Re: [alto] General comment about draft-penno-alto-protocol-03.txt

[stefano previdi <sprevidi at cisco.com>]

Hi Martin,

<sorry for being late>

On Aug 7, 2009, at 4:33 PM, Martin Stiemerling wrote:
> Hi Stefano,
>
>
> > -----Original Message-----
> > From: stefano previdi [mailto:sprevidi at cisco.com]
> > Sent: Tuesday, July 28, 2009 2:43 PM
> > To: Martin Stiemerling
> > Cc: alto
> > Subject: Re: [alto] General comment about draft-penno-alto-protocol-
> > 03.txt
> >
> >
> > On Jul 28, 2009, at 2:37 PM, Martin Stiemerling wrote:
> >
> > > Stefano,
> > >
> > > > On Jul 28, 2009, at 2:01 PM, Martin Stiemerling wrote:
> > > > > I've just learned in the session that there is actually a loss of
> > > > > functionality in the merger between Proxidor & P4P: the sorting
> > > > > server-based oracle is just gone.
> > > >
> > > > Proxidor capability of ranking/rating paths is still there and
> > > > functionality
> > > > is entirely preserved.
> > > >
> > > > See section 4 of the draft for details.
> > >
> > > hmm, which part of section 4?
> > >
> > > I assume that still the sorting feature *at the server*
> >
> > I'm not sure I understand  "at the server"...
> >
> > if you refer to the case where:
> > - client send list of addresses
> > - server rank/rate/sort them
> > - server replies with an ordered list
> >
> > then, this is what we do in section 4.
>
> Section 4 is about path rating but not about "rating at the server"  
> as you've described above and I was also referring to. Or I cannot  
> find the part where it is defined.

As Obi already explained on a separate email, the method for computing
path rating is not specified and can be either on-the-fly or through
pre-computation. We didn't put any detail on the computation itself as
we believe it is not in the scope of the protocol specification.

thanks.

s.

>  Martin