My input on the issue of TXT vs. RR.
When it comes to DNS, I am not enough of a cog to know the technical issues
related to TXT vs. RR.
However, from a developer standpoint, unless there will be some
programmatic algorithm to automate the process of determining which is used
to query a domain policy, there should be ONE record type, not two. Again,
if there will be two, the logic needs to be in place to automatically
determine what is to be use. The last thing you want is for a MARID client
to be doing two or more different record lookups.
Personally, I have trouble understanding why anyone would start a new design
with a "kludge" concept. But again, I don't know why the TXT vs. RR is an
issue. Maybe would be kind enough to sent me a personal message with the
basic issue between the two.
Finally, I hope I am not being redundant but I would like to point out again
that the majority of the mail is spam and while MARID made help protect the
spoofing of domains, most lookups will be failures (NXDOMAIN or none). This
efficient issue should not be ignored.
--
Hector Santos, Santronics Software, Inc.
http://www.santronics.com
----- Original Message -----
From: "william(at)elan.net" <william(_at_)elan(_dot_)net>
To: "Michel Py" <michel(_at_)arneill-py(_dot_)sacramento(_dot_)ca(_dot_)us>
Cc: "MARID" <ietf-mxcomp(_at_)imc(_dot_)org>
Sent: Friday, June 11, 2004 7:53 PM
Subject: RE: Alternative to TXT or new RR was: Comments on
draft-ietf-marid-core-01 xml use
On Fri, 11 Jun 2004, Michel Py wrote:
william(at)elan.net wrote:
[..] I was joking because while this may seem like a way out,
in no way do I think it should be done or that it'll ever fly
by IETF top guys. We should never attempt to completely take
over somebody else record, that is a very bad precident for IETF.
Then, which one out of the three alternatives I posted yesterday would
you recommend (or add your own)?
1. We do it quick and dirty with TXT.
2. We do it right and slow with a new RR type.
3. We say that we'll go with 1 as a temporary
measure and with 2 in the long run.
I'd prefer #2 actually.
I think it would not be as slow as you predict because greater majority
of mail operators are not using microsoft os (> 80%) and will be able to
use new MARID record type quite quickly. Having situations that unix folks
are able to do it while microsoft can not will put greater pressure on
microsoft (plus their charimans's comments that spam problem would be
"solved" by 2006) and might even cause to have patch for dns resolver &
dns server released before next OS.
That said I'd be ok with option #3.
--
William Leibzon
Elan Networks
william(_at_)elan(_dot_)net