On Wednesday 14 July 2004 15:17, Markus Triska wrote:
> > 2. Its main developer (Tom Lord) is desperately in need of cash, as he is
> > currently unemployed. (or at least was the last time I checked).
> Visit http://gnuarch.org/ for more information. While he is de facto
> unemployed, as you say, he puts it more brightly:
> Are these "after hours hobby projects" or what? In fact, no -- since early
> 2002, these projects are what I do. I don't have a day job that subsidizes
> this work. Although I'm now working on developing some start-up projects,
> in the meantime...
> By the way: Tom Lord is also working on a new implementation of Scheme
> (Pika Scheme), supporting Unicode. Considering that we could use Tom's Pika
> Scheme instead of TinyScheme, and that he can work on Pika Scheme by living
> on donations that come from his Arch project, it follows that Arch is
> without a doubt a proper sub-project of the GIMP.
Another implementation of Scheme? Aren't the ones in:
enough? Or isn't any of them better suited as a starting point?
That's one problem in Scheme: there is a plenthora of different
implementations, each of them different, and none of them as fully usable as
Perl (along with CPAN), Python (along with the standard library and other
libraries), etc. In these languages, the implementation is the standard, and
instead of having a minimalistic and useless standard, and tons of developers
with minds of their own creating competing implementations, there is one
development team, many halo developers creating extensions, bindings, and
support code, and generally a much better usability.
I'm not really a great believer in concentration of efforts when it comes to
open-source projects. But the Scheme situation is ridiculous.
This is not the only problem with Scheme. Paul Graham set out to resolve the
problems with the various LISP dialects, in creating Arc:
>From what I read about it, so far, it seems like it has the right direction.
It will also put LISP more up-to-par with the other agile languages. (Perl,
Python, Ruby, etc.) However, there hasn't been a trace of an implementation
or even a well-defined spec, yet.
> I'm therefore all the
> more for GIMP to win.
Do you mean that you're all the more for Arch to win? I doubt the GIMP prize
money (assuming we win) is going to go to Tom Lord.
> Gimp-developer mailing list
> [EMAIL PROTECTED]
Shlomi Fish [EMAIL PROTECTED]
Knuth is not God! It took him two days to build the Roman Empire.
Gimp-developer mailing list