To: John F. Dowd who wrote (35196 ) 12/9/1999 11:57:00 AM From: Daniel Schuh Respond to of 74651
Sun, Microsoft and the standards wars theregister.co.uk Then again, if you look back a little, there's always plenty of laughs to spread around. Microsoft, funnily enough, hasn't come under anything like as much fire vis a vis standards as Sun has, despite having done everything in its power to keep Windows out of the standardisation business. But the arguments go back a long way. To get the whole story straight, we have to go back to 1992, when Apple had announced its Open Collaboration Environment. Microsoft hastily responded with the concept of a Windows Open Services Architecture (WOSA), which was described at the time by Forrester as "nothing more than a backgrounder put out by Microsoft's PR department", but it did result in "an awful lot of saluting and much less shipping" as one wag put it at the time. Novell responded with a white paper on Systems Applications Services, but it was Sun's proposal for WABI, a Windows Application Binary Interface that caught the imagination, since it provided a way to give Unix users the ability to run Windows applications in an open environment, without the need for Dos or Windows, and without emulation or any change to the Windows application. Shock horror in the Microsoft camp, with Redmond threatening litigation, claiming that Windows intellectual property rights were being violated. Apart from Sun's traditional antithesis to Microsoft, Sun advanced the WABI approach because Microsoft would not offer it any acceptable terms for licensing Windows source code, although it had licensed Insignia, Bristol and Locus. Microsoft was upset when identical 486s running the same benchmark, one with WABI and the other with Windows, showed that WABI out-performed Windows by 50 per cent. If Microsoft needed a prod to move it in the direction of 32-bit APIs, this was it. How unusual. DR-DOS worked better than MS-DOS, according to the dread Microsoft email anyway, and WABI ran Windows programs faster than Windows. So of course, Microsoft had to compete by other means. Microsoft was further embarrassed when in 1994, Sun, Novell, Microsoft and the European Commission met at Gleneagles in Scotland ostensibly to discuss the future direction of the desktop environment, but Sun made the event the launch of the Public Windows Interface (PWI) that was to be submitted to standards bodies. Nigel Burton, then Microsoft's manager of Microsoft's solutions development group admitted that Microsoft "should have done a better job" of documenting Windows, but "the idea of relinquishing ownership of the Windows API" was not open to discussion. Sauce for the goose is apparently not sauce for the gander in this case, but that's life. WOSA was not open at all, it transpired. In 1994, the PWI proposal was submitted to ECMA. As can be easily imagined, Microsoft had some strong reasons to get its revenge on Sun, which brings us back to the recent events. Sun did not deliver the Java 2 Standard Edition to ECMA on 1 December because of its concern that although ECMA protected patents in submissions, it did not provide protection for copyright or other intellectual property rights. Sun did not wish to encourage Microsoft's amorous embrace-and-extend moves, so decided to withdraw. It is still possible, but unlikely, for ECMA to continue Java standardisation. Sun remains a member of ECMA, and has been very active, proposing not just the API for Windows, but also ECMA Script and ECMA Object Data Interfaces.