> I don't think this threat is tangible enough to be a real problem, but I
> see the argument. I think the convenience is worth the slight risk,
> though.
well, it's not really a matter of convenience, since it is trivial for
all users who want to to add . to their path ...
for that reason, it really makes no difference, although i agree that
since these are cs accounts, executing from . is pretty common, and one
would be expected to be clueful enough to put commonly executed personal
programs in a special directory that appears before . in the path to avoid
that type of problem ......
although, for organizational reasons, it makes sense for such
'user-defined' path to be set in a different place from the
'system-defined' path, and therefore for . to not be included in the
system-defined path ... so it's not that stupid to leave . out and let
users put it in, nor is it uncommon ...
-dave
______________________________________________________________________________
dave caputo | And you may ask yourself \ ceci n'est ___ /
dvc3@pantheon.yale.edu | Am I right?...Am I wrong? \ pas une ___ //~~/
caputo@cs.yale.edu | And you may say to yourself \ pipe [ ]_// /
http://glazunov.dc.yale.edu | MY GOD!...WHAT HAVE I DONE? \ `---' /