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

[David Harrington <[e-mail]>]

Which reinforces the point that the title and intro are misleading if this
is just a temporary "snapshot of current thinking" document.

I personally am not convinced publishing this as an RFC is beneficial -
submitting for RFC publication starts a series of reviews etc. that might
effected by asking for those external reviews directly. Publishing as an
RFC, even an Informational one, tends to cast things into stone to a
degree. And this document doesn't sound ready for that step.

If the document is very clear that it is a temporary snapshot of current
thinking for the WG, maybe publishing as an RFC could be justified, but
I'M clear it meets the bar for RFC publication, even with the requested
changes to title and intro.

My $.02
--
David Harrington
[e-mail]
[phone]





On 9/19/13 5:33 PM, "Pete Resnick" <[e-mail]> wrote:

>Understand that I still haven't done my review of the document, and that
>I was the one who pushed for an APPSDIR review:
>
>On 9/19/13 3:59 PM, Brian E Carpenter wrote:
>> Reading this thread, I perceive big differences in understandings
>> of the scope of the document (and of the WG).
>>
>> IMHO the document (and the WG) are about making the networking layer
>> (layer 3), and routing, work consistently in zero-management home
>> networks. That inevitably spills over into DNS.
>>
>> The document (and the WG) are not about making the application
>> eco-system work consistently. That is completely absent from the WG
>> charter; after all, it's an Internet Area WG.
>>
>> I believe the draft meets the charter goals. It's certainly a snapshot,
>> and should be labelled as such, but it isn't intended to stray much
>> outside layer 3, and shouldn't.
>>
>> Whether work is need in the application eco-system for home networks
>> is a separate discussion.
>>    
>
>Let's be fair: The charter talks about tackling requirements in name
>resolution *and* service discovery. Once you get into those areas, but
>especially the latter, you're dealing with the needs of applications.
>Most of the work in this WG requires INT area expertise, but APP issues
>are inevitably going to pop up. I think to say that the draft shouldn't
>stray much outside of layer 3 is a bit naive. But since we're being
>fair: I've tried to keep an eye on things going on in homenet and to
>attend homenet meetings myself, and I've tried to encourage other
>applications folks to do so, and I've failed miserably on both counts.
>In large part, the former has been due to time and scheduling issues,
>and for the latter I clearly was unable to make the case to applications
>folks that this work *did* require cross-area review. What else is new
>in the IETF lately?
>
>It sounds like we have a path forward for this document. But I've always
>been wary about INT groups doing things for applications without serious
>APP area input. (Lists of bad things available upon request.) Blame all
>around, including myself, for not making that situation better.
>
>pr
>
>-- 
>Pete Resnick<http://www.qualcomm.com/~presnick/>
>Qualcomm Technologies, Inc. - [phone]
>
>_______________________________________________
>homenet mailing list
>[e-mail]
>https://www.ietf.org/mailman/listinfo/homenet