Re: [Seamoby] comments on the paging protocol assessment draft

["James Kempf" <[e-mail]>]

Yoshi,

> > But do you think it is realistic to expect the paging team to rate
> > the design based on this single sentence when the name of the
protocol
> > implies a connection between the subnet and paging area, and the
> > rest of the design is described in terms of having a paging agent
(which
> > controls the paging area) in the last hop subnet?
>
> We also have the following desription:
>
> 6.4.3.  Other States in Paging Agent
>
>    The following additional information MAY be maintained.
>
>    o A list of Access Routers through which Paging messages are
carried
>      on traffic channel.
>
>    o Any L2 dependent information needed for performing L2 paging.
The
>      L2 dependent information MAY include the list of Access Points
that
>      are not able to receive and process Paging messages at L3.  Note
that
>      how to page Hosts through those Access Points depends on L2 and
out of
>      the scope of this document.
>
>      /* Authors' note: if an Access Point is able to receive and
process
>      Paging messages at L3, then it can be a separated Paging Agent
and
>      SHOULD NOT be included in the above list. */
>
> I think this is enough to indicate independency between paging area
> and subnets.
>

I agree that the first bullet implies that there may be multiple
subnets per paging agent, and thus per paging area,
but to me this looks like an afterthought
and not a fundamental feature of the design. I expect that was
also the case for the person who performed the requirements
comparison on the document. It was difficult enough for
the assessment team to wade through the details of
each protocol in the alloted time. I don't think one can expect that
they should infer that a requirement is satisfied based on
a few hints here and there. If a requirement is intended
to be satisfied, then it should be clearly stated how. At
best, this bullet and the previous statement would
rate as a rather deep underspecification on this
requirement.

> I misunderstood the meaning of multiple dormant modes such that it
> would mean multiple dormancy states, e.g., Active/Sniff/Hold/Park
> modes in the case of Bluetooth.  Also, I had been thinking that
> time-slotted paging at L3 only relates to the following requirement,
> and that it is weaker than other requirements since it uses SHOULD NOT
> instead of MUST NOT.
>
> 4.13.   Future L3 Paging Support
>
>    Recognizing that future dormant mode and wireless link protocols
may
>    be designed that more efficiently utilize IP, the IP paging
protocol
>    SHOULD NOT require L2 support for paging.
>
> I guess this requirement seems to have been weighted higher during the
> judgement process.
>

It was not a first priority requirement, but it was deemed important.
As I said in SLC, we went through a process of normalization
on our comparison criteria, and some proposals ranked higher
on certain requirements than others.

Would you find it satisfactory if, as Jari suggested, some text
is inserted into the assessment draft explaining how the final
decision was made?

            jak


_______________________________________________
Seamoby mailing list
[e-mail]
https://www1.ietf.org/mailman/listinfo/seamoby