I reckon the real problem in this scenario is the inclusion of the
user=phone parameter in the address-of-record in the first place. If the
parameter is used to signify that the number is known to be a PSTN number,
its only value is to indicate the fact to someone who wants call the
address-of-record. It doesn't seem to be a very satisfactory mechanism for
this purpose, but if it is used, the original problem could be solved by
registering two address-of-records, one with the parameter, and one without.
In our products we do a similar thing when we register with a digits-only
user component _and_ a more general user component, in order to cater for
callers who can only specify digits, and to cater for users who can specify
more cuddly names.

Barry Desborough

> -----Original Message-----
> From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]]
> Sent: 04 January 2002 05:05
> To: Shail Bhatnagar
> Cc: Jonathan Rosenberg; Barry Desborough; sip-implementors
> Subject: Re: [Sip-implementors] user=phone in registered addresses
> 
> 
> 
> 
> Hi,
> according to bis-05 statement as below:
> 1384 When the registrar has determined that the client is 
> permitted to make
>  the request, the registrar MUST
> 1385 extract the address of record from the To header field of the
> REGISTER. Note that the registrar MUST
> 1386 extract the entire To header field URI in order to use 
> it as an index
> in the location service.
> 
> It says entire URI in To- header has to be used as index. 
> That too with
> "MUST". Therefore both the addresses will have to be different
> registrations.
> I think this statement in draft bis-05 adds complexity.
> Statement according to previous draft where it didnt give any explicit
> statements on "how TO header has to be taken as key while doing
> registrations".
> This would let the implementer have his own policy as just considering
> "user@domain" which was recommended in bis-04
> 
> regards,
> Shetti
> 
> 
> 
> 
> Shail Bhatnagar <[EMAIL PROTECTED]> on 01/04/2002 10:06:46 AM
> 
> To:   Jonathan Rosenberg <[EMAIL PROTECTED]>
> cc:   Barry Desborough <[EMAIL PROTECTED]>, sip-implementors
>       <[EMAIL PROTECTED]> (bcc: Shrinivas 
> Shetti/HSSBLR)
> Subject:  Re: [Sip-implementors] user=phone in registered addresses
> 
> 
> 
> 
> Jonathan, Are you suggesting that following 2 registrations
> should update the same record
> 
> To: <sip:1234@proxy;user=phone>
> To: <sip:1234@proxy>
> 
> Is an implementation free to treat them as same or different ?
> 
> thanks,
> Shail
> 
> Jonathan Rosenberg wrote:
> >
> > Barry Desborough wrote:
> >
> > > Any opinions on this matter?
> > >
> > > Another manufacturer's UA is registering, using 
> user=phone in its To:,
> > > From:
> > > and Contact: headers. My UA is a simple POTS access 
> device. All the
> user
> > > can
> > > do is enter telephone digits. When I try to call from my UA to the
> other
> > > device, omitting the user=phone parameter, the registrar 
> can't resolve
> > > the
> > > URL.
> > >
> > > Now it seems to me that the user=phone parameter is only 
> useful in the
> > > Contact: header - the registration should be done without this
> parameter
> > > in
> > > the To: and From: headers. After all, the registered address is a
> pretty
> > > abstract value - all that really matters is that the registrar can
> > > resolve
> > > the Contact: URL from it. Is this reasoning correct, or 
> am I missing
> > > something?
> >
> > This has been a continuing source of confusion, and is 
> currently logged
> > as issue #281, in fact.
> >
> > Generally, the mapping of an incoming request URI to obtain 
> a key into
> > the DB of registered contact is a matter of local policy, 
> but it needs
> > to be done in a consistent and coherent manner, in order to avoid
> > interop problems. This particular item, use of user=phone, 
> has been a
> > source of trouble for some time.
> >
> > I believe that the correct thing to do in this particular 
> case is that
> > the registrar should not user the user=phone when using the
> > address-of-record from the REGISTER. Furthermore, it ought to be
> > stripped from the incoming r-uri, although it probably 
> should not have
> > been there in the first place if the caller has no idea about the
> > meaning of the numbers being entered. Generally, the 
> user=phone is in
> > there if the number is known to represent a PSTN number, rather than
> > just a string of digits that are of local significance.
> >
> > -Jonathan R.
> >
> > --
> > Jonathan D. Rosenberg, Ph.D.            72 Eagle Rock Avenue
> > Chief Scientist                         First Floor
> > dynamicsoft                             East Hanover, NJ 07936
> > [EMAIL PROTECTED]                 FAX: (973) 952-5050
> > http://www.jdrosen.net                  PH:  (973) 952-5000
> > http://www.dynamicsoft.com
> >
> > _______________________________________________
> > Sip-implementors mailing list
> > [EMAIL PROTECTED]
> > http://lists.cs.columbia.edu/mailman/listinfo/sip-implementors
> _______________________________________________
> Sip-implementors mailing list
> [EMAIL PROTECTED]
> http://lists.cs.columbia.edu/mailman/listinfo/sip-implementors
> 
> 
> 
> 
_______________________________________________
Sip-implementors mailing list
[EMAIL PROTECTED]
http://lists.cs.columbia.edu/mailman/listinfo/sip-implementors

Reply via email to