Re: [homenet] APPSDIR review of draft-ietf-homenet-arch-10

[Ted Lemon <[e-mail]>]

Dave Cridland and I had a bit of a side conversation about this, which started out with me being a bit shirty, but eventually produced some useful output (I think) which I have included below (with Dave's permission):

On Sep 19, 2013, at 10:40 AM, Dave Cridland <[e-mail]> wrote:

> Ted Lemon wrote:
>> On Sep 19, 2013, at 10:06 AM, Dave Cridland <[e-mail]> wrote:
>>> I reiterate for you: If the IETF publishes this document as it stands,     it will look like, and be taken as, the official "Home Networking     Architecture for IPv6". Change the title (at least) to avoid this. I even     proposed alternate title texts. How much more constructive would you     like?
>> 
>> That is the goal of the document.   And that is what the document is.   The problem with your suggestions is that they propose that we give the document at title that does not describe what it is.   Perhaps a better title would be "Preliminary Home Network Architecture for IPv6."
> 
> That would work as well, I'm fine with that.
> 
> A note saying that at the time of publication, this topic was under active work by the homenet WG would be good too, but IETF tradition seems to be to avoid such things.
> 
>> But my criticism of your criticism about publishing the document stands. The point of the document is for the IETF to say "this is what we are trying to do."   It's entirely appropriate for this information to be available to the IETF as a whole, and not just to the homenet working group.   AFAIK the homenet working group has IETF consensus to do this work, and there is no competing work.   So it would IMHO be wrong for the working group to propose an architecture and not publish it—to do so would be essentially to shortcut the IETF consensus process.
> 
> There's been a few cases over the years where a document has been published knowing full well it's not complete, or that there would be a replacement within the lifetime of the working group, I know. But all the ones I can immediately think of were cases where the document could clearly provide value to people outside the IETF, and the document stood on its own.
> 
> This document is, to my eyes, borderline for this. In point of fact, you don't seem to be arguing it should be thrust in front of the noses of every TP-Link and so on; you talk above about making it available to "the IETF as a whole". I appreciate an RFC (and the publication process surrounding it) acts as a useful mechanism for internal publicity and awareness, of course, but I'd be more comfortable if this were handled by an explicit note to [e-mail] asking for cross-area review.
> 
> However, I'm certainly willing to stand in the rough here.
> 
> What I do suspect is that a number of external parties, including most particularly consumer-grade ISPs, may well read this with interest, and I think that kind of audience could use a few more hints (and less hints to the contrary) of the actual status of the document.
> 
> Sent with [inky: <http://inky.com?kme=signature>]

Hm, okay, now I think we're getting somewhere.   I think the reason the IETF should publish this is that there is no clear "outside" to the IETF—everybody is welcome.   So part of the goal of publishing it ought to be to put a stake in the sand that CPE vendors will pay attention to, and possibly criticize.   That won't happen with a working group document.   But I think you are right that it would help to be clearer about that.   Do you mind if I forward this discussion (that is, the contents of this particular email message) back to the original distribution?   I'm the one who took it private, but I don't want to presume.