Liberate OEMs, open Windows APIs, says MS critic Reback

MS Remedies The lawyers speaking at Ralph Nader's Which Remedies? meeting had a better grasp of the situation than the economists, but those from a traditional antitrust law background did not seem to appreciate just haw obdurate a company Microsoft would be in compliance, or even if it were split. Those like Gary Reback who had crossed swords with Microsoft in court were in a better position to appreciate what could work. Glenn Manishin of Blumenfeld and Cohen (and author of the SIIA remedies document) explained why he considered structural relief preferable to conduct remedies. The latter left serious risks for enforcement, as well as the interpretation of the scope and definition of any consent decree [as has been seen already]. There is inconsistency in the use of terms like horizontal and vertical splitting of Microsoft and its products. Horizontal splitting is mostly used to describe remedies where initial line drawing is required to split operating systems from applications and content. It has already been seen how Microsoft has been able to play with the boundary between the OS and application with the "integration" of IE into Windows. One of the undesirable results of such horizontal splitting would be to leave Microsoft with OS monopoly pricing power, and the potential to return to its old ways. Vertical splitting seems generally to mean dividing Microsoft into Baby Bills. Manishin saw this as a clean mechanism for eliminating anticompetitive incentives, and avoiding intrusive government oversight. With so-called vertical divestiture, there is no need for judicial line drawing or product definition. Manishin thought that the risk of OS fragmentation was largely illusory and would be offset by new compatibility-enhancing products. He was not concerned at the complexity of the employee and stock-splitting issues. If there was one thing that united all present at the meeting, it was that the government should not be involved in technical line drawing. Microsoft has deliberately mischaracterised this as the government designing software, of course. It seems that in the ongoing secret discussions between Microsoft and the DoJ (which cannot really be regarded as negotiations, in view of Microsoft's intransigent attitude), Microsoft is only willing to concede some minor conduct remedies which would not eliminate any of the seriously anticompetitive practices. There was also much cynicism as to whether the court could prove to be effective in the detailed supervision of Microsoft's conduct, so that conduct remedies were not generally favoured by those present. Essential facility arguments were mentioned, but not examined in detail. Manishin said that reasonable and un- discriminatory access to the essential facility was needed, but someone has to determine what is reasonable. Gary Reback of Wilson, Sonsini, Goodrich & Rosati has proved himself to be a formidable legal opponent of Microsoft on several previous occasions. He noted that there was concern amongst Microsoft acolytes that Microsoft had done such a poor job of presenting its case. Reback's response was: "That's the truth: that what you see in court is the true Microsoft ... There's no surprise about that." Reback thought that the best remedy would be one that did not require judicial monitoring. It was widely admitted that the DoJ did not do a good job of monitoring, since it is not a regulatory agency, but is a law enforcement group. Consequently, monitoring consent decrees is not high on its agenda. The business conduct remedy ("thou shalt not ...") would bring all manner of problems, Reback considered. "Thou shalt not bundle" would not work, he said, noting that the same week Microsoft claimed to be having settlement talks with the DoJ, it announced it would bundle its media player with Windows. If the remedy does not address such issues, what is the point of the trial, he asked. Source code licensing would create problems, since if compatible versions of Windows were produced, they would be playing catch-up all the time, and wouldn't do very well because nobody is prepared to build a business around them. Nor would MIS directors be willing to stake their companies on them. This argument by Reback is of course the answer to Lebowitz' theoretical excursion into the economics of more versions of Windows. In theoretical mode, Reback said that so far as the cost of competition was concerned, if it were true that the additional costs of breaking up Microsoft might result in 6 per cent higher costs, as Lebowitz had suggested, it should also be born in mind that there are enormous costs in trying to keep up with Microsoft's products. Reback thought it worth 6 per cent to have competition. In addition, the cost of maintaining Windows 9x and NT compatibility was high for users. Game, set, and match to Reback. The cost of the Microsoft monopoly was very great: trying to keep up with Microsoft's products, with Microsoft adding and then subtracting APIs during development. Reback suggested that the goals should be: to liberate the OEM channel (freedom to bundle by the OEM); unbundling relief (to stop what happened to Netscape from happening again); access to information (so that Microsoft cannot prevent others from producing software to run on Windows, or benefit from undisclosed calls for example); proscription on acquisitions, for a period of time (to prevent Microsoft buying, for example, the number four player, and putting numbers one to three out of business). Microsoft had made it very difficult for hardware manufacturers to make new equipment work with Windows, since the manufacturer is beholden to Microsoft for an API. Microsoft should not have the right to decide what is developed - basis of capitalist model: that's what capitalism is. Joe Simons of law firm Rogers & Wells noted that bad remedies can be worse than not bringing a case at all, since the miscreant sees it has got away with it, and becomes even more anti-competitive. He noted that Microsoft is apparently willing to agree to a consent decree containing some conduct issues, but was not willing to give up its power to "innovate" [which seems to be synonymous with "monopolise"], nor would it accept any break-up of the company. It was appropriate to talk now about the remedies needed, since there could be a consent decree at any time. Judge Sporkin received hardly a mention at the meeting. [Sporkin had opposed the first Microsoft consent decree, since in his view it was too narrow, but he was overruled by the Court of Appeals. Should there be a consent decree again, it is likely that there would be many responses during the Tunney Act period when there is an opportunity for public comment. The Court of Appeals in DC is a very political court that has shown itself to be opposed to antitrust law. It ignored the provisions of the Tunney Act that were designed to stop cosy deals between the DoJ and a miscreant when it overruled Judge Sporkin on spurious grounds. However, this time around, if there is a consent decree, it is less likely that the court could get away with the same trick, without there being blood on the streets of Washington.] Simons noted that most large companies have antitrust compliance programmes, but it had emerged during the trial that Microsoft had no such programme. The problem that Microsoft presented was that there was not just a disagreement about the law: Microsoft would not accept the court's result, as had been previously evidenced when Gates said that the consent decree would change nothing at Microsoft. Microsoft had flagrantly (and contemptuously) provided copies of Windows without IE that would not work, when it had been shown that it was possible to remove IE functionality from the so-called integrated product. ®