe they have a case.
RW:
Richard Hawkins adds...
I'm surprised that there have been so few suits so far. Expect them.
David Niemi says...
There are many problems with abuse of copyrights and patents these
days, but this case is not the right place to try to solve them.
Richard Hawkins adds...
Yes; don't expect anything novel in areas outside of antitrust. Also,
revoking the copyright isn't realistic. There's no real way to do
this, it would be a taking, and really wouldn't accomplish anything:
all it would mean would be that those portions of the binary owned by
microsoft could be copied freely; it wouldn't make the source code
available so that changes could be made, and it wouldn't allow the
copying of the licensed code within windows.
Don Weightman responds:
So maybe we'd need to have an auction. (I like the idea of a Dutch auction
for three to five buyers.) But auctioning off the source code without more
doesn't really get at all the sources of Microsoft control, like the
knowledge inside employees' heads of how all that code fits together, what
the interconnection and interoperability kinks are, and where the bodies
are buried. So if you had a spinoff of employees, as well as the code
asset, this would look like divestiture; see my answer to #6.
-----
A "Bonus Question Series" Richard Hawkins wanted to answer...
John Murdoch asks:
Which Victim Does the Anti Trust Act Protect?
The original Anti Trust Acts were developed to protect small farmers and businesses from predatory actions by trusts -- Standard Oil regulating who got railroad tank cars, for instance, so small oil producers couldn't compete.
Richard Hawkins responds:
There was also concern about consumers. Modern law (after Bork &
Posner started pointing out how silly the law had become) only worries
about the effect on consumers. As an example, the government filed
suit in Brown Shoe when Kinney, a retailer, and Brown, a manufacturor,
wanted to merge. Neither had more than 5% of their respective markets,
but the government sued as this would allow the combined entity to sell
a product of comparable value at a lower price than their competitors.
Oh, horror. I'm glad that the government was there to protect us from
low prices on quality goods. The merger was blocked. Bork cites this
case as a leading candidate for the worst antitrust decision of all
time. Today, the outcome would have been different--the lower prices
would mean that the merger was pro-competitive, and the effects on
competitors be damned.
John Murdoch:
Can Microsoft argue that Judge Jackson is misapplying the law?
Richard Hawkins:
Certainly not. The judge hasn't applied *any* law yet; so far he has
merely determined what happened.
John Murdoch:
Jerry Pournelle cogently points out that Microsoft's presence in any market
category has consistently driven prices lower. He particularly points out
that Microsoft has been remarkably aggressive in providing tools and support to
developers -- handing out developer tools for free to anybody who even looked like a programmer.
Microsoft has spent zillions of dollars providing tools and support to tens of thousands of small businesses through their ISV and Microsoft Certified Solution Provider program. Other vendors that have emulated those programs don't provide nearly as much, and charge much, much more for their programs.
Richard Hawkins:
Jerry Pournelle is one of my favorite authors, particularly his "Fall
of the West" type work from the 70's, and his editing of anthologies.
His descriptions of working with DOS and later Windows hardware were in
part responsible for my buying Macintoshes again rather than switching
more than once.
Here, Jerry completely misses the point. He is describing how Microsoft obtained the Windows monopoly. In fact, obtaining a monopoly through vigorous competition is quite legal, and these actions would fall into that category. The problem is, that this has exactly nothing to do with the current case.
---------- Next week: KDE Developers
Call me a capitalist swine, but, owning stock in the evil empire, the 5 baby bills would be the most profitable by far. It also might go a long way to getting MS-Office into Linux -- the major reason people claim stay with Windows. I know that there are plenty of alternatives, I have used some, but for mass adoption it will take more of a push. Look at the humble PC; it was nothing until Big Blue said it was OK for business to use.
If (and that's a huge IF) M$ code was to be opened, it would suffer from the same maladies that plague Mozilla... Only more so.
Namely, the pre-release size of it. NT2k is gauged at what? 40 million lines of source? How many Linux hackers would be willing to drop their passion and actively work on poring over that much code?? Once their curiosity was satisfied, most would just smile, nod, and go back to coding for Linux.
Sure, some would persist, a small enclave would make the betterment of the WinAPI and WinOS into a crusade, but for the most part, it's not happening. How many people would be willing to scale a poorly designed, but built sky-scraper in an attempt to fix the bugs?
Add to that the very M$ condition of there being a lot of licensed code (which would not be part of the disclosure) without which the rest is useless, and the release of Windows source to the public is a non-possibility.
-- What you do today will cost you a day of your life.
It seems clear to me that most legal scholars fail to understand the fundemental drivers of the software industry. I guess being a 'legal scholar' tends to focus one on the legal issues (Duh). Unfortunately the practical issues of either a breakup or of licensing Windows are very scary. The 'experts' answering here in /. do seem to understand that. Still many others in the legal world seem to miss the point entirely.
For example, in an Infoworld article published on the nineth, one such expert by the name of Horvath was quoted as wanting to force the licensing of Windows so that different companys could create competing versions. He claims the following result:
When I read something like the above I immediately realize the person making these statements doesn't understand how things really work at all. The actual result would be a fracturing of the API's (Application Programmer Interfaces) where non-compatible versions of Windows would proliferate accross the landscape. This would be followed by a shakeout as one version becomes dominant and gains the largest amount of marketshare.
This would happen because the software industry tends towards standards, and the company that owns the standards (or seems to) owns the marketplace. So, if you are a windows licensee that wants to own the market, the first thing you do is something Microsoft calls 'Embrace and Extend'. This is where you create a version of the operating system that is compatible with the standard, but which has unique enhancements only your version supplies. You do this because you know your competitors are also doing it. This has always happened, it will always happen...
Usually the winner at this game is the one most willing to court the software developers (people like me). Because it is our work (applications) that makes an operating system something people want. Because we are the ones who recommend the Client and Server platforms our companies use. Because without us there would be no software industry. And, most importantly, because we tend to move as a herd towards the one platform that guarentees us the most potential users.
I remember the bad old days of the early 1980's very well. At least 15 competing PC platforms (some little more than toys) and no standards at all. The reason the IBM PC platform running DOS and then Windows became the primary developement platforms is not because they were the best. It was because they were good enough, because they were correctly marketed to programmers and because there was a network effect that brought in more developers as the number of users grew.
My point is this: The software industry will always tend towards a single OS with the vendor of that OS owning an effective monopoly. However this doesn't mean the vendor of that OS can rest on their laurels, as the other constant of the software industry is change. The reason Microsoft was dominant for so long is simple; they understood these facts.
I do not think you can 'end' the Microsoft monopoly by breaking up the company or doing anything that allows the API's to remain closed. However, as was pointed out by several of the /. legal experts above, opening the source is an unlikly remedy because of the licensing issues it raises and the fact it would represent a 'taking' by the government.
So, how about something completly different? How about finding a way to get Microsoft to fund the development of competing Open Source operating systems? I am not certain what form this would take, but I do think it is the only real answer here other than goverment oversight of all Microsoft contracts. Everything else we might do will, because of the natural forces of the software market, tend to the same situation we have now. Personally I would rather the natural monopoly was in the hands of an Open Source group of one kind or another...
Jack
- -
Are you an SF Fan? Are you a Tru-Fan?
After reading this article, I definitely think the way to stop Microsoft and future Microsofts is to open up the API.
An open freely available universal API would give the consumer real choice in which operating system to use. Every OS could run any application. A consumer would choose their OS and then be able to choose their applications independently. I would no longer have to use win32 just because the applications I need to do my work only run on win32.
As most of the Win32 API can be found just by installing the Win32 SDK, the real issues here are change control and ownership. An independent board should carefully control all changes made to the API. Benefits and consequences should be weighed and voted on before any change is published. The API should be owned by the public and not by any company. Even the company/agency that controls change should not be the owner.
I believe that there should be an independent computer API company/agency created. Microsoft probably should fund it as part of their fine from the anti-trust trial. The new API could be based on Win32 (God forbid :) or it could be entirely new. A system like what NT is using now, internal API with Win32/Posix translation layer, could be used as an implementation model. The real problem with an entirely new API will be transitioning exiting Win32 applications and development houses. Another part of the MS equation could be a translation layer from Microsoft for Win32 to the Universal API.
Finally, the wine project has been moving toward providing this utopia, but one of my big concerns about wine is that once wine achieves complete compatibility, Microsoft will introduce some new piece and we'll be right back where we started. For example, wine was making very good headway on 16 bit apps just as Windows 95 was released, and although I don't think MS would change their API again (at least not until IA-64) they could introduce a new common control or some other important piece.
The Linux community is the champion of Open (and often free) computer systems. Instead of focusing on how to punish Microsoft, let's use this to create something good.