Notes and Consensus from Interim Meeting in Hong Kong

Jordi -

Just a quick note -

On 3/15/06 12:56 PM, "JORDI PALET MARTINEZ"
<jordi.paletconsulintel.es>
wrote:

...snip...

> 
> So the question is again, IF we want to have a fair
comparison:
> 1) Technical detailed comparison of different aspects
among both of them,
> pros and cons (i.e., overhead, prefix delegation, DNS,
etc.).
> 

The items given as examples above do not appear to be points
of contention
as recently discussed on the list. I think the technical
aspects of the
solutions are well represented in the interim meeting notes
and references
given by the presenters. Our PS focuses on completeness of
solution
(signaling, encapsulation variants, OAM including MIBs,
failure detection,
etc) and ability for a protocol to solve the H&S
scenarios. The recent
discussion on the list is also not focused on what is in the
PS but, on the
availability of an XP client for host based softwire
initiation/termination
as it is related to TTM.


> 2) How much time it will take for router/concentrator
vendors to implement
> TSP, and if it will happen in case we choose it.
>

This question isn't within our problem statement. Others
have given examples
of TSP and L2TP (supporting V6) router/concentrator support.
We cannot make
any statements about effort, time or other resources
necessary to implement
foo vs bar in any environment we are not familiar. I propose
that we attempt
to use the lists of current known and/or soon to be released
(announced)
implementations.

 
> 3) How much weight we give to the support in routers
vs. the support in
> clients.
> 

This is a good question. We did not address in our interim
meeting. We did
not discuss weighting the proposed solutions for their
ability to solve host
vs "CPE" based softwire initiation/termination
in the H&S scenarios.
Although I believe it has been shown that both proposed
solutions solve and
are deployed in both scenarios.



Thanks

-DWard


_______________________________________________
Softwires mailing list
Softwiresietf.org
http
s://www1.ietf.org/mailman/listinfo/softwires