[R6RS] Case-sensitivity

Michael Sperber sperber at informatik.uni-tuebingen.de
Sun Feb 19 03:39:13 EST 2006

[Finally catching up with Wadler's Law now ... :-} ]

Anton van Straaten <anton at appsolutions.com> writes:

> Working with an XML dialect in which many tags have to be quoted with 
> |...| or "..." would significantly detract from the usability of 
> something like SXML, IMO.  The code would look more cluttered and less 
> readable, which undermines one of the benefits of using SXML over XML.

Well sure, but that's XML/SXML---it's a specific embedded language.
But what about other languages with different kind of identifier
syntaxes?  Say, ML or Haskell.  XML may go away again at some point in
the future, and get replaced by a case-insensitive language.  I just
don't think this one example is sufficient justification to do
make that kind of sweeping change.

> What's your opinion about providing a mechanism to choose between 
> case-sensitivity and insensitivity?  In the archive-reading I've done, I 
> don't recall having seen discussion of that.

I think that's dodging the issue.  Maybe this is what we want to do
(seems to've worked for Matthew), but I'm afraid of the complications
due to it.  People are likely to get confused about what mode they're
in (I get confused all the time with PLT Scheme), and get funny
results when they mix case-sensitive and case-insensitive code
inadvertently, especially when using macros.  Some module will export
a mixed-case identifier, and somebody in case-insensitive mode is
going to scratch their head.  It all smells suspiciously of the worst
aspects of Common Lisp.

Consequently, I'd rather be case-sensitive than case-ambivalent.

Cheers =8-} Mike
Friede, Völkerverständigung und überhaupt blabla

More information about the R6RS mailing list