CLIM mail archive

[Prev][Next][Index][Thread]

1 Re: Using Processes in CLIM0



1While I agree that having a notion of processes in CL is a good thing, I
think that CLIM should be part of an overall ansi standard that also has
some endorsed process model. CLIM-SYS then makes no sense; it should be just
part of CL. (much as the error system or CLOS is). At worst, it should be in
it's own process package. The reason for this is that processes have wider
application than windows (or at least potential different, if overlapping
audiences). Having some window system define the process model is a bit
inverted. Having some window system qualify the process model is another
thing (that I would approve of).

As for what I would like, I'm partial to the symbolics implementation of
processes, but can live with adaptations of it, such as are in Allegro. The
main thing is to have a notion of stack groups, which processes are built
on. This allows for implementation of notions such as dynamic closures and
coroutines that I would very much like to take advantage of in a portable
fashion.

----
 Brad Miller		U. Rochester Comp Sci Dept.
 miller@cs.rochester.edu {...allegra!rochester!miller}


0,,

References:

Main Index | Thread Index