Re: CAPWAP BOF follow up: nmrg - CAPWAP

["James Kempf" <kempf@docomolabs-usa.com>]

> > Well, I think it is different in that there are vendors stepping up
saying
> > they want interoperability.
>
> funny, that is not actually clear from discussion on the mailing list.
> i would not call a protocol which merely downloads a code image to
> contribute much to real interoperability.  but maybe i am missing
> something.  so i am prodding the list to tease out some protein.  it's
> moving slowly.
>

The code downloading part is the most suspect aspects of LWAPP. Your
comments on the list have been quite helpful. What I tried to do with my
message was be somewhat more clear about areas where a working group could
have a potential for success and not (but unfortunately missed the code
loading part). I did not intent to trigger any suspicions of black
helicopterism, but if there were, I think your clarifications should help to
clear that up. Thanx much for your participation in the list, it goes beyond
what one can reasonably expect of an AD at this stage.

> >> so i won't be able to load balance in a multi-vendor environment?
> > The thought is that the protocol is all that is required for
> > interoperability. In practice, that might not be so.
>
> i am greatly relieved </sarcasm>.
>

:-)

Seriously, though, the 3GPP UTRAN cellular protocols are standardized for
interoperability, but the load balancing algorithms are up to the vendors.
In practice, they achieve good load balancing because the vendors know what
they are doing, and spend a lot of time and money developing their
algorithms. From Bernard's comments, it sounds like some of the 802.11
vendors either don't know what they are doing or aren't spending enough
money on the problem, so an interoperable protocol might not, in practice,
do much to help. Which is what I meant.

> > With my Docomo hat on, I can certainly say that it would be
> > immensely useful for Docomo as a service provider to have an
> > interoperable protocol that would allow this kind of mix and
> > match between APs and ARs.
>
> except i can not actully tease out from the list discussion exactly
> what kind of interoperable protocol that would actually have a
> multi-vendor deployment is being proposed.
>

The discussion doesn't seem to be focussed on a good description of the
problem.

> but this whole discussion needs to move there.  and just the
> technology, not the personalities.
>

Yes, you're right. Unfortunately (or perhaps fortunately) I'm spending most
of my days touring beautiful Tuscan hill villages and my evenings enjoying
Tucsan food, and if I were in my right mind, I would probably not have
looked at email until I got back (and dealt with the spam volume then). But
when I get back, I'll talk with Pat and see whether its possible to tease
out a problem statement.

Also, BTW, I got some email from an EE Times correspondent and spoke with
him on the phone for about a half hour. He had the misconception that IETF
had decided to standardize LWAPP. I quickly dissuaded him of that, and tried
to put him straight about where the discussion is at. I don't like talking
to journalists (especially when I am on vacation) but my suspicions were
that he had some misconceptions and they were confirmed; hopefully, I
haven't managed to make them worse.

            jak