On Tuesday, 20 July 1999, Sean Cribbs wrote:
>I have Heise's code. It doesn't work on the new JDK.
>
>I am running JDK1.2.1 on Solaris 2.6.
Either it runs all of the old code or it doesn't. People want code that
works. They don't want excuses. They want to hear that your code still
works. And it will survive the next half-dozen upgrades.
I'm sorry the old code doesn't work. I don't want this kind of thing to
happen each time we upgrade to a new version of JOS. Sun Microsystems
believes that the new JDK is compatible enough to run all of the old code.
And then, they have a long, long list of exceptions.
The old code I wrote doesn't work on the new JDK. I believe that this is a
threat to an open source project. Why should we write a bunch of code for
Java 2.0 (1.2) code if it isn't going to work on Java 3.0?
It stings. It hurts. It's rude. Sun Microsystems should be ashamed of
themselves. They use a practice that doesn't work. It doesn't have to be
that way.
That's why we should not immitate Sun Microsystems when we create a JOS
API. That is why it is so important to try to make long-lasting interfaces
for these JOS-related things.
I don't know what else to do about it. I keep hammering away at the bad
attitudes. It is not acceptable to build reasonable incompatibility.
Reasonable incompatibility means that it doesn't have to work. Heise's code
should work today on a JDK 2.0 just as it did when it was written. Isn't
that what Write Once really means?
_______________________________________________
UI maillist - [EMAIL PROTECTED]
http://jos.org/mailman/listinfo/ui
