RE: "*" and "#" signs in the userinfo of SIP URI

Hi, 

>Okay, I understand, but then are you opposed to
recommending the IESG
approve the draft as is?

I think that we shall only have one mechanism of
representing # and *.

But, from what I understood you are proposing to remove the
E and F
part, right, and use # and * instead?

I don't think your draft needs to talk about what has to be
escapted
etc, because that is handled by 3261.

Regards,

Christer


> > -----Original Message-----
> > From: Christer Holmberg (JO/LMF) 
> > [mailto:christer.holmbergericsson.com]
> > Sent: Friday, March 09, 2007 7:48 AM
> > To: Brian Rosen; Henning Schulzrinne; Paul
Kyzivat
> > Cc: iptelietf.org; Juha Heinanen
> > Subject: RE: [Iptel] "*" and
"#" signs in the userinfo of SIP URI
> > 
> > 
> > Hi,
> > 
> > >>Is question 2) really a show-stopper for
your draft? I 
> assume it is 
> > >>based on the SIP-URI syntax, and that is
what need to be 
> changed if 
> > >>we want to allow un-escapted #.
> > >Well, I am opposed to changing SIP-URI syntax
without an extensive 
> > >discussion in sip to make sure we understand
the consequences.  I 
> > >think they are significant, but I don't know.
> > 
> > You should not change any syntax in your draft in
any case 
> - because 
> > the issue is generic.
> > 
> > Regards,
> > 
> > Christer
> 
> 

_______________________________________________
Iptel mailing list
Iptelietf.org
https://
www1.ietf.org/mailman/listinfo/iptel

On Mar 9, 2007, at 9:01 AM, Christer Holmberg ((JO/LMF))
wrote:

> I don't think your draft needs to talk about what has
to be escapted
> etc, because that is handled by 3261.
>

I think a reasonable mechanism is to focus on the
"input" character  
set and encoding, before URL encoding. The latter is largely
a  
mechanical process that's invisible to the user, except when
writing  
URLs on napkins. If we later (3261bis) change our minds and
say that  
# does not need to be escaped, "old" (escaped)
URLs will still work  
and have the same meaning.

Personally, I think using E and F is a hack that will likely
lead to  
mistakes and confusion since nobody will be able to remember
whether  
E represents # or * (or that they are at all connected).

My suggestion: Briefly describe the issues surrounding # and
say that  
it needs to be %-encoded. Done. (There doesn't seem to be a
need to  
encode *.)

Henning

_______________________________________________
Iptel mailing list
Iptelietf.org
https://
www1.ietf.org/mailman/listinfo/iptel

On Mar 9, 2007, at 9:06 AM, Henning Schulzrinne wrote:

>
> On Mar 9, 2007, at 9:01 AM, Christer Holmberg
((JO/LMF)) wrote:
>
>> I don't think your draft needs to talk about what
has to be escapted
>> etc, because that is handled by 3261.
>>
>
> I think a reasonable mechanism is to focus on the
"input" character  
> set and encoding, before URL encoding. The latter is
largely a  
> mechanical process that's invisible to the user, except
when  
> writing URLs on napkins. If we later (3261bis) change
our minds and  
> say that # does not need to be escaped, "old"
(escaped) URLs will  
> still work and have the same meaning.
>
> Personally, I think using E and F is a hack that will
likely lead  
> to mistakes and confusion since nobody will be able to
remember  
> whether E represents # or * (or that they are at all
connected).
>
> My suggestion: Briefly describe the issues surrounding
# and say  
> that it needs to be %-encoded. Done. (There doesn't
seem to be a  
> need to encode *.)
>
> Henning
>
>
I agree with Henning.
DaveO.

> _______________________________________________
> Iptel mailing list
> Iptelietf.org
> https://
www1.ietf.org/mailman/listinfo/iptel

_______________________________________________
Iptel mailing list
Iptelietf.org
https://
www1.ietf.org/mailman/listinfo/iptel