Re: Case Sensitive File Systems...
@ the AC who said: "Multi-lingual case-sensitivity involves characters that a westerner couldn't even recognise again the next day"
Case conversion isn't a big deal, actually, because most of the world's writing systems are actually single-case. It's the various flavours of Roman/Greek/Cyrillic scripts that have this oddity of using a big and small version of the same letter in a way that conveys additional meaning (e.g. "he helped my uncle jack off his horse" does not have the same meaning as "he helped my uncle Jack off his horse.")
The idea of two coded symbols with the same semantic value is a peculiarity of European writing. Blame the Christian monastic copyists, who found it too tiring to write the angular Roman and Greek letters and instead developed a rounded script more suited to faster transcription of those texts. Cyrillic stems from Greek, and thus are also coded as a two-case script. Armenian is the only other cased script, and there's (tenuous and contentious) claims that its writing system stems from Greek, but in any case its writing was also part of the Christian tradition of monks copying Latin and Greek manuscripts.
There are some unexpected case conversions, but only really within Latin, because it's used for so many languages. The traditional capitalisation of the Dutch word "ijsbeer" is "IJsbeer"; The German word "Fluß" when written all-caps is "FLUSS", and if you're Turkish, 'i' and "I" are two separate letters, whose opposite cases are, respecitvely, 'İ' and 'ı' (that latter one catches most naive text processors, and people who accidentally imbue their C code with the user's locale...). You can code canonically for each of these (even the Turkish example can be unambiguously accounted for by coding 'i' as "dotless lowecase letter I", "combining overdot").
But my point is that once you get outside the reach of the Christian monastic tradition, you find that just about every other language gets buy with just one letter case, once you realise that rearrangement (Indic scripts), clustering (Indic and Thai) and presentation forms (Arabic) are not letter cases, but rather different appearances of the same letter.
Biting the hand that feeds IT
