Damning MS with the facts III – the browser battle

The third part of our analysis of the DoJ's case

MS on Trial The DoJ's and plaintiff States' Joint Proposed Findings of Fact is the first time the DoJ has been able to set out logically its evidence concerning Microsoft's attempt to split the browser market. It would be wrong to portray the issue as being Netscape versus Microsoft because the focus is on whether Microsoft has used illegal anti-competitive practices to the detriment of consumers, and not whether Netscape has suffered. That's a primary reason why Microsoft's attack on the AOL-Netscape-Sun deal and the subsequent fishing trip through the associated documentation was a red herring, used by Microsoft to find out business secrets, rather than because it truly needed to see the documents to defend itself. The DoJ did not go into battle on behalf of Netscape, but rather to defend the principle of fair competition (and the resulting benefits this brings to consumers), as embodied in US antitrust law and particularly the Sherman Act, sections 1 and 2. To be sure, there are many other issues where the DoJ is challenging Microsoft, but the primary issue remains the alleged illegal monopolisation. Microsoft's primary justification for its action of incorporating IE into Windows 98 was that it had always "integrated" technology into MS-DOS, that this had been planned before Netscape existed, and the Netscape knew about the plan. The difference between incorporating a browser and incorporating memory management, compression and the like is that this time there was the possibility that Microsoft's operating system monopoly would be successfully challenged. As a consequence, Microsoft drew out all it claws and attacked everything in sight. Gates recognised that non-MS browsers threatened to commoditise Windows. Gates' May 1995 "Internet Tidal Wave" memo is not particularly perceptive, except that he identified Netscape as " a new competitor 'born' on the Internet". Windows general manager Ben Slivka's report "The Web is the next platform" was more to the point: "We should be extending the Web with as many Microsoft technologies as possible, even if we have to modify those technologies in ways not original [sic] intended by their designers". The DoJ lists seven further examples where Microsoft recognised the threat posed by browsers to its monopoly. Although the incident in which Netscape Chairman Jim Clark, in "a moment of weakness" in late December 1994, tried to re-open negotiations about Microsoft paying Netscape a flat fee "of a couple of million dollars to take us out of the game" is known, the circumstances at the time have not been explained, although we do know that Clark made the move without consulting the Netscape board. Gates had announced Marvel (MSN) in November 1994 and there was much speculation that Microsoft would be able to dominate what was then the online services business because of the power it was expected to have as a result of having the MSN icon (exclusively, it was then thought) on the desktop. It was very likely that Clark was intimidated by the market strength that this was expected to bring MSN. In the event of course, Microsoft did not succeed with MSN as an online service. Clark deal - why didn't MS bite? What remains unexplained is why Microsoft rebuffed Clark's attempt at negotiations. The DoJ merely says that "the discussions [with Clark] went no further and were not renewed". At the time, Microsoft was negotiating to license Mosaic, and agreed this two weeks later. Logically, even if negotiations for Mosaic were well advanced, and Microsoft had started building a team to turn Mosaic into IE (and it is known that very little work was done on Mosaic to create IE 1.0), the prize of a neutralised Netscape would have far exceeded the alternative. Possibly some Microsoft egos were at stake, but if the opportunity were really there, Microsoft's failure to move seems quite illogical. The first suggestion from Microsoft that Netscape should not compete was at a meeting on 2 June. The DoJ says "Microsoft suggested that Netscape consider abandoning its independent base of platform-level browsing code on Windows 95 and merely build on top of Microsoft's code". Barksdale, Netscape's CEO, subsequently visited Microsoft and had discussions with Dan Rosen, Paul Maritz and Nathan Myhrvold about the possibility of Microsoft bundling Navigator. Reardon, in his deposition, salivated at the "lots of value-added stuff beyond HTML rendering in Navigator". Barksdale had evidently not appreciated that Microsoft wanted Netscape to use Microsoft's browser code and just build applications on top of it. Barksdale even suggested that Microsoft "distribute [Navigator]". There is further evidence in Microsoft's internal documents after the 2 June meeting that Microsoft's strategic plan was to convince Netscape to abandon Navigator. Rosen emailed Maritz optimistically, expecting Netscape to concede, but Maritz did not agree that "Netscape is 'ready for a broad strategic relationship' ". On 21 June, at a meeting between Microsoft and Netscape, Microsoft "proposed to Netscape that the two firms divide the browser market" with Netscape ceasing to develop a browser for Window 95 and Microsoft not producing IE for Unix or the Mac. It was at this meeting that the idea of a "line" was developed. Although Microsoft has made a crafty case for questioning Andreessen's actions concerning his notes for the meeting on 21 June 1995 in which browser market splitting was discussed, the DoJ has provided a full account that counters Microsoft's innuendo. The DoJ has put all the documentary evidence into chronological sequence. Daniel Rosen, Microsoft's lead negotiator, wrote an internal Microsoft email to Maritz and others on 14 April 1995 asking "Will they [Netscape] cede the client [Navigator] and its standards to us?" Co-opting Netscape On 15 May, Rosen noted in another email to Maritz that "our goal should be to wrest leadership of the client evolution from [Netscape]". Thomas Reardon of Microsoft was even more direct in an email on 25 May to Rosen that an exchange of protocol specs [for browsers] "would really be a veiled effort on our part to move [Netscape] off the Windows client". Maritz replied in an email to Gates and Rosen that it was "imperative ... to co-opt Netscape" and that he was open to any strategy to achieve that result. The deposition of Chris Jones, an IE product manager who played a key role in the 21 June meeting for Microsoft, confirmed that "Oh, I believe there was a discussion of that nature [market splitting]". This essentially scuppered Microsoft's claim that no such discussion occurred. It is important to distinguish between perfectly legal persuasive efforts by Microsoft to ask Netscape to drop Navigator, and an illegal proposal of market splitting, which goes against the user interest. This is one of the most important things to emerge from the evidence, and would make it hard for an appellate court to overturn Judge Jackson if he rules for the DoJ. The judge will also have noted that Microsoft changed its story. In its Proposed Findings of Fact, Microsoft did its best to undermine Andreessen's credibility, but the clinchers are that he is known to be a very fast typist (and talker, come to that); that his notes were distributed quickly after the meeting (making it rather difficult to have concocted them); that a copy was sent to Apple the next day; and that all the other evidence from the time supports this. Only Microsoft's subsequent testimony attempts to reinterpret what happened. Barksdale was convinced that Microsoft's purpose was "to get access to certain code and APIs necessary for ... product development". Rosen gave his account of the meeting in a draft email the same day, noting that "Much of the conversation centred on how the lines would be drawn" between the respective roles of Microsoft's and Netscape's roles in a divided market. But it was Jones' email comments that confirmed what had been put to Netscape and was damaging to Microsoft since his story closely confirmed the accounts by Barksdale and Andreessen. He wrote: "Because of our priority to own client and server platform, if they can agree to use our client code on Win95 and use our BackOffice and NT APIs, and promote these as the solutions ..." then Netscape would be neutralised. Rosen emailed Gates saying "the test of this alignment will be Netscape's agreement to use Microsoft's client code on Win95" and ending that Netscape "seemed to embrace this strategy". Reardon, Maritz and Gates were sceptical that Netscape would agree to the proposal, but the email chain shows without any doubt that market splitting was proposed by Microsoft. When Microsoft realised the antitrust problems that had resulted, in the DoJ's words, it made an "after-the-fact assertion that its market division proposal was simply exploring forms of legitimate cooperation" that was a pretext and "contrary to the evidence". Microsoft's contention that "it was not trying to get Netscape out of the browser business is erroneous" and that "even on Microsoft's version of the facts, its proposal to Netscape was a naked attempt to eliminate platform-level competition and to divide markets". Rosen's testimony was "inconsistent", the DoJ added euphemistically, and proceeded to devote nine pages to a detailed examination of the "inconsistencies". Nor does Microsoft's story that it was engaging in joint venture discussions correlate with the evidence, the DoJ claimed with 13 pages of detailed document and testimony citation. Has Netscape agreed to Microsoft's market division proposal, the DoJ noted, Microsoft's OS monopoly would have been maintained and a Microsoft monopoly in the browser market would have resulted. Of course, since the evidence was concluded, Microsoft has achieved a browser monopoly in new PC browsers since a monopoly is usually defined as having 70 per cent of the market. Developments have moved faster than the courts, with the result that IE is now dominant, Netscape's circumstances have changed radically, and there is no easy way in which the so-called IE-Windows 98 integration can be undone without harming consumers. With the prospect of an appeal before there is any final resolution of the matter, it could just be that Microsoft's own effort to evolve Windows 98 into the NT code base will flounder before any final legal conclusion is reached. At the same time, some measures by the court to curtail Microsoft's expansion by acquisition, some enforced divestment, some controls of licensing and pricing, and some further erosion of the hardcore executive staff (and particularly its chairman) could finally bring about a gentler, kinder Microsoft. ® Other Sections Part II Part I DoJ damns MS with facts Complete Register Trial coverage

Other stories you might like

  • Hive ransomware gang rapidly evolves with complex encryption, Rust code
    RaaS malware devs have been busy bees

    The Hive group, which has become one of the most prolific ransomware-as-a-service (RaaS) operators, has significantly overhauled its malware, including migrating the code to the Rust programming language and using a more complex file encryption process.

    Researchers at the Microsoft Threat Intelligence Center (MSTIC) uncovered the Hive variant while analyzing a change in the group's methods.

    "With its latest variant carrying several major upgrades, Hive also proves it's one of the fastest evolving ransomware families, exemplifying the continuously changing ransomware ecosystem," the researchers said in a write-up this week.

    Continue reading
  • What do you mean your exaflop is better than mine?
    Gaming the system was fine for a while, now it's time to get precise about precision

    Comment A multi-exaflop supercomputer the size of your mini-fridge? Sure, but read the fine print and you may discover those performance figures have been a bit … stretched.

    As more chipmakers bake support for 8-bit floating point (FP8) math into next-gen silicon, we can expect an era of increasingly wild AI performance claims that differ dramatically from the standard way of measuring large system performance, using double-precision 64-bit floating point or FP64.

    When vendors shout about exascale performance, be aware that some will use FP8 and some FP64, and it's important to know which is being used as a metric. A computer system that can achieve (say) 200 peta-FLOPS of FP64 is a much more powerful beast than a system capable of 200 peta-FLOPS at just FP8.

    Continue reading
  • Meta's AI translation breaks 200 language barrier
    Open source model improves translation of rarer spoken languages by 70%

    Meta's quest to translate underserved languages is marking its first victory with the open source release of a language model able to decipher 202 languages.

    Named after Meta's No Language Left Behind initiative and dubbed NLLB-200, the model is the first able to translate so many languages, according to its makers, all with the goal to improve translation for languages overlooked by similar projects. 

    "The vast majority of improvements made in machine translation in the last decades have been for high-resource languages," Meta researchers wrote in a paper [PDF]. "While machine translation continues to grow, the fruits it bears are unevenly distributed," they said. 

    Continue reading

Biting the hand that feeds IT © 1998–2022