RE: WG Review: IPv6 Operations (v6ops)

["Hesham Soliman (EAB)" <hesham.soliman@era.ericsson.se>]

Hi Margaret, 

Before we officially disband the ngtrans WG, we will have
to make a decision about what to do with all of its WG
IDs.  There are a number of choice, but the main
options are:

        - The draft is reverted to an individual submission.

                This is the default choice for documents that
                don't find a new home.  Most of the ngtrans
                drafts will probably fall into this category.

                There is a further choice here -- whether to 
                change the name of the draft immediately, or 
                leave it as draft-ietf-ngtrans-*...  New
                revisions of these drafts would need to be
                done under the authors' names, as ngtrans 
                would no longer exist.

        - The draft is moved to v6ops or to another WG.

                In order for this to happen, the document
                would need to fit within the charter of
                another WG, and there would need to be a
                consensus within the WG to accept the 
                document as a work item
.
                I expect that we will officially accept 
                a few of the ngtrans work items (ones that 
                clearly fit into the v6ops WG charter) as 
                v6ops work items in Sunnyvale.

        - The draft is immediately expired.

                I don't think we will do this with any of the
                ngtrans work items, unless we have active 
                documents whose authors would prefer to have
                them expired, rather than reverting them to 
                an individual submission.

I realize that these aren't great choices.  The authors of ngtrans
work items have worked very hard to produce good documents that have
been accepted by the ngtrans WG and reflect WG consensus, and these
choices will represent an official step backwards for many of those 
documents.

=> Understood. In this particular thread I was trying
to make the case for the EGP/IGP-based tunnelling
because I think it's very useful/needed and supported
by all the major router vendors. In this case, I think
that the picture is not too grim because hopefully
they can be moved to existing WGs in the routing area.
I hope that some communication takes place between
the chairs/ADs to see if this is feasible. 

I have some concerns about other drafts, like for 
example ISATAP, which is also a very useful tool and 
does have deployment cases. 
I'm just not sure how concrete one has to be to
"prove" that deployment will need something, when 
the actual deplyment hasn't taken place on a large
scale. To a large extent there will be some speculation,
which is no different to the justifications that 
were behind these drafts when they were wrote in the
first place. If we wait for people to deploy networks
and then discover that they are missing pieces then
we're too late. Let me give you an example, if 
6to4 was written now, it would have suffered the same 
destiny as ISATAP or BGP-based tunnelling. But because
it was written a long time ago, it is actually an
RFC. I'm not sure that 6to4 is a better solution
for inter-domain tunnelling than BGP-based tunnels.
I could be completely wrong´, I hope so....

Hesham