Murray S. Kucherawy wrote:
-----Original Message-----
OS vendors could have supplied it earlier.
The question is not whether the nameservers themselves support it (we know these days that they do), but rather the user interfaces that many people have to use to create records. For example, several web-based "domain control panel" things for virtual web site hosting know how to add the common types, but not others. This has even challenged deployment of DKIM, which doesn't have its own RRType and re-uses TXT, because it's strange for some people to post a big base64 blob into a web form for which the label is "hostname/address" or some such.
These are valid widest total infrastructure support goals, but I don't
believe its (and I believe RC5507 and Patrick is stating) are valid
arguments against the usage and promotion of using RR types. Well
understood, but not an excuse to abandon the ideal because in my
personal view a "Simplicity" factor has emerged with many ISPs
offering WEB-based DNS management frontends. I don' think that means
it wasn't possible for them to support new RR type record input - some
simply took the path to least support resistance.
Any-hoo, that is why i poise the question in regards to what are the
new "ideals."
Just consider, Murray, it isn't a big deal in 2006 for SPF/MARID
efforts, then it would had never came up as a suggestion as a IETF/DNS
community endorsement appeasing idea to help mode the I-D for what
become RFC 4408. After all, SPF already had a long list of IETF key
cogs who were militantly against it - and some who are still are now
deciding its BIS future.
The essence I am getting from limited WG participant leading these
efforts, is that it doesn't matter any more, don't bother to even
consider it any more. Its a waste of time.
--
Hector Santos, CTO
http://www.santronics.com
http://santronics.blogspot.com
_______________________________________________
Ietf mailing list
Ietf@xxxxxxxx
https://www.ietf.org/mailman/listinfo/ietf