Also, as I think Olin confirmed a few days ago, there's little new
work being done with MIT Scheme.
Well, Stephen Adams is working hard on a new compiler for the system, and
there's a big emphasis in the new code to supporting Windows, so that might
turn out well.
It has a lot of
features, some of which might be handy (I recall seeing some X/MS
Windows support), some of which aren't to me (first-class
environments).
Some of the major players in the MIT Scheme development story had
the attitude that MIT Scheme was purely for what they wanted it. "You could
cut the wire that connects us to the net; I would not care at all,"
I heard once.
A problem I have with this is that it makes it hard to "mine" the
MIT Scheme sources for useful code, because it's not always R4RS,
and the library dependencies are kind of a snarled mess.
I made an effort *not* to be this way with scsh, but I am not sure
I succeeded. Strict R4RS is a unusable subset. I use multiple-value
returns all over the place. My libraries are a principled collection
of S48 modules... but so what? S48 lacks browser tools to help you
decipher structure from the module information. A lot of scsh's more
ambitious macros use unportable low-level Clinger-Rees macros.
(And, of course, some of the major players in the MIT Scheme development
story *do* care about outside use. But resources are always limited.)
Please note that I am certainly not an official spokesman for the
MIT Scheme folks; and I could easily have overlooked something.
My perceptions are limited in scope.
-Olin
|