This article is more than 1 year old

MS attorney retreated in face of impenetrable defence

The problem seems to have been that the more he asked, the more mired he got

Edward Felten's evidence has proved to be technically the most important, so far as a core issue of the case is concerned, providing technical detail that was lacking in previous testimony. The DoJ has not been wise in its failure to provide at least a brief overview of the subject areas to be covered by its witnesses, with the consequence that what was disappointingly absent from Professor Farber's evidence (that he had not looked at the Windows 98 source code) was subsequently covered by Felten. The transcript shows that early reports of Felten's cross-examination do not truly reflect the important events in the court room. It is clear that many reports are adopting a tabloid approach of looking for drama rather than significance. This is well-seen in media reports of comments by Judge Jackson, where he admonished Dave Heiner of Microsoft for playing word games with Felten and trying to make him "make a slip of the lip" by repeatedly asking the same questions. But more significant was Felten's control of Heiner, so that Heiner's questions not only failed to gain points, they also tended to land Microsoft deeper in the mire when Felten elaborated. It is a disaster for a cross-examiner constantly to be surprised by the answer to factual questions. Heiner was badly briefed by Microsoft, and may be finding it difficult to get enough time from Microsoft specialists to gain a sufficient understanding of the issues. Just possibly, he is being fed a story line by Microsoft staff that is far from the truth. Alternatively, he may find the subject matter too technical. It may well be that Neukom (or Gates, who is rumouredly pulling the strings behind the scene) decided that Heiner was not doing well and should be curtailed. The significance of Felten's evidence was that he called Microsoft's technical bluff, and won. He drew attention to Microsoft's efforts to sabotage his now-famous prototype removal program, which could be seen as a sign of Microsoft's concern that the program might suddenly find its way onto the Web and be used by dissidents. It was also a crude attempt to discourage other efforts to emulate Felten's approach. It is amusing, but hardly surprising, that Microsoft does not have sufficient confidence about IE to believe that its enslaved Widows 98 users would prefer to stay with IE. At no time did Felten lose his cool, and he gave no ground at all. When asked which software in Windows 98 allows the user to browse the Web, Felten replied: "Are you really asking me to go to through the 14 or 18 million lines of code one by one and tell you which is this and which is that?" Heiner failed to find a toehold on the vertical rock face that he attempted to climb, and fell. What was lacking in Felten's evidence, and was not drawn out in Philip Malone's redirect examination, was a generalised account of the packaging tricks that Microsoft used to weld together Windows and IE. The DoJ has a hard job making a legally sound case that Microsoft should not be allowed to merge Windows and IE. The merger becomes an illegality if it can be shown that the way in which it was deliberately anticompetitive. The files at issue are particularly shdocv.dll, mshtml.dll, urlmon.dll and wininet.dll. Few can doubt, particularly after Felten's evidence, that Microsoft has indulged in some very devious ways of combining different functionality in files, so that relatively few can be identified as being specifically IE files. The DoJ may not have done enough yet to prove that Microsoft combined files in the way it did to make the removal of IE as difficult as possible. The raw material for a better summary of packaging tricks is mostly in Felten's testimony, but it requires a clearer presentation of the principles involved. The particularly childish aspect of Microsoft's approach to Felten's removal program was the mock seriousness with which Microsoft treated it. Microsoft continually turned its deaf ear when Felten said that Microsoft had better resources than him to construct an IE removal program (and of course to produce a version of Windows 98 without it). Instead, Heiner treated Felten's prototype removal program as something that it was not intended to be. Microsoft had evidently examined it very closely since September when Microsoft was first given a copy of the program. Some very silly questions resulted. For example, Heiner claimed that html page rendering was two to 300 per cent slower. Felten called his bluff, and said that he had been using Windows 98 with IE removed and Navigator installed instead every day for many months, and that his experience was inconsistent with Heiner's claim. Heiner also tried to make a big issue about how small the removed parts of Windows 98 were, ignoring many clear statements by Felten that his objective was to demonstrate a point: that IE functionality could be removed from Windows 98; that another browser (including IE) could then be installed; and that Windows functionality was unchanged. Felten said clearly many times that there was no technological reason for Microsoft to package IE with Windows 98, and that Microsoft failed to provide user choice about browsers. ® Complete Register trial coverage

More about

TIP US OFF

Send us news


Other stories you might like