Anton van Straaten
anton at appsolutions.com
Fri Feb 17 13:43:22 EST 2006
Michael Sperber wrote:
> I'm concerned about our decision to make Scheme case-sensitive.
> So---do you agree we should have a rationale on this? If so, what is
I agree that it would be useful to have an "official" rationale
available, even if it's not in the R6RS itself.
Although I usually prefer to work in a case-insensitive language, I
don't think that simply sticking with case-insensitivity would be good
enough. For example, case-sensitive symbols in Scheme programs are
necessary when working with case-sensitive XML data using SXML. The
same goes for any other embedded sublanguage which maps to some
case-sensitive external language.
Ideally, I would like to see a standardized way to choose between the
two options, such as the #cs and #ci expression markers in PLT.
I had previously thought that requiring support for both modes might be
too demanding as a requirement for R6RS. However, on c.l.scheme today,
Jens Axel Søgaard raised another issue:
> From an implementation point I don't think it is an either-or. In
> order to support r5rs-code, implementations need to keep some kind of
> case-insensitive mode.
At least in the short term, then, implementations are likely to support
both, which makes me think that standardizing the mechanism for choosing
between modes could make sense.
More information about the R6RS