Kaffe's MS-Extensions Will Hurt Java
Godmar Back
gback at pa.dec.com
Fri Jun 25 10:08:46 PDT 1999
Regarding that discussion about MS's extensions:
As I've already said, these things are low in importance and priority
that it almost doesn't justify a discussion about them, but there's
one thing I don't understand.
Why do people believe that MS's extensions are worse than Sun's
definition of Java? Let me see: if indeed Transvirtual released
them under an Open Source license, then it would be just like Sun's
Java... except, of course, Sun's Java is not available under an
open source license. So it seems what we're having here is that
company A defines a proprietary standard and releases an open source
implementation of that standard. Company B defines a proprietary
standard but does not provide an open source implementation.
Because people think that company B is more likable, company A
is accused of trying to pollute company B's standard.
Somebody also mentioned Sun's standardization process. Let's
look at an example. Real-time Java, for instance. Several companies
[1]
have worked hard during the last years to develop and agree on a
standard for real-time Java (PERC). They had DARPA funding, they all
had
experience in developing real-time systems, they drafted and
discussed a specification. Late in the process, when it came time to
decide whether NIST should adopt that specification, Sun lobbied hard
and the proposal was turned down even after an initial vote had
approved it.
At the same time, Bill Foote from Sun presented Sun's idea of
real-time Java at JavaOne 1.5 weeks ago. After spending half an
hour explaining general concepts, such as priority inversion and
generational garbage collection, he concluded by pointing out that
Sun is working on a specification, and that people should check
their website periodically. [2]
Asked by the audience if Sun was working with the Realtime
Java Working Group, Sun said that if others want to talk about
real-time Java, then others can talk about real-time Java.
If you check Sun's website [2], you'll see one link and one slide
presentation. Countering competition with slideware is a practice
that's not uncommon in this industry: it discourages early adopters,
seeds FUD and can help secure a company's position in the
marketplace when it does not have a product that can compete.
Why should we (on this mailing list) care whether company A or
company B does it.
- Godmar
[1] http://www.j-consortium.org/rtjwg.html
[2] http://www.rtj.org/
More information about the kaffe
mailing list