On 5/18/2004 10:05 AM, Eric A. Hall sent forth electrons to convey:
On 5/18/2004 11:55 AM, Matthew Elvey wrote:
On 5/18/2004 9:26 AM, Eric A. Hall sent forth electrons to convey:
1) collision with every other expiremental RR encoded into TXT,
Any specifics on one that matters?
Take a look at RP which uses TXT RR targets that are sometimes bound to
the associated domain. Every other experimental use is equally valid.
Looking. I'll grant you that every experimental use is equally valid.
SPF works just fine with RP. RP is more commonly defined a separate type:
dig elvey.com RP returns
...
;; ANSWER SECTION:
elvey.com. 7201 IN RP postmaster.elvey.com. .
...
See RFC 1183 <http://www.dns.net/dnsrd/rfc/#rfc1183>.
Even if dig elvey.com TXT is correct for RP, this still isn't a
problem. Say matthew.elvey.com were my MARID-protected domain.
Oh, no - I'm using 2 different experimental TXT RRs. No problem - take a
look.
Collision with other TXT RRs isn't a problem.
You do not own TXT, and no you cannot have exclusive use of it, which is
what your question implies.
What part of I don't expect exclusive use of it don't you understand?
No it does not; see my new comment below.
You snipped this text I wrote from my post, referencing the 'magic
number'-like start to a record:
I'd bet that other users of TXT would, like SPF does, typically mark
their records, and as pointed out, the typical SPF record is
30-something bytes, and any particular record does not need to be any
longer than that.
And yet you YELL, as if I'm the one not listening.
2) even if you move the RR to some other point in the tree, it's
a bad idea to be extra liberal with the encoding. large DNS
messages and answer sets have to leave the UDP-based
What large DNS messages? The evidence presented shows that SPF records
are weighing in at way under 512 B.
IRRELEVANT IF THERE ARE MULTIPLE TXT RESPONSES IN THE SET, INCLUDING THOSE
WHICH HAVE NOTHING TO DO WITH SPAM
You are claiming that TXT is used in important ways with packet sizes
that are very close to 512B - so close that they can't fit a handful
(~15) more bytes .
I claim that's false, and provide example evidence in
matthew.elvey.com's records of why that's all that's needed.
Again, not applicable. SPF doesn't require packets > 512B.
SETS OF TXT RRS ARE THE PROBLEM
If you *YELL* really loud, people will believe you?
Hmm. Let's try it. *ANYONE WHO SENDS ME SPAM TODAY WILL DIE IN 5 DAYS.*
The biggest 'delay' is in getting application developers to work around
native resolver API limitations.
Wow. So a worldwide rollout to anything like most of the DNS servers
and resolvers for 100 million domains using a new RR is infeasible in
the near future in the real world, based on the deployed software out
there, or likely to get out there.
Rolling out new RRs in modern servers is pretty simple. The problem is
with the client-side resolver APIs, which only offer a handful of
pre-defined query types.
But the fraction of the servers out there that are so modern that they
support new RRs is what?
I mentioned the resolvers.
BTW, Apropos the Yahoo's DK I-D wannabe-
http://antispam.yahoo.com/domainkeys/draft-delany-domainkeys-base-00.txt
posted to this thread:
It fails to meet IETF requirements for publication as an I-D, or for
consideration by this group.
This text is not allowed:
"This document may not be modified, and derivative works of it may not
be created. This document may only be posted in an Internet-Draft."