Re: [6lo] Spencer Dawkins' No Objection on draft-ietf-6lo-paging-dispatch-04: (with COMMENT)

[Spencer Dawkins at IETF <[e-mail]>]

Hi, Pascal,

On Fri, Sep 30, 2016 at 4:36 AM, Pascal Thubert (pthubert) <
[e-mail]> wrote:

> Hello Spencer;
>
> Thanks a bunch for your review. Please see in line;
>
> When responding, please keep the subject line intact and reply to all
> email addresses included in the To and CC lines. (Feel free to cut this
> introductory paragraph, however.)
>
> ----------------------------------------------------------------------
> COMMENT:
> ----------------------------------------------------------------------
>
> I found this text
>
>    A Page (say Page N) is said to be active once the Page N Paging
>    Dispatch is parsed, and as long as no other Paging Dispatch is
>    parsed.
>
> somewhat unclear. Is it saying
>
>    A Page (say Page N) is said to be active once the Page N Paging
>    Dispatch is parsed, and remains active until another Paging
>    Dispatch is parsed.
>
> ?
>
> [Pascal Thubert (pthubert)] Yes, and I like your sentence above better
> than the original. The temporal aspect (your "until") still remains to be
> clarified, as meaning "as the packet headers are being processed from the
> first to the last octet. Do we need to indicated that or is the implicit
> good enough?
>

It's good enough for me :-)


> I wasn't quite sure what "so far" meant in this text (and temporal
> references in RFCs that live forever are somewhat confusing, anyway).
>
>       As a result, there is no need so far for restoring the Page 0
>       parsing context after a context was switched to Page 1, so the
>       value for the Page 0 Paging Dispatch of 11110000 may not actually
>       occur in those packets that adhere to 6LoWPAN specifications
>       available at the time of writing this specification.
>
> Would this be just as correct with "so far" deleted, or am I not
> understanding the point you're making?
>
> [Pascal Thubert (pthubert)] I meant at the time of this publication, there
> is no known standard that has a case where page 0 would need to be restored
> after switching to page one. Does removing the so far express that
> correctly?


I think so.


> Thanks for explaining why you're choosing "Specification Required" as your
> IANA policy.
>
> [Pascal Thubert (pthubert)] The bottom line is that for most of these 6Lo
> networks, there is no equivalent to ethertype. We were already cornered
> with the ITU that started using some escape codes without IETF agreement.
> Now we are opening a very large namespace, we want it to be used by many
> communities beyond IETF, but we also indicate that we wish the IANA to
> manage that namespace like the IEEE does for ethertypes; and we wish that
> non experimental values are registered based on some standard action not
> just anyone asking for one. Now, this question leads to another. Should we
> reserve one page, say page 15, for experimentations?


That sounds reasonable to this outsider.

Do the right thing, of course :-)

Spencer