By William Fellows
As expected, Sun Microsystems Inc will use ECMA as its route to the standardization of Java at ISO. In this way Sun hopes to be able to retain control over the future development of Java which it would have had to give up under new rules adopted by ISO’s JTC1 committee and applied to the PAS process that Sun has now abandoned. ECMA standards can be fast-tracked to ISO standardization through a relationship ECMA has with ISO which Sun’s VP Jim Mitchell described as unique.
Whatever that means the fact is more than 100 ECMA standards have made it through to ISO standardization this way. Mitchell also claims there are no maintenance issues involved in this process and its ECMA submission leaves control of the specification in the hands of Sun’s Java Community Process, the vehicle for extending the Java platform specifications. ECMA’s only maintenance will be of a passive kind, fixing the wording of the spec itself. Sun’s already submitted a Java 1.2.2 specification to ECMA which will be presented to a meeting of the group’s general assembly in Kyoto, Japan on June 24. ECMA is expected to vote on the spec in December. Then it goes to ISO for fast track adoption.
However it’s not all plain sailing for Sun. There will be a straight yes/no vote at ISO among the national bodies represented. In 1996 Microsoft Corp was able to shoot down another ECMA standard, the Public Windows Inititaive, at this stage, thus preventing it from becoming an ISO standard. The PWI was a Sun effort to get Windows APIs put into the public domain. Sun argues that this time around there is no IP issue. Microsoft was able to mount a successful campaign against PWI at ISO on this issue. Java software president Alan Baratz said the company will be satisfied even if Java gets no further than ECMA standardization; at least it’s a standard of some kind. Sun has gone so far as to hire the former DEC exec who was responsible for helping PWI get shot down.
There’s also the issue of Microsoft, Hewlett-Packard and others – mostly small ISVs – called the Real Time Java Working Group who are trying to get their work on a real-time Java clone taken up by a standards body. Sun has been able to head them off so far, most recently at a vote by NCITS, the National Committee for Information Technology Standardization, not to sponsor the group’s work. However it has been clear for some time that efforts to come up with something more acceptable to NCITS are proceeding and Sun’s latest move is sure to galvanize a more concerted effort by HP and Microsoft to get control of Java. Baratz said at most their efforts will amount to a last desperate attempt to revive Chai [HP’s embedded Java clone which it has licensed to HP] and derail Java standardization. Baratz says their efforts are aimed at fragmenting Java and thereby undermining its cross-platform value proposition. He says they won’t succeed, pointing to the failure of previous Microsoft efforts to standardize its own technologies, including ActiveX.
Standards veteran Jim Isaak believes HP, Microsoft et al will take advantage of the change in Sun’s standards plan by gathering support for some kind of a Java consortium which will nail down precisely what they want to achieve. That might include establishing a well formed Java specification they could promote as a standard. They could take that standard to another standards body, or even apply for PAS status to submit it to ISO. It’s perfectly acceptable, Isaak notes, to take other people’s work forward for standardization. It was a precedent created by none other than Sun itself when it took Microsoft APIs forward to become an ECMA standard as PWI. Isaak says the so-called Real Time Java Working Group chose badly when it opted for NCITS as its standards route. He advised against it, but notes that perhaps IEEE might be a more amenable to its goals. But only if it protects Java’s crucial element: platform independence. Isaak says that Sun appears to care more about maintaining control of Java development than it does about standardizing the technology. Isaak says that when he worked at IEEE he tried to get a Java standard initiative going when the technology was in its very infancy, but that Sun didn’t want to play.
Meantime, he wonders whether ECMA might not be endangering its own credibility as a standards development organization by sponsoring Java. ECMA is a category A SDO, meaning it can submit its own standards directly to ISO’s fast track process. He points out that ISO’s JTC1 group does not actually require control over maintenance. It requires control to be provided through a ‘due process’ that allows any interested parties a voice in the process. Traditionally ECMA was also an organization with a ‘due process’ that provided a means for any affected interest to participate in the maintenance of their standards. Is ECMA is willing to forgo some of the due process considerations in returns for getting a ‘hot item’ like Java? It’s a question for ECMA members.
Content from our partners
Isaak also speculates that Sun’s new Java standardization plan could also prove to be self destructive. He says US guidelines (OMB A119, Public Law 104-113, and prior FTC guidelines) all define common elements of ‘due process’ that are required for US Government adoption of Volunteer Consensus Standards. A document endorsed by ECMA – such as the Java spec – that fails these criteria will be challenged in any such use. In this way ECMA lose credibility as standards setter. Moreover, if Sun succeeds, they set the stage for any number of corporations to promote their specifications as standards though similar means, and this opens the doors for more powerful players than Sun … quite likely to Sun’s detriment.
