On Feb 22, 2009, at 6:10 PM, Andrew Reilly wrote: > On Sat, Feb 21, 2009 at 09:03:30PM -0800, Guillermo J. Rozas wrote: >> - The great inconvenience and lack of respect for existing users >> which the change causes/caused. Even though I much prefer case >> insensitivity, I would not argue for C to change. It would be >> too disruptive. >> Why are Scheme users afforded any less consideration? > > This is the bit that I don't understand. I thought that the > argument for case-insensitivity was to prevent (presumably > 'other') programmers from using symbols that differ only incase, > because that would be confusing. So it follows that good, > existing code does *not* use symbols that differ only in case, > because that would be confusing. So from the point of view of > existing code, case-sensitive or case-insensitive ought not to > matter.
No. If you can rely on a case-insensitive language, you can use case as part of your conventions, and those conventions might use different case in different uses (like English with beginning of sentence). Ken Dickey showed such an example. That's a tool you remove when you make the language case sensitive, besides introducing 2^n possible spellings that all differ, even though that's not very common. But I have seen some C programs that rely on case sensitivity in confusing ways, and they were not trying to be obfuscated. Furthermore, just like syntactic sugar (of the macro variety or not) leads to cancer of the semicolon (or the square bracket, as the case may be), in case-sensitive languages, people start using horrible typing conventions such as caml-case. _______________________________________________ r6rs-discuss mailing list [email protected] http://lists.r6rs.org/cgi-bin/mailman/listinfo/r6rs-discuss
