From mboxrd@z Thu Jan 1 00:00:00 1970 From: Ken Raeburn To: guile-emacs@sourceware.cygnus.com Subject: guile and emacs Date: Sat, 28 Aug 1999 11:21:00 -0000 Message-id: <199908281821.OAA14563@raeburn.org> X-SW-Source: 1999-q3/msg00000.html Hi, gang. Sorry you haven't heard from me in a while. Here's what's up: My RSI is improving, but not gone. I'm starting to put more time into Guile-Emacs again. Emacs 20.4 is out, and a bunch of redisplay changes and other changes are now getting merged into the Emacs sources. I've also got a chance to merge in some of my initial changes for making the main body of the Emacs code less aware of the Lisp internals. That'll be what I focus on next. At http://www.xemacs.org/Architecting-XEmacs/index.html there are some ideas some of the XEmacs folks have come up with for improving the Emacs/Lisp interface, which appear to be independent of the Emacs/XEmacs split. In particular, Michael Sperber is looking into moving XEmacs to a general Scheme base. Not specifically Guile, for two reasons: 1) a more general Scheme interface allows later retargeting if desired, or giving the user the option of multiple back ends for comparison, and 2) he's got some "issues" with Guile's implementation decisions, including the use of conservative GC. What he's saying makes sense, so I'll be looking a bit at separating further into "Lisp" vs "general Scheme" vs "Guile" bits what I've done so far. (GC will be an interesting issue though; I've got some thoughts on that.) Mike's a researcher in programming languages, and has been working on (among other things) Lisp->Scheme translation, with a prototype translator already in the works. We might be able to build on this. There's also occasionally some relevant discussion in comp.lang.scheme and comp.lang.lisp; that's how I got in touch with Mike. Jim Blandy has been working in Japan with the MULE folks to figure out a good way to do multilingual support in Guile. Sounds like it'll be functionally similar to the Emacs MULE implementation, but not identical. He's proposed an interface, so I may start experimenting with a stub (unibyte-only) version of that interface. A snapshot of the code I've been working on is at http://www.mit.edu/~raeburn/guilemacs/guilemacs.tgz if you want to play with it. It includes CVS admin files, so if you've got CVS and an FSF account, you'll be able to do updates starting with that version. (For those of you I've already told about it, don't bother fetching it again, it's the same old snapshot.) The post-20.4 changes have been pretty huge. I'm going to try to get some of my more basic changes merged into the main sources before trying to merge that stuff back into my CVS tree. That's it for now... Ken