This article is more than 1 year old

Final MS witness sinks slowly, boringly

The gold-plated MIT economist don't give good rebuttal

MS on Trial Gross generalisations based on insufficient knowledge of the industry were the hallmark of rebuttal evidence by Dean Richard "Schmalensee" Schmalensee. It fell to Michael Lacovara to examine him for Microsoft, an unenviable task. Dean Schmalensee had a ready denial for anything he was asked to deny, and confirmed everything he was asked to confirm. At $800 an hour, for hours and hours, that's pretty easy work. It was not realistic for Schmalensee to say in one sentence that IE has zero incremental cost to consumers, and in the next sentence that Windows 95 "was available for a price less than $63" implying this was a consumer price. He went on to claim that "Windows 98 is available ... for $2 more, on average". Lacovara tried to repair the earlier damage caused by the defence's refusal to define an operating systems market, because on any definition it can immediately be shown that Microsoft is the monopolist. His tactic was to ask Schmalensee to trot out Microsoft's latest definitions of an OS and a platform. The Dean on platforms Schmalensee said: "The term 'platform' is more evocative than the term 'operating system', I must say. A platform is something on which other things rest. And in this industry, a platform is a piece of software that provides functionality - APIs, in the parlance. It can be used by the writers of other software. So an API provides a platform on which - sorry - a platform provides an interface, a set of functionality, a set of APIs, and a set of standards, but, traditionally, a set of APIs that can be used by other programs. It can be used by other software developers. An operating system operates the computer, schedules the process, runs the disk drive, runs the printer, manages the interfaces and so forth. "Operating systems, typically, are platforms. Many platforms are operating systems. But, conceptually, there is a difference, and an important difference." It was particularly interesting that Schmalensee did not specifically list browsing as a characteristic of operating systems. Schmalensee twittered on about how including more features to Windows raised the barrier to competitors, but suddenly his song was stopped by Judge Jackson: "Would any of these features function as a platform?" he asked. Schmalensee waffled about Quarterdeck's DESQview having some multitasking. In fact of course, Quarterdeck released a multitasking windowing environment called DESQ in May 1984 that could run several DOS programs simultaneously: this was later re-released as DESQview. It is also worth noting that Microsoft used its Windows network device driver to sabotage DESQview and prevent it being loaded into higher memory. Of course, Microsoft's own software experienced no such problems, and Microsoft knew it would take a great deal of reverse engineering to unravel this, causing a critical delay between Microsoft's release of a product and Quarterdeck achieving compatibility with the new Microsoft product. The consequence was that Quarterdeck stopped marketing products that rivalled Microsoft's. Therese E Myers, CEO of Quarterdeck, told the WSJ in December 1992: "We found it to be too difficult. The issue was quite a bit that OEMs had to be loyal" to Microsoft. Microsoft produced some confidential documents from the AOL-Netscape merger, and quoted Goldman Sachs' due diligence report as saying that Netscape would have 100 million users by 2002, and that it already had 50 million users. Due diligence is a procedure that is used to protect companies against shareholder suits as a result of an acquisition. Firms like Goldman Sachs are sometimes used to undertake this, in the belief that heavyweight merchant bankers must be right. The reality of course is that they lack industry experience and are more prone to error than well-informed people in the industry. The Dean on design AOL witness David Colburn had made it clear that AOL's plans were not yet firm, so that Goldman Sachs' forecast was not merely unfounded, it was irresponsible as well. Schmalensee claimed that users were not restricted from adding programs to the start menu, in direct contradiction of earlier evidence. Schmalensee also offered his interpretation as to how the plaintiffs believe IE should have been designed: "I think they believe that Microsoft should have developed Internet Explorer, not as a part of the operating system, but as a stand-alone application." This is an interesting attempt at deceiving the court: Microsoft of course used Spyglass Mosaic as the browser in IE1, with very little change. IE was and is being developed as a separate product, and not as part of Windows. The welding togther, as everybody now knows, has been done by Microsoft to drive Netscape's browser from the market. There was a great deal of criticism by Schmalensee of DoJ witness Franklin Fisher's testimony. Although he was marginally better than Schmalensee, in both cases their technical evidence is so flawed that it should not have been admitted. It was completely unconvincing for Schmalensee to go against admissions from Microsoft's witnesses that the distribution of the Netscape browser was readily achievable by downloading, as though this was as good as OEM distribution. A collection of Fisher's writings was offered as a defence exhibit, but were not accepted by the court. The DoJ's David Boies successfully objected that if they were for the purpose of trying to impeach Fisher, they should have been used earlier. It turned out that Fisher was in court at the time, presumably there in an attempt to curtail some of Schmalensee's wilder excesses. Naughty Schmalensee also claimed that Colburn had said that Netscape had not produced a componentised browser, when in fact Colburn had said that a beta version was delivered to AOL in September 1998. Lacovara failed to get accepted 25 pages of how (in the view of Schmalensee's young helpers, for it was clear that he does not get his hands dirty in these matters), Netscape was not foreclosed from distributing its browser through the ISP channel. Boies objected to this being against the principle of rebuttal testimony, and Judge Jackson agreed: he ruled the exhibit inadmissible. Of course Microsoft will have to pay for the work anyway. So far as predation was concerned, Schmalensee said that if successful, it would have the result that competitors were removed as a constraint on behaviour. Since that is what had happened innumerable times (All those DOS utilities, DR-DOS, OS/2, Navigator...) Schmalensee was quite wrong to claim that there was no predation. Later, he said: "The appropriate test [predation] has two parts... the first involves an act that is a deliberate money-loser, as the phrase is there, pricing below costs in the classic instance. And the second component of the test is that that act makes sense - only makes sense - as part of a strategy that (a) will extinguish or has extinguished competition; and that (b) will permit recoupment of the losses after competition has been extinguished." Judge Jackson began to side with the DoJ, noting that a smaller competitor cannot afford to compete at a low (or negative) price indefinitely, and referred to this as "long-term predation". There seemed little doubt he had made up his mind about this. Schmalensee gayly continued lauding the fact that a key Netscape document that Microsoft had obtained during its raid on the AOL-Netscape merger documentation was under seal, thinking that he could attach any innuendo he wished to it. It was too much for the judge, and he promptly unsealed the key information. This showed that Netscape was "willing to spend up to $10 per download (in a high case scenario) via promotions, contests, incentives and advertising to stimulate downloads." This was hardly surprising, and of course arose because Microsoft had decided to price IE negatively, forcing Netscape to do the same. Schmalensee controverted Gates' statement that IE was a "no-revenue product" by claiming it generated revenues "as a part of Windows". But steady on, Schmalensee, where's the revenue for the Apple version, and the stand-alone versions? By the morning of Schmalensee's second day, he must have been feeling a little tired, since he launched into a "suppose we have a market with ten operating systems of comparable size..." but that didn't work because his client was a monopolist, whatever it said. Schmalensee also forgot that he wasn't a computer scientist, or even an engineer, and suggested that "Microsoft combined the functionality of its MS-DOS products with the functionality of its Windows 3.x products, and some new features. It wasn't combining code. It was combining capability. And produced and integrated --an integrated product. "It offered additional benefits to consumers and to ISVs." Lacovara encouraged Schmalensee to witter on about the future, but it had little relevance, since the Complaint concerned what Microsoft had done, although any remedy would need to be effective in the future. Microsoft needed to show that there was no applications barrier to market entry, in order to show there was no monopoly in an antitrust context. Schmalensee claimed there was no barrier, and thereby confirmed his ignorance of the industry. Schmalensee threw in the PalmOS as a rival to Windows CE, but did not mention that the synchronisation programs for transferring data to and from a PC ran under Windows. The defence runs out of steam Lacovara ran out of useful things to ask, so continued by elaborating the defence rather than rebutting earlier evidence. His purpose was to take as much time as possible, in order to put Boies under pressure to conduct his cross-examination quickly, as the judge wanted the testimony finished last week. It didn't help. Dean Schmalensee had not produced a coherent argument that was going to help Microsoft' case. Microsoft, as usual, has only released a few exhibits, mostly impressions of web pages. There are a few things of interest however. A summary of the proposed AOL-Netscape deal to the AOL board on 17 November 1998 gives an interesting private assessment of Microsoft's situation in the enterprise: "[The deal] offers us a 2-year window while Microsoft is handicapped by its weak traffic [on MSN], poor NT scalability, and lack of directory technology." AOL's strategy was to keep users on its own web site for as long as possible, which was no surprise. Another exhibit shows how Microsoft could break even on IE if the price of Windows were increased. So much for Microsoft never ever charging for IE. ® Complete Register Trial coverage

More about

TIP US OFF

Send us news


Other stories you might like