On Sun, Aug 3, 2014 at 10:39 AM, Doug McIlroy <doug@cs.dartmouth.edu> wrote:

Because the effects of buffering are not transparent, it would be
wise to make it an optional optimization, not a default.

It's not just efficiency, it's also behaving compatibility with other implementations. Or were you going to argue this with e.g. the ANSI C committee too?

brandon s allbery kf8nh sine nomine associates

allbery.b@gmail.com ballbery@sinenomine.net

unix, openafs, kerberos, infrastructure, xmonad http://sinenomine.net