No-one's claiming that, AFAICS. The v=spf# string neatly takes the
mimimum slice of TXT that it can work with, leaving it otherwise free
for its current uses.
It means that any query for the TXT record for a domain is goign to end up
returning the SPF record plus everything else that is there. SPF is not the
first spec to propose using the TXT record for a protocol purpose, but it is
the first one that is likely to be widely used across domains.
The problem is that a lot of people already using TXT records inside domains
are going to suddenly find that their TXT queries are longer by the length
of the SPF record.
This in turn is likely to end up causing a lot of queries to exceed the 500
byt limit on UDP queries and TCP/IP fallover.
If anyone's yet seen even a partial collision with TXT
records starting
with something approximating v=spf, please point it out.
The collision is on retrieval, as I explained.
Once SPF takes off we can get its own record type, but until it does,
that won't happen - we might as well ask for a specific
record type for "domain owner's geek code".
Hoping to one day get a RR is the worst solution of all. It means that the
spec will be considered provisional until you get a RR - which is
realisticaly five years off.
Beside which, of course, TXT works. _spf doesn't (or at least when I
asked before how to put an underscore into a record name in
BIND, no-one could answer).
Just do it.
The underscore convention was invented by Paul Vixie who wrote BIND and is
also the author of the SRV draft.
Can you expand on that (or give me some pointers to look elsewhere)
purely for my curiosity?
later today I hope...
-------
Sender Permitted From: http://spf.pobox.com/
Archives at http://archives.listbox.com/spf-discuss/current/
Latest draft at http://spf.pobox.com/draft-mengwong-spf-02.9.4.txt
To unsubscribe, change your address, or temporarily deactivate your
subscription,
please go to
http://v2.listbox.com/member/?listname(_at_)©#«Mo\¯HÝÜîU;±¤Ö¤Íµø?¡