Plans for 2.8 - GNOME Managed Language Services?
- From: Ryan McDougall <ryan mcdougall telusplanet net>
- To: desktop-devel-list gnome org
- Subject: Plans for 2.8 - GNOME Managed Language Services?
- Date: Thu, 25 Mar 2004 22:05:44 -0700
I think its about time we seriously considered promoting other languages
to first class status, along beside C. We need C for speed in core GTK+
libraries, but as computers become faster, human time becomes more
important than computer time, and high level languages vastly improve
programmer productivity, as well as encourage new developers.
I think that the best way to include managed languages in the GNOME
desktop in a fair, open and free way, is to mandate a GNOME Virtual
Machine + Object/Type System and let the individual language designers
choose to target our platform at their desire. We would only provide the
base system, and let the implementors worry about JITting their pet
language to our VM. I would also like to see this done through
freedesktop, so it will be a "linux thing" not just a "GNOME thing".
The plus side of a unified Object/Type system is we can provide our
own .NET style high-level library stack which would be usable from any
language. The code reuse and power of a coherent and unified library
that works seamlessly in any compliant language (al la .NET) is
*extremely* tempting.
With the refusal of Sun to open up Java, that pretty much excludes it
from consideration as the base of implementing VM+obj system. That
leaves only three options:
1. Make our own solution - perfect except for the *huge* amount of work
this option represents.
2. Parrot VM+Object system - possibly not fast enough for statically
type languages
3. mono CLR+CLI - everything we want, but invented by microsoft
I think we should pursue the above options in the reverse order. We
should use mono's run time VM and CLI object system as a core part og
GNOME, but not dictate what language uses the components, whether it be
mono's C# compiler, or Java over IKVM, we won't care. Secondly,
exploring parrot should be an interesting investigation, provided its
developers are interested in working with us.
Ideas? Comments? We need to be ahead of the curve on this if we want to
gain ground on MS; Longhorn is going to be impressive when it finally
comes. ;)
Cheers,
Ryan
[
Date Prev][
Date Next] [
Thread Prev][
Thread Next]
[
Thread Index]
[
Date Index]
[
Author Index]