> It was my take that Mr(Ms) +11234567890 ...

Note that I closed the initial portion of my message body with "=JeffH". But 
also note that it take clues/hints/traditions/conventions outside of the rules 
of the set of email RFCs to identify what might be the nominally socially 
correct "name" (note "names" are not always "identifiers") to use to refer to 
me. And you'll get different answers from different entities. (some would 
perhaps suggest "obnoxious" or something similar).  ;)

Anyway, from my perspective, either is fine for the purposes of this discussion.

Also note that I copied my message to "[EMAIL PROTECTED]", which by 
the way, will indeed "reach" me via email, even though the "local-part" is 
indeed a valid +1 PSTN number, but not one I "control", beyond it's association 
(in this context) with kingsmountain.com. Hm, I may need to enable SIP-based 
comms for it....


Paul Kyzivat wrote in reply to Adam Roach::
 >
 >                         But there are a lot of people who seem to think
 > it is fine to take any user part that looks vaguely line a phone number
 > and treat it as such.
 >
 > It was my take that Mr(Ms) +11234567890 specifically feels that way.

Actually, no, that is not what I was attempting to say.

Rather, what I was attempting to say is that..

   It is nominally not correct (for some definition of correct)
   to take a "user" part (aka "local-part" in RFC2822 parlance),
   from the "[EMAIL PROTECTED]" sub-construct of the "SIP URI"
   construct defined in RFC4474, and interpret its real-world
   attributes, or attibutes it might have in another protocol context,
   outside of the current protocol context (i.e. SIP),
   unless hints/extraContext/whatever are also present in the
   latter protocol context -- regardless of the syntactic
   construction  of said "user" part.

I also agree with a converse of the above..

   It is nominally "a leap-of-faith" to take a "user" part
   (aka "local-part" in RFC2822 parlance), from the
   "[EMAIL PROTECTED]" sub-construct of the "SIP URI" construct
   defined in RFC4474, and interpret its real-world
   attributes, or attibutes it might have in another protocol
   context,  outside of the current protocol context (i.e.
   SIP) -- regardless of the syntactic construction  of said
   "user" part -- unless, hints/extraContext/whatever are
   also present in the latter protocol context.


Paul Kyzivat went on to say:
 >
 > My understanding is still that even
 >
 > "sip:[EMAIL PROTECTED];user=phone" and
 > "sip:[EMAIL PROTECTED];user=phone"
 >
 > are *different* and should be
 > announced as such. Certainly the routing rules are different for
 > reaching the two of them, and that can be significant.

There was sentiment in the room here @Philly/IETF-71 yesterday morning, that 
the "user" part of those two above identifiers SHOULD be treated by UAC/UA/UE 
as the "same", and that various present SIP RFCs at least imply such (I don't 
know myself, need to research it), though it is not currently consistently 
implemented that way.


 > IMO if you want an address that can be displayed simply as a phone
 > number, then you should use a TEL for it.

Sure, for header fields such as To and Request-URI, but the issue here is what 
to assert in the From header field such that one can sign it (and other SIP msg 
components) and have some sort of linkage back to a trust anchor.


=JeffH
aka [EMAIL PROTECTED]

ps: I didn't send my original message from my actual @messaging.sprintpcs.com 
email address for two reasons..

1. I didn't want that email address and phone number published on the web in 
email archives.

2. I didn't want to receive the SIP mailing list as SMS messages on my 
PDA/phone (scary).

---
end











_______________________________________________
Sip mailing list  https://www.ietf.org/mailman/listinfo/sip
This list is for NEW development of the core SIP Protocol
Use [EMAIL PROTECTED] for questions on current sip
Use [EMAIL PROTECTED] for new developments on the application of sip

Reply via email to