Re: [DNA] WG LC comments on draft-ietf-dna-goals-01.txt

[Pekka Nikander <pekka.nikander@nomadiclab.com>]

>>    1. I think we should make clear in the Introduction that
>>       we are here only dealing with IPv6.  That is clear from
>>       the context, i.e. our charter, but it should also be
>>       clearly stated _early_ in the document.  (Or did I just
>>       miss it?)
>
> The name of the draft is 'Detecting Network Attachment in
> IPv6 Goals' and I thought that was explicit enough. :-)
>
> Do you still think we need an additional statement? If so,
> I will add it in the Introduction.

Please do.  The change may be as simple as changing "IP"
on the second line of introduction to "IPv6", just as a
reminder.

If you want to put there more, I would suggest also changing
the second last line of introduction:

>    Detecting Network Attachment (DNA).  DNA schemes should be precise,

s/Attachment (DNA)/Attachment for IPv6 (DNA)/

Or, alternatively, to add to the end of introduction:

    This draft considered the DNA procedure only from IPv6 point
    of view, unless otherwise explicitly mentioned.  Hence, the
    term "IP" is to be understood to denote IPv6 by default.

Or something like that.


[My issue #2 removed as we seem to agree on it.]

>>    3. With regard to goal G10, which currently states that ...
>>
>>>      DNA schemes should take into consideration the case where a host
>>>      is attached to more than one link at the same time.
>>
>>      ... should we add "or multiple times to the same link"?
>
> G10 is about the special case that a host is attached to multiple links
> at the same time and hears different RAs from different links.
>
> I see few reason to add "or multiple times to the same link". I think
> that's a different case and we may obscure the main issue, 'multiple
> attachment', if we add that.

Well, I was thinking about the situation where the host is
connected through several interfaces to the same link.
In some way, this is a special case of a host being connected
through several interfaces to several different links.

I am not sure how to formulate that properly, but IMHO the whole
continuum should be covered by G10.  I don't see it covered
by the other goals.

--Pekka