Christian Huitema wrote:
> I believe you are wrong. On obvious counter-example comes if
> the sender
> has no privilege with the sender's ISP, but the receiver has these
> privileges. In that case, the packet is marked as "priority high,
> traffic type X" by the sender; the sender ISP's rewrites that to
> "priority standard, traffic type X"; the receiver's ISP rewrites to
> "priority high, traffic type X."
>
Assuming that the domain ingress would check the SLA with the eventual next
domain after the domain egress, in addition to the SLA with the immediate
traffic origin, yes. This in the diffserv specs?
Anyway, your example assumes existence of an indication for the traffic
type, that does not exist. If it would, it would open opportunities for some
interesting DoS attacks, wouldn't it?
Another way to a similar scenario would be for the receiver to arrange
partial path reservation from the ISP in question onwards with explicit
signaling. I'd guess GPRS with the radio bearer QoS extended to the GGSN
would suffice as an example. A variation of RSVP could do the same for any
IP terminal. I'd believe this will come up in NSIS, when it gets going. But
this would be flow specific, intserv-like stuff.
> By the way, many people seem to believe that diffserv necessarily ties
> with monetary exchanges. That is not necessary, if for example you
> implement "alternative best effort" and get 2 best effort classes, one
> for "high throughput, low loss", and one for "low delay, possible
> losses."
>
This assumes that the service got from the 2 classes is equally useful (or
that the applications that can benefit from these different classes are
equally useful to the end users). Would not work if the other is clearly
more useful than the other. And even if both classes are equally useful, and
there is actual benefit of the 2 best effort classes, the operator's would
be fool not to try charge extra (maybe an extra $1/mo?) for offering this
choice. Customer retention can also be expressed in monetary terms.
People are willing to pay (more) for the more predictably successful
"dial-in experience" to their ISPs.
At least a year ago the phone operators in U.S. still charged $1/mo for
*tone dialing*... Could save by subscribing to pulse dial only :-)
I think that by definition, if there is differentiation to the user
experience, there needs to be monetary difference somewhere.
Jarno
--------------------------------------------------------------------
IETF IPng Working Group Mailing List
IPng Home Page: http://playground.sun.com/ipng
FTP archive: ftp://playground.sun.com/pub/ipng
Direct all administrative requests to [EMAIL PROTECTED]
--------------------------------------------------------------------
