> I [...] discovered that on some architectures there are source files
> that will result in object files that differ in only the case of
> their name.  [...]

> What do you think?

I do not expect this to be a popular viewpoint.  I do not expect it to
be the one the Project will adopt; I'm not even sure it's one the
Project _should_ adopt.  But you asked. :-)

I think that filesystems that conflate names that differ only in
alphabetic case are barbarisms that should be worked around when
necessary but never catered to.  If you have to deal with one,
interpose a mapping layer (Eunice comes to mind as an example).  I'm
surprised cygwin doesn't do this - I would most certainly expect it to,
even to the extent of calling it a bug if I discovered it didn't.

If nothing else, the notion of case-folding is charset-dependent.  Are
0xdc and 0xfc the same except for case?  In 8859-1 they are.  In 8859-7
they are not.  In 8859-8 neither one is even a letter.  (At least if
what references I have at hand are to be believed.)

/~\ The ASCII				der Mouse
\ / Ribbon Campaign
 X  Against HTML	       mouse@rodents.montreal.qc.ca
/ \ Email!	     7D C8 61 52 5D E7 2D 39  4E F1 31 3E E8 B3 27 4B