Re: [Isis-wg] I-D Action:draft-ietf-isis-layer2-05.txt

[David Allan I <[e-mail]>]

Hi Robert: 

I think there is nothing wrong with shipping pre-standard implementations of any protocol.

DA> Wild agreement. There is something wrong with suggesting having shipped it makes it a done deal...

Of course vendor takes a risk here if by going via standards process the spec will change based on the input from WG. Then he has to worry about interoperability issues with his pre-standard implementation.

DA> Exactly, but as near as I can tell based on how this happened and the comments from the proponents of OTV, they are not interested assuming that risk. They simply want the code points. I have a problem with that, at least on standards track.

Remember how tag switching surfaced ?

If you want to go this path I am sure you are very well aware that there is a really long list from any vendor to publish a draft when an implementation is ready.

DA> Sorry, not quite parsing that statement.

I don't want to start any sort of flame war here on that. But I think this was never a showstopper in IETF to have an implementation. Of course authors can not use it as justification of any sort and here I am with you and others.

DA> exactly, there are well understood ways in the IETF to be good citizens and still do proprietary things, and keep them that way for that matter...

So considering both drafts on the table I think draft-hasmit-otv-00 is a very good individual contribution which does in fact has a lot of technical merit.

DA> That's nice, but I don't necessarily see that as relevant to this discussion.

On the draft of draft-ietf-isis-layer2-05.txt I think the best way would be to indeed separate the new encoding into a separate draft, submit as individual contribution (there is still some time before next IETF in Maastricht and ask for agenda item on this or other WG to present.

DA> I presume you are referring to splitting out the OTV components. I of course believe they do not belong in draft-isis-layer-2, but before they are even considered by IS-IS WG in another form I actually think draft-hasmit needs a home, such that the IS-IS extensions are not simply a schema with no defined semantics in some enduring form (cf. simply some expired draft at some point in the future)...The mere existence of an individual submission is not justification for a WG to take on an item...

Would you and other's agree that this could be the best way out from this thread ?

DA> I think I've been clear, there is really two possibilities. One is OTV needs to be adopted as a standards track item, charter change yadda yadda (and this is more than just the IS-IS TLVs), to which L2VPN looks like the most appropriate home, otherwise it is BOF time. Once it is on track to produce a document both IS-IS WG and IANA can reference, it can progress from there. The other alternative is informational track. Those are the choices we all have when we choose to introduce technology to the IETF, it's up to the OTV proponents which path they take...

Anything else seems patently incomplete and a bad idea...or inappropriate...

Rgds
D


> Hi Victor:
> This is not progress and innovation, this is simply */bad behavior/* 
> and as such should not be rewarded....
> Please go away and publish something informational track and request 
> your own IANA code points if you are now so genuinely concerned about 
> the "common good", you obviously weren't when you started shipping....
> D