This is the mail archive of the guile@cygnus.com mailing list for the guile project.
| Index Nav: | [Date Index] [Subject Index] [Author Index] [Thread Index] | |
|---|---|---|
| Message Nav: | [Date Prev] [Date Next] | [Thread Prev] [Thread Next] |
Jim Blandy <jimb@red-bean.com> writes: > As far as Guile's size is concerned: > > What we're trying to address here is not really a technical issue. > It's more of a psychological one: developers will hesitate to include > Guile in their applications if they feel, on a gut level, that it's > too large. > > There are two ways around this, that I see: > > 1) Partition Guile, to separate the essential from the inessential. > This makes it easier for people to see what they're really paying > for, and to buy in only as much as they want, and to gradually > increase their buy-in over time. > > 2) Give Guile features they want to use, regardless of its size. > Tk plays this role for Tcl. > > Hopefully it doesn't have to be one or the other (useful features + modular). I actually like the idea of slimming down guile, and when hobbit is able to compile any arbitrary guile code, it would allow lots of bits of guile to be written in scheme without any major penalty, which, IMNSHO, is a very good thing; c code can be a nightmare to manage and maintain; in guile, sometimes it's a bit of a nightmare to figure out what it's doing ;). It probably isn't one of the biggest problems with guile right now, though. I think it gets mentioned a lot because guile is slow in some areas, and the default reaction is to think slow == bloated and large. Provided we have a brainless way to package up an application with the bits of guile it actually needs, it hopefully won't cause everyone to eventually suffer because of it. And, there should be a guile-lotsa-added-stuff executable, for those of us who love languages with the kitchen sink :). <rant> It isn't a mindset that I can really understand: among useful languages, you'd be hard pressed to find one that could be considered small. Lisp & friends often get slapped about in this way because they aren't the default on a lot of systems (this is the real issue, I think), while the same people complaining about their 'enormous' size have no problem using the c library; the enormous chunk which is the c++ library; java (the ugh); perl. If guile could become a standard component of linux/*bsd/hurd, the functionality would suddenly be the issue, not the size (it's already on the system, so no one really has to know or care how big it is; I'd think python has experienced something of the same situation, and has really benefitted from RedHat using and including it with the system). The bit that I don't like is that you have to sort of mislead people, but I guess that's how marketing works. </rant> -- Greg, how about "Cindy Crawford will be all over you if you use guile and drink GNU beer?" ;)