[R6RS] proposed library/syntax-case changes

dyb at cs.indiana.edu dyb at cs.indiana.edu
Mon Aug 28 13:59:27 EDT 2006

> http://scheming.org/cgi-bin/mailman/private/r6rs/2006-April/001155.html
> Which I intended to mean exactly what I take you to mean now.
> To this, you objected strenuously.  (And if you track down my
> follow-ups, you can see that they differed on exactly this point.)

Not exactly.  Before I had a chance to respond to this, you also suggested
adding an eval-libraries form to the library header that restricts the set
of libraries that library-environment could name.  It was the combination
of a tight restriction on available libraries and no static restriction on
which are actually used by a particular call to eval that I didn't like,
because I felt that the eval-libraries restriction alone was rather
useless.  This led to the-library-environment, which I rather liked. 
Unfortunately, that had to be tossed in favor of something less static to
accommodate phased library semantics.  Now I'm back to not liking the the
"for run" (nee eval-libraries) restriction, all the more so because Aziz
Ghuloum has given me a fresh perspective on the negative consequences of
the restriction, as I described in the first note of this thread.

> This wouldn't be a problem in and of itself, if you hadn't convinced
> me of your position in the meantime.

Sorry about that.


More information about the R6RS mailing list