On Thu, Jul 03, 2008 at 09:23:58AM +1000,
Mark Andrews <Mark_Andrews(_at_)isc(_dot_)org> wrote
a message of 32 lines which said:
No sane TLD operator can expect "http://tld"; or "user(_at_)tld"
to work reliably.
[Mark, you used non-RFC2606 names, the IESG will put a DISCUSS against
you.]
I agree but it is not the point: an email adress like
bortzmeyer+ietf(_at_)nic(_dot_)fr is legal and works but not reliably (there
are
many stupid broken Web forms which refuse it and tell me it's not
valid).
http://example is legal and should work. If it does not, it may
indicate a broken implementation.
But where should it resolve to? "example.example.net."
or "example."? Under what circumstances?
I suspect there are still mail configuations
around that will re-write "user(_at_)tld" to "user(_at_)tld(_dot_)ARPA".
There are many broken mail configurations.
Should we be writting a RFC which states that MX and address
records SHOULD NOT be added to the apex of a TLD zone?
No. A TLD is a domain like any other and we should not write special
rules for them.
Names with and without dots already have different semantics.
Should we be writting a RFC which states that single label
hostnames/mail domains SHOULD NOT be looked up "as is" in
the DNS?
I hate special cases.
TLDs are already a special cases in so many ways.
Mark
--
Mark Andrews, ISC
1 Seymour St., Dundas Valley, NSW 2117, Australia
PHONE: +61 2 9871 4742 INTERNET: Mark_Andrews(_at_)isc(_dot_)org
_______________________________________________
Ietf mailing list
Ietf(_at_)ietf(_dot_)org
https://www.ietf.org/mailman/listinfo/ietf