Re: Plans for 2.8 - GNOME Managed Language Services?


First of all - before I say anything here
(1) I do not speak for my employer
(2) I am reluctant to get caught up in this discussion, and may feel a need to bow out over some sensitive issues, particularly where my opinions differ from those of some Sun management. So please try to bear with me if I don't pursue all responses.

That said...


I guess you feel that if an app was written in Java, any VM could run
it, which belies the assumption that what we want is a *language*, and
what people are concerned about is the VM *implementation*. I think its
the other way around. ...

Once again, this boils down to the question, "Do we want to be the
language or the platform?"

I think that both are important to some of the people and groups with strong feelings about this. Sun would almost certainly wish to use its own runtime VM on its bundled systems, for a multitude of reasons. Also any group with an investment in Java may tend to see C# as a "competitor" or at best redundant, and so would have issues with shipping/supporting a C# code base.

Of the two, I think for Sun in particular the VM issue is probably the touchier one. If it's possible to write GNOME code that can run on a free runtime but which can also run on a proprietary runtime, then we'll avoid the thorniest issues and give the corporate interested parties the most flexibility.

Also - and this is important to bear in mind I think - RAND licensing isn't free, it presents real problems to free distributions. Why? A "reasonable" license fee conflicts with free software distribution, since it can't be collected, and making a special allowance for free software / non-commercial use etc. is inherently "discriminatory". (For this very reason W3C largely reversed its position with respect to RAND licensing of its standards, in favor of royalty-free licensing, a couple of years ago).


If we're going to use choice of C# or Java, plus GNU Classpath, what is
the point of the C# part? It loses support for GNOME, in order to get an
incremental improvement in language syntax.


I agree with Havoc here, from a pragmatic viewpoint.

- Bill

[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]