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

[SM <[e-mail]>]

Hi Ted,
At 14:30 19-09-2013, Ted Lemon wrote:
>There won't be a fight—I'm just trying to figure 
>out what the right thing is to do.   The 
>situation is complicated by the fact that as 
>Brian said, the document really does focus on 
>layer 3 and naming, and isn't intended to go any 
>wider than that.   To us, that's what "IPv6" 
>means, but I think we failed to account for the 
>possibility that other readers might think it includes the application layer.

The document discusses about addressability.  As 
Pete pointed out there's an application 
perspective to that.  There are two alternatives 
right now, i.e. what Dave Cridland suggested or 
looking into the organization of the 
document.  There may be more alternatives if there is input from other people.

> From my perspective, I think accounting for the 
> application layer is a good idea—it's just not 
> something that this document sets out to do, 
> and indeed I don't think this document by 
> itself can do that, so that needs to be 
> excluded by what we mean by "IPv6 Home Network 
> Architecture."   As Brian said, we really are 
> talking about the bottom of the stack here.

The bottom of the stack approach is not that 
practical when the parts have to be put 
together.  In my opinion it leads to complexity.

>I would actually encourage you to look at the 
>MIF architecture document (not yet adopted, but 
>should be soon), which is still an intarea 
>document, but really addresses the apps layer 
>problem more directly: 
>https://datatracker.ietf.org/doc/draft-anipko-mif-mpvd-arch/
>
>I say this not only because I think it's a good 
>and interesting document, and not only because I 
>would like your input, but also because I 
>believe it relates pretty closely to what's 
>going on with the homenet architecture document, 
>and I'm realizing that by exposing you to 
>partial snapshots of the larger picture, we are making your work harder.

I thought about MIF when I was reading the 
draft.  The home networking would likely be using 
split views for DNS.  I have to go through the 
DHCPv6 specifications and some of the MIF work 
before I can comment on the MIF architecture 
document.  The work is not that hard. I'll take 
the liberty not to explain why or else I will be 
volunteered for IETF work.  :-)

Best regards,
-sm