naming conventions, was: Case sensitivity

> it would be nice if some day Maxima users on capable platforms
> could use Greek and their native language without prejudice

Absolutely.  I have been involved with Unicode for many years -- and
even gave a paper on Unicode issues to a conference on Greek typography
a few years ago, so I am certainly aware of the issues.

Of course, whatever we do should extend cleanly to full Unicode.  And of
course, mathematicians use both uppercase and lowercase alpha as well as
uppercase and lowercase latin "a".  And of course, whatever case
conventions we have for Latin should extend to Greek and other
alphabets.  (I was unaware of the situation you mentioned where Latin is
case-insensitive and Greek is case-sensitive.  And there is no
implementation excuse for it either -- the mapping tables for Unicode
are available and easy to use.)

However, it is important to remember that Unicode is a *character
coding* convention.  It does *not* define the meaning of the characters.
Just because "b" and "B" are different characters in Unicode does not
mean that the word "belfry" refers to a different object than the word
"Belfry" (which might simply be at the beginning of a sentence).
Conversely, Unicode does not define all possible variants which might be
useful in mathematical texts -- although in a radical departure from
Unicode's original philosophy (based on UNIfication), a lot of these
variants are being added above the Base Multilingual Plane.  For
example, "1D706, MATHEMATICAL ITALIC SMALL LAMDA".  But anyway.

All this being said, and despite some nagging reservations, I guess I'm
now convinced that Maxima needs to be case-sensitive.  There is just no
clean way of making the mathematical symbols capital-a and small-a or
capital-lambda and small-lambda distinct (as they are in common
mathematical usage) while considering "diff" and "Diff" to be the same.

        -s

PS The business about Titlecase does *not* mean that "some
Eastern-European languages that have _three_ cases"; just that there is
a funny situation related to the Croatian (Latin alphabet) digraphs
"dz", "nj", and "lj" which are encoded as a single letter for unique
transliteration with their Serbian (Cyrillic alphabet) equivalent, but
which must appear as "Dz" etc. at the beginning of a capitalized word,
not "DZ".  Same issue with polytonic Greek prosgegrammeni (the iota
subscript becomes an *ad*script in titlecase).  These are the ONLY cases
in Unicode where titlecase is distinct from uppercase.