Tim Wilkinson states:
By providing a solution as Open Source we both unify
the two waring Java parties and enable developers
to choose their development tools, the extensions they want to use, and
allow deployment on any platform supported by Kaffe. You can now write
J++ code on Visual Studio and deploy on Redhat Linux.
and
I want Java to be controlled by the Open Source movement - standards
set by the people who actually use them.


This is a delightfully enticing argument. It's also utter nonsense.

The Open Source movement supports standards because they know it is the glue that ties the Internet together. They know that sharing a common standard is what allows the various different systems or programs to communicate and share data. They also know that a standard doesn't work by just letting everybody extend it at will, then watching to see whose extensions are being used by the most people.

If we followed Tim's argument, then the Mozilla project should drop its support for the W3C Standards, and return to the bad old days where Netscape adds their own HTML extensions, and Microsoft adds theirs, and whoever captures the most websites is the winner. Or, maybe GNU and Linux should go beyond their support for POSIX, to add support for Microsoft C++ extensions - after all, lots of people use MS C++.

You may not like Sun's licensing. You may not like Sun's control over their source code. But that is a separate issue from the Java Standard.

The Java Standard is thoroughly, and accurately documented, and is free for anyone to use. This is in stark contrast to the Windows API, with all its secrets and inconsistencies, which is where Java will head if you start accepting arbitrary changes from Microsoft.

There is an established procedure for maintaining and updating the Java Standard. Maybe that procedure is too Sun-centric for you, but in that case, work on establishing a new procedure - don't just arbitrarily extend the standard to suit one company (especially a company that is openly opposed to the basic objectives of Java).

Maybe you feel the Java standard is missing something, or that it changes too slowly, but that's not a problem - as you know, the Standard also provides ways to extend the language (through new classes or JNI), while still remaining consistent with the Standard. Microsoft, of course, ignored the documented approach, in favour of methods that tend to corrupt the language, and break WORA. Furthermore, MS's changes were completely unnecessary.

Do you know of any cross-platform standard that has survived an onslaught by Microsoft? Did C++ become a cross-platform standard, as was originally intended? Can we even define a common HTML standard anymore? When Microsoft promised to support the Windows API on Unix, did they continue that support, or was it just temporary, with the ultimate goal of paring it back, and forcing naive developers to move to Windows (see Bristol's lawsuit against Microsoft ). Microsoft opposes cross-platform standards, because standards allow people to make choices that might not include Windows.
 

Where Java is concerned, Microsoft has stated that they intend to:

     "kill cross-platform Java by grow[ing] the polluted Java market."
 

Tim is going to help them.
 

Shame on you, Tim. I won't be using your product.

Mike Cornall
[EMAIL PROTECTED]
 

Reply via email to