Not Quite, Not Yet

It doesn't take an expert to tell you when a JVM doesn't support Java version 1.1. All you have to do is point it at some 1.1 applets and wait for the quiet explosion. That's exactly, of course, the experience Netscape prepared for us last week with the release of Communicator 4.0.

This week, with Sun's HotJava remaining the world's only "1.1 capable" Web browser, Netscape has yet to respond to our inquiries about the matter. Symantec has also passed the buck, claiming that its Just-In-Time compiler-based JVM, as supplied to Netscape, is Java 1.1 compliant, but that Netscape itself has failed to supply the appropriate supporting classes to make it work with 1.1 applets.

That's a plausible but disappointing explanation, and it leaves the developer-at-large waiting for an amount of time very loosely specified (it's now due "sometime this fall"). While we hate to be tough on Netscape about it, the company did claim it would be here by now.

Symantec, incidentally, was proud to inform us that its very same JVM had been licensed by Sun for inclusion into the official Sun JDK 1.1. Interesting.

In the Meantime...

Quite undeterred, the rest of the Java industry marches on.

Sun, interestingly enough, has managed to run afoul of one of the most prominent standard-setting organizations in the world. Due to the flurry of acronyms and hype about the decision, it's been hard to make sense of the ramifications, but in fact we consider the consequences to be rather negligible.

Economic interests have a way of rendering standards bodies unimportant, and the Web itself is perhaps the best indication of that.

However, for the curious: the U.S. group responsible for American positions on international standards failed to stamp approval on Sun's Java proposal to the ISO/IEC--a very venerable international standards body that has no doubt affected our lives in countless ways we don't even realize.

The issues surrounding the rejection basically relate to Sun's desire to continue to own Java and control its development--quite an ordinary thing for Sun to want to do, and quite an unusual state of affairs for the standards bodies in question.

Regardless of the philosophical merit of the arguments put forth, Sun has blamed the decision on Microsoft's influence with the committee, and Microsoft is doing what it does best: gloating.

On another front, Microsoft, while already being threatened with revocation of its Java license by Sun (for not honoring the Java standard), has introduced J/Direct--a new set of classes coupled with JVM modifications that will only exist in the Microsoft JVM with the stated goal of providing an "alternative" to the current system within Java for interacting with native application libraries.

Microsoft has said some interesting and amusing things on the subject, and we'll relish in eviscerating Microsoft's latest attempt at hijacking the Java standard in coming weeks.

Sun and Netscape, meanwhile, continue to focus their attention on the rather dubious matter of Foundation Classes. For those keeping count, there's now three sets of them: Microsoft's Application Foundation Classes, Netscape's Internet Foundation Classes, and Sun's Java Foundation Classes.

In short, these are all the fruits of another attempt to subvert the Java standard--in this case, by attempting to jump ahead of it. All of these packages offer "enhancements" to the 1.0 API (typically interface-related), which largely duplicate the capabilities of the 1.1 standard.

This, too, has provided us with a lot of humorous reading material, and we'll be back to study the phenomenon in depth soon . . . so stay tuned.