Re: [6tisch] removing the 'e'

[Rene Struik <[e-mail]>]

Hi Pascal:

The 802.15.4rev draft is, as the wording suggests, a *draft* 
specification. One should not give the impression as if this would be a 
stable draft, let alone a specification cast in stone. Thus, referencing 
to this as "802.15.4-2015" puts people on the wrong footing. Moreover, 
already anticipating that this draft reaches the end station in 2015, 
even prior to start of sponsor ballot, casts some doubt as to how 
receptive one would be to consider comments received during sponsor ballot.

My only comment right now is that, in its current form, removing the "e" 
is premature. This may change over time, but one should not let 802.15 
folks be the arbiter of this, but users of this specification, including 
6TiSCH.

BTW - ZigBee mostly uses the 802.15.4-2006 specification and most 
6lowpan documents also refer to that specification.

Best regards, Rene

On 3/31/2015 3:07 AM, Pascal Thubert (pthubert) wrote:
> Hello René:
>
> So far, 6TiSCH has not expressed requirements on TSCH. We have taken it as a base and built upon it. So I can hardly relate your words to something specific.
>
> - The base that we have considered for the architecture is very abstract since we had a goal to design beyond 802.15.4 whenever possible. From Pat's presentation, I have trouble believing that this text is affected at all, but then I'm interested in any specific point, and it is still time to fix Archie's text.
>
> - For the more specific draft, a good suggestion could be to ask the Interest Group to review the minimal and 6top interface drafts and assert whether and how they are affected by the change to 802.15.4-2015, so as to promote an action either on the IEEE or on the IETF side. I gathered from Pat that they are not, but a formal review from the I-G could help close that discussion.
>
> Pat, do you think that can be done?
>
> Pascal
>
>
>> -----Original Message-----
>> From: Rene Struik [mailto:[e-mail]]
>> Sent: lundi 30 mars 2015 20:22
>> To: Pascal Thubert (pthubert); [e-mail]
>> Subject: Re: [6tisch] removing the 'e'
>>
>> Hi Pascal:
>>
>> I think the proper approach is that one investigates whether an updated version
>> of a cross-referenced specification is suitable and, if so, one can adopt this.
>>
>> To me, it is not a given that the revision effort of 802.15.4 that is currently on its
>> way will automatically satisfy requirements 6TiSCH has.
>> I am wondering how we would know, since so far technical changes have never
>> been discussed/socialized within 6TiSCH itself (see also
>> http://www.ietf.org/mail-archive/web/6tisch/current/msg02853.html).
>> Anecdotal evidence regarding extensive changes makes me somewhat
>> pessimistic here.
>>
>> We should use a "trust but verify" approach here. Anything else would be
>> imprudent. I think this more or less reiterates Subir Das's point.
>>
>> So, I am not in favor of implicit, blind trust; only in favor of explicit, verified
>> trust. Unfortunately, this means I think the recommendation you put forward
>> below is imprudent.
>>
>> As a final note:
>> a) not sure why the 802.15.4e amendment would be incomplete. This seems to
>> be in direct conflict with the note on page 1 of the 802.15.4e-2012 standard
>> (NOTE-The editing instructions contained in this amendment define how to
>> merge the material contained therein into the existing base standard and its
>> amendments to form the comprehensive standard).
>> b) Not sure how one can already predict now when the current revision work will
>> be end and result in a revised standard. Assuming that this will be in 2015 (with
>> sponsor ballot not having started yet) seems somewhat premature.
>>
>> Rene
>>
>> On 3/30/2015 11:23 AM, Pascal Thubert (pthubert) wrote:
>>> Dear all:
>>>
>>> Pat presented at the WG on Thursday his recommendation to remove the 'e'
>> after IEEE802.15.4 in our current charter and WG Documents (slides at the end
>> of http://www.ietf.org/proceedings/92/slides/slides-92-6tisch-2.pdf ).
>>> The net effect would be that when a new version of the standard is published -
>> 2015 should be available soon -, our normative references will implicitly pick
>> that latest version and inherit the fixes that were made since then, as well as the
>> additional PHYs on which 802.15.4 TSCH can operate.
>>> Pat indicated that this procedure is the expected one when referring to IEEE
>> documents. Subir noted that at the IETF we normally have reference on
>> particular versions, and a change in an RFC is a new RFC thus a new reference,
>> and the IEEE form of reference may create an issue with IETF practice. The
>> chairs will need to validate this before proceeding.
>>> On the question whether the change is desirable, Pat made the case that 15.4e
>> is just an amendment, which is an incomplete reference, and that it yields some
>> incorrectness that is now fixed in the upcoming 2015 version of 802.15.4.
>>> All in all the group agreed that the change is desirable. We are now coming to
>> the list to confirm the consensus. If you have an issue with removing the 'e'
>> please speak up now.
>>> Cheers,
>>>
>>> Pascal and Thomas
>>>
>>> _______________________________________________
>>> 6tisch mailing list
>>> [e-mail]
>>> https://www.ietf.org/mailman/listinfo/6tisch
>>
>> --
>> email: [e-mail] | Skype: rstruik
>> cell: [phone] | US: [phone]


-- 
email: [e-mail] | Skype: rstruik
cell: [phone] | US: [phone]