Re: Comments on draft-nielsen-v6ops-3GPP-zeroconf-goals-00

[Tim Chown <tjc@ecs.soton.ac.uk>]

Hi,

I can see the issue with 3gpp being in urgent need of a solution, but then
we could also say that (as far as I have read) that community has already
adopted isatap.   So the requirements are kind of being written up in
retrospect.

My view would be that if isatap meets the 3gpp zct requirements, fine, but 
we could still have the requirements in one document, since the requirements
should be definable for all four scenarios in one doucment (general 
assumptions, list of all requirements, then one section per scenario stating 
which requirements MUST or SHOULD be met for each scenario).

Currently we seem to have a 3gpp text with superfulous included requirements,
and a general zct text that includes 3gpp that - if we keep two texts -
should remove the 3gpp parts to avoid duplication (and probabe inconsistencies).

I think the chances of a single solution are now slim, if 3gpp runs with
isatap(*) (out of immediate need) while the WG develops a more generic "single"
solution applicable to all scenarios.    I think we should accept this fact,
but not let it cause divergence in the recorded requirements texts.

Tim

(*) Though there is a chance that isatap might be the general solution, I
    suspect this is unlikely.

On Mon, Nov 08, 2004 at 02:46:37AM +0100, Karen E. Nielsen (AH/LMD) wrote:
> Hi Tim,
> 
> The 3gpp draft could have been a separate chapter of 
> draft-suryanarayanan-v6ops-zeroconf-reqs-00.
> 
> I think it has been explained in length on the list why it isn't, but let
> me do it once more:
> 
> "Procedural":
> There's a certain urgency to solve the 3gpp case and as the 3gpp
> requirements, spelled out in the previous version of
> this document, was thought to be in a mature state it was judged
> best to keep this separate to avoid 
> the delay and the "blur" that could arise from integrating this into
> the generic zeroconf document.
> 
> As you point out in a different mail - then it is difficult to keep track of
> the generic requirements from the environment specific ones in the generic 
> zeroconf document.
> 
> Technical:
> The constrained conditions of 3gpp: bandwidth, round trips times (and costs)
> singles out the 3gpp environment compared to the other cases considered. 
> 
> These conditions may have implications that are in conflict with some of the 
> requirements for advanced features of the other scenarios. 
> 
> Something else:
> Having said this, it has been pointed out by many that given that the 3gpp document
> is indeed about 3gpp only, some "generic" zeroconf text should be removed
> and furthermore more text should be added on the explicit 3gpp deployment scenario.
> 
> BR, Karen
> 
> > -----Original Message-----
> > From: owner-v6ops@ops.ietf.org [mailto:owner-v6ops@ops.ietf.org]On
> > Behalf Of Tim Chown
> > Sent: Monday, November 08, 2004 1:49 AM
> > To: v6ops@ops.ietf.org
> > Subject: Comments on draft-nielsen-v6ops-3GPP-zeroconf-goals-00
> > 
> > 
> > Hi,
> > 
> > I started reading draft-nielsen-v6ops-3GPP-zeroconf-goals-00, but I'm
> > not clear why we need this text rather than just reading
> > draft-suryanarayanan-v6ops-zeroconf-reqs-00 for the 3GPP part?
> > 
> > (In theory, draft-suryanarayanan-v6ops-zeroconf-reqs-00 could/should
> > list all the requirements, and each of the 4 scenario sections - 3gpp,
> > isp, unman, enterprise - should point at which requirements 
> > they carry?)
> > 
> > Will we thus see more simialr docs for unman, ent and isp?  
> > 
> > I'm confused :)
> > 
> > Tim
> > 

-- 
Tim

North American IPv6 Task Force Technologist Seminar
More info at http://www.ipv6seminar.com/