On Mon, 2006-09-01 at 16:36 -0500, Paul Kyzivat wrote:

> 
> David Stuart wrote:
> > Hi, I have a follow up question to the question below.
> > 
> > Let's say I send a request, which results in a redirection with the
> > following contacts:
> > 
> >      1. [EMAIL PROTECTED];q=1.0
> >      2. [EMAIL PROTECTED];q=2.0
> >      3. [EMAIL PROTECTED];q=3.0
> 
> First, q-values must be in the range [0.0-1.0] so I am going to assume 
> all your q-values have been divided by 10.


Ooops, my mistake. That was a manually typed example.



> According to 3261 you should merge the last one in according to q-value 
> among whatever candidates are left when you get it. (Your 2nd choice.) 
> So the relative order would depend on how many of the others you have 
> completed.
> 
> But this has been discussed and considered problematic. The problem is 
> that q-values are only ordinals, not cardinals, so mixing values from 
> different sources makes no sense. Also, the order then depends on 
> whether proxies recurse or redirect.


Yes, I see.

Thanks, that's very informative. I'll look into RFC3841 to see if it is
any easier to implement that portion.


Dave
-- 
David Stuart, SIPquest
Email: dave (at) sipquest (dot) com
Phone: 254-8886 x234  Web: http://www.sipquest.com/
Address: 300 - 350 Terry Fox Drive, Kanata Ontario, K2K 2P5


_______________________________________________
Sip-implementors mailing list
[email protected]
https://lists.cs.columbia.edu/cucslists/listinfo/sip-implementors

Reply via email to