On Wed, 2010-02-03 at 23:07 -0500, Daniel Herring wrote: > On Wed, 3 Feb 2010, Robert Goldman wrote: > > > "The best is the enemy of the good." --- Voltaire > > So far, I have only written about pragmatically solved problems. Do you > really want to hear what I think might be best? ;) > > > > There seems to be a strong sentiment from at least part of this mailing > > list that ASDF loading should be reorganized so that some minimal core > > functionality is loaded, and more can be loaded on demand. > > > > I'd like to go on record as opposing this. I see the arguments for > > tidiness involved, but I think there are some powerful arguments against > > doing this: > ... > > Compared to make or autotools or cmake or bjam or ... ASDF is very small > in functionality, footprint, etc. Except for users having trouble > finding docs, it is also rather easy to install. > > Here's an approach that would make a lot of sense and solve numerous > backward-compatibility issues. The full ASDF toolset may be big and hard > to install [see below]. But a simple command drops a self-contained file > customized for a particular system. Thus end users don't even have to > install ASDF to use it! > > This approach also solves the bootstrap issue -- users can bootstrap using > the full functionality of ASDF. So it actually installs just like any > other library. > > - Daniel > > P.S. Any similarity to the configure.ac -> configure transform is purely > accidental. ;)
Did you forget an attachment ? -- Stelian Ionescu a.k.a. fe[nl]ix Quidquid latine dictum sit, altum videtur. http://common-lisp.net/project/iolib
signature.asc
Description: This is a digitally signed message part
_______________________________________________ asdf-devel mailing list [email protected] http://common-lisp.net/cgi-bin/mailman/listinfo/asdf-devel
