btw: this isn't flame bait... if it's too off topic, since no one else
is participating, I'd be happy to discuss these things in private
emails.
On Tue, 12 Sep 2000, Petr Novotny wrote:
> Well, Fortran's rules for indexing come from the same teapot.
Yes, and we all know how much attention every programmer pays to memory
architecture these days. Writing a program to display today's date --
sure, why not include a web browser in that! What's a GIGABYTE of
RAM these days? :-) (Humor? not funny? sorry.)
> is not a cycle for www (a record) but might be a cycle for mail (mx
> record).
Agreed... but, alas, cycles/loops vs efficiency is still a weak argument
-- in my camp. If you hose your MX cnames, you deserve your lost/queued
mail. Hardly seems any different than 1 month TTLs and then changing
things and dealing with that fallout for a month.
> There _are_ dinosaur programs which actually choke on
> MX->CNAME.
Question: Why are there not so many non-dinosaur programs that
actually choke on MX->CNAME?
> > HOW MANY MAILERS REFUSE TO ACCEPT BARE CARRIAGE RETURNS?
> Looking for a flame, aren't we?
No. Although it's not qmail... per se... I've constantly had to tell
people "Gee, I run qmail... if your system can't send mail to me, but
other systems seem quite capable of it, I'd tend to put the blame on
your system." And then I get back the usual "but I can send mail to
other systems just fine." to which I respond, "but are they running
qmail"
Most people tell me to "fix" my system. Of course, I refuse. It's
really bad when one of those people is your boss... and he's a little
tweaked that he can't seem to send mail to you. Ya know?
> How many mailers are "transparent-reliable" - I mean guaranteed to
> leave a message intact?
I'm not really sure I follow you.... perhaps the soon-to-be de facto
standard of having the HTML formatted body of the message as an
attachment doesn't give you a warm fuzzy feeling? and we all know that
visualization of HTML is left up to the display agent. *shudder*
Scott
