|
|
|
|
Sun's Java is, after all, the foundation of most of the Web services solutions offered by the WS-I's inaugural board members. Even casual observers must have been wondering how an important interoperability organization could exist without Sun being afforded status equal to Microsoft and Java 2 Enterprise Edition licensees IBM, Oracle, BEA, and Hewlett-Packard. Two months later, and still no Sun. What happened? Did Sun's invitation get lost in the mail? Or was a decision made to exclude Sun from the board? And could such a decision be part of a larger plot to marginalize Sun's Web services stature? This much is known: Sun was invited. But the nature and the circumstances of that invitation make it clear that someone or some group decided not to include Sun as a board member. The idea for something like the WS-I had been floating around for some time. During the fall of 2001, the idea was being discussed at business meals and industry events. By January, IBM and Microsoft officially hatched the WS-I and were lining up its board members. Oracle's invitation to be a WS-I board member arrived sometime in early January, according to senior director of Oracle9i product marketing, John Magee. Meanwhile, Sun was kept in the dark. It would n't be until February 4, just two days before the public announcement of the WS-I, that Sun vice presidents Marge Breya (SunONE) and Richard Green (Java) were informed of the WS-I and invited to join with a less-than-board member "contributor" role by IBM's Director for eBusiness Standards Strategy Bob Sutor. Many other companies were offered membership during the same two-day period, according to both Sutor and his Microsoft counterpart--.Net Platforms Strategy group director Neil Charney. They have insisted that Sun wasn't singled out for a last minute invitation. But that doesn't address the question of why Sun, perceived by many as a leader in Web services, wasn't invited to join earlier as a board member.About a week later, at the launch of Microsoft's Visual Studio .Net, Bill Gates extolled Microsoft's membership in the WS-I while deriding Sun's lack of participation. What Gates neglected to point out was that Sun hardly had any time to evaluate the proposal and was at that point considering membership. Both Microsoft's Charney and IBM's Sutor deny any intent to marginalize Sun. Both insist that they sought members for the WS-I's board who, historically, have demonstrated leadership in Web services. Reflecting on history, Charney said, "If you asked Sun, you wouldn't necessarily get a positive answer about Web services." Sutor concurred, adding that "When it comes to the XML-based standards that most people consider to be a part of Web services, Sun does not pop to the top of that list of people." While Sutor concurs with Charney, history may concur with neither. Sun is virtually synonymous with one of those XML-based standards: XML itself. XML is the oldest of the four core Web services protocols and the only one that wasn't hatched recently from an IBM or Microsoft lab. XML has been under the stewardship of both the World Wide Web Consortium (W3C) and the Organization for the Advancement of Structured Information Standards (OASIS). Sun appears to have been deeply involved in XML's evolution as a protocol long before Web services came into vogue. From 1998-2000, Sun's Director of Technology Development Bill Smith was president of OASIS' board of directors. Sun's current manager of XML industry initiatives, Simon Nicholson, is on that board. According to documents on the W3C's Web page for Web services activity, Sun was active in the earliest of the W3C's birds-of-a-feather sessions for the XML protocol. The site also lists a white paper by Sun's Jon Bosak as one of the first significant drafts on XML's promise for the Web. When pressed for specifics on why they considered Sun's behavior unbecoming of a Web services leader, Sutor and Charney cited Sun's early positions on the three other core Web services protocols (UDDI, SOAP, WSDL). But they made no mention of XML. Sutor claimed that Sun dismissed UDDI's usefulness because the company viewed the Web services directory as a threat to Jini, its own Java-based technology for locating services on a network. But the record suggests otherwise. In a press release from Sept. 6, 2000, IBM, Microsoft, and Ariba not only announced the launch of the UDDI Project, but listed Sun as one of the "thirty-six industry leaders" that supported it. Sutor offered no other specifics as to why he felt Sun wasn't behind UDDI. He did, however, provide details on the WSDL protocol. Sun, he said, failed to get behind the submission of certain Web services protocols to W3C. Last year, according to Sutor, he asked Sun's Smith to support the WSDL submission and Smith refused. If Sutor did make the request, it wasn't a casual call to an acquaintance. Sutor and Smith are credited with working famously together under the guise of OASIS to produce the ebXML specification for the United Nations. While Smith was unavailable for comment, the only evidence on public record that might corroborate Sutor's contentions are the initial submissions to the W3C for both SOAP and WSDL. Unlike the first public document of UDDI.org, neither lists Sun as one of the many initial backers of the two protocols. On the other hand, Sun went public shortly thereafter about its enthusiasm for IBM's support of SOAP. Within two months of the W3C SOAP submission, IBM donated a Java-based implementation of the specification called SOAP4J to the open source community. IBM's June 1, 2000 press release, announcing the initiative, quotes Sutor as saying that "IBM is a leading advocate of ensuring that the Internet foundation remains truly open and that the Internet doesn't become a battleground of competing, vendor-specific 'control points.'" Days later, during a presentation at the JavaOne conference in which Sun's XML roadmap was presented (see Webcast), Sun's Duncan Davidson called SOAP "a very interesting piece of technology to a lot of people." Noting IBM's efforts to make SOAP "a truly standard solution that's not controlled by any one vendor," Duncan added, "We really support them in doing so and we're putting engineering resources to help make that happen." There's virtually no evidence to suggest that Sun had failed to embrace at least three of the four Web services protocols--XML, UDDI, and SOAP. If the lists of initial backers are the proof points, the only controversy concerns the remaining protocol--WSDL. As it turns out, the list of the nine WS-I board members--Accenture, BEA, Fujitsu, Hewlett-Packard, IBM, Intel, Oracle, and SAP, reads like a who's who of those that initially backed the W3C WSDL submission. All except for Accenture. If Sutor and Charney were using the list of UDDI, WSDL, and SOAP backers as a benchmark for Web services leadership and therefore invitations to be founding members of the WS-I, that doesn't explain why Accenture--one of the nine WS-I board members--rose to the top and Sun didn't. Like Sun, Accenture (then Andersen Consulting) was one of the original 36 backers of the UDDI Project. Also like Sun, it doesn't appear as an initial backer of WSDL or SOAP. Likewise, four of the six remaining companies that received invitations to be board members--HP, Intel, and Oracle--also appear on only one of the three lists: The initial list of WSDL backers. In fact, if initial support for UDDI, WSDL, and SOAP were the criteria used to decide who would be WS-I board members and who wouldn't, then Ariba and CommerceOne are mysteriously absent as board members. In addition to IBM, Microsoft, and SAP--three of the nine WS-I board members--Ariba and CommerceOne are the only companies that appear on all three lists. Neither received an invitation to be a WS-I board member. While Sun executives won't speculate on the criteria used to select WS-I board members, they're quick to point out that Sun is extremely active in all aspects of Web services. In addition to Sun's demonstrated involvement in XML and UDDI, Sun's chief technology evangelist Simon Phipps said that Sun has also been in the thick of SOAP and WSDL. According to Phipps, "The record speaks for itself. Sun was present the first day it was publicly possible for us to join the W3C's working groups in support of the SOAP and WSDL protocols." While the minutes of those meetings are tucked away behind a members-only section of the W3C's Web site, another page demonstrating Sun's involvement in Web services isn't. According to the W3C's Web Services Architecture Working Group roster, a Sun employee--Chris Ferris--is as close to the top of the list as a participant can get: He's the chairperson. Microsoft's Charney cautioned against relying on single litmus test such as who was on what list. For example, Charney criticized Sun's absence from one of the industry's first Web services interoperability events as an indicator of Sun's unwillingness to work with the rest of the industry. Citing conflicting commitments, Sun officials acknowledged that they missed the event but said the company will be present at all future interoperability bake-offs. According to Charney, "We looked for people seeking alignment around Web services and we wanted to keep it down to smaller group of companies that had unique perspectives and were complementary to each other. Accenture, for example, already had significant efforts underway to help its customers deploy Web services." What do you think so far? Is the WS-I a power play? Should Sun have a seat on the board, or should it join as a contributor? Or, do you have that elusive piece to complete the puzzle? Let David know what's on your mind by writing to him at david.berlind@cnet.com. Read "IBM's unfolding power play," where David points out that there could there be more to the WS-I's history than meets the eye. |
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||
|