Re: [Gnu-arch-users] is there demand for itla?

[Tom Lord]

    > From: Ian Duggan <address@hidden>

    > Ok. It's starting to shake out a bit, then. itla is an overarch
    > runtime, and overarch is a library of best practices. Those are
    > excellent targets to aim for.

    > My only concern is whether the choice of guile will attract a
    > large enough set of developers to make overarch grow quickly/be
    > useable to a large population.  I like scheme/lisp and all, but
    > just not sure how many people out there can hack on it. You
    > don't think it would be a limiting factor?


Not really.  There is a kind of lisp renaissance going on these days.

I'm not too fond of Guile in particular but then I'm not too fond of
any of the free implementations.  As a place to start, the factor that
"it's a GNU standard" is as persuasive an argument as any.  It's
sorely tempting to pick systas instead just because it has everything
I know I need but, alas, it also isn't as widely deployed as Guile and
is closer to an R4RS rather than R5RS Scheme.

Doesn't matter much, though.  Unlike Python, Perl, and Ruby -- Scheme
is _comparitively_ (far from perfectly) specified language.  ITLA will
port fairly easily across many implementations, present and future.



    > If Arch truly is at a crossroads and attention/money is what
    > will move it forward, then we should focus on these things.

Got any clever ideas :-)?


    > On the funding route, perhaps we should consider looking to
    > linuxfund.org for funding. I'm not sure what's involved in this
    > but arch seems like a great candidate to me.

I've followed linuxfund.org for quite a while.  I don't think it will
cut for lots of reasons that would be impolitic to enumerate.  It's a
nice theory.

pubsoft.org is another one.


    > 2) Windows port. This is a big one. 

I side with those who say that the only reason to offer windows
support for anything is purely tactical: the immediate, tangible
payoff has to be higher than the cost.   


    > Well, I don't think there's any danger of SVN reinventing
    > it. They'd have to start over with a different approach
    > methinks.

Sort of, but not quite.   You can morph continuously between the two.

The net result will be _vast_ amounts of wasted time on svn features
that are crap (and that transitively waste the time of users using
them) --- but the final product can wind up being rather more arch-
like than not.


    > I don't see how you could transistion from a centralized
    > repository approach to a distributed model.  

I do.  But I don't want to give them many more hints than I already
have.   At least not gratis.


    > You just end up
    > with something like Clearcase which with replicated repositories
    > instead. But I get your point.

Nah.  They can do far better than that.


    > > As for apache-style voting: nah.  We aren't quite there yet.  The arch
    > > community is too small and there are some loud voices who don't quite
    > > have the arch vision.  It would, I strongly suspect, go widly awry.

    > What exactly is the "arch vision" 

If I could sum it up in a sentence or two I would.

>From your perspective, I think you have to note that I've resisted a
heck of a lot of suggestions since early 2002 and the net result is
that people like you are lately more interested in arch.  Even if you
can't quite resolve the pattern in your field of vision, you should be
able to tell that there is a coherent and potent pattern there.

-t