"Guillermo J. Rozas" <[email protected]> says:
> Macros were something to be used sparingly, when absolutely necessary.
In general I've long since given up following Guillermo-vs.-Eli (more
because of too much included text from past messages than because the
topic isn't interesting :-), but this caught my eye.
\begin{rant}
Before R5RS, macros were ad-hoc, unportable, and ill-behaved in edge cases.
No question about it, hygienic macros are a big intellectual advance,
raising the possibility that those things could change.
But there has been a /huge/ cost. This case-sensitivity discussion is
just one of many in which useful features of what I still think of as
the actual Scheme language (before the macro overlay) founder on weird
incompatibilities with macros. (The REPL!!!! Give me a break.)
Unlike some other things, this is /not/ the fault of R6RS, although R6
has moved further down the road built in R5. So it's at least somewhat
independent of the recent contentious issues.
We now have the macro tail wagging the Scheme dog.
I don't have a technical fix to offer, but it would make me very happy if
the Steering Committee announced a policy that if the needs of macrology
conflict with the needs of traditional macro-free Scheme practices, it
should be the macro standard that has to jump through hoops to fix it,
not the core Scheme part of the standard.
I know this is unlikely.
\end{rant}
_______________________________________________
r6rs-discuss mailing list
[email protected]
http://lists.r6rs.org/cgi-bin/mailman/listinfo/r6rs-discuss
