Re: initial issues

[Ran Atkinson <rja@extremenetworks.com>]

At 17:58 13/02/01, Ben Black wrote:
>  If I come to see
>that GSE-like approaches are indeed not up to the task at hand, I too will
>argue against them.  What I would hate to see, however, is a refusal to
>investigate similar proposals to GSE or even to other, rejected solutions
>simply because they were shot down years ago.

        IMHO, folks interested in the GSE Architecture advancing
need to undertake some work to persuade those who are open-minded
but currently sceptical:

1) Closely examine the GSE Analysis I-D, then expand and clarify
the original GSE Architecture document from several years
ago to specifically address the questions and issues (including
any misunderstandings) that are raised in the GSE Analysis I-D.
The GSE Analysis I-D is quite constructive if viewed as input
on what parts of the previous GSE documentation needs more
editing.

2) Those same folks need to create one or more GSE Engineering
documents (e.g. precisely how to modify TCP, UDP, & SCTP; 
precise specs for the other protocol changes that are needed).

3) At least one instance of running code, while not strictly
required by process, would go A Long Way towards helping the
sceptics understand that GSE was viable to code up and make
work in real networks.

I'd suggest that such interested GSE folks undertake this not
on this particular mailing list, but perhaps as a Design Team,
returning here AFTER at least both Architecture and Engineering
document revisions exist that could be subjected to broader review.

Just 2 cents from the right coast,

Ran
rja@extremenetworks.com