Hi Scott,

> First, a lot of this is general comments on goal, motivation
> and problem statement.  This should all be put up at the front, as
> context.  I would have the order in front be Introduction, then Problem
> Statement, then "Proposals" with "Abbreviations" being the first
> subsection under that.


I have no problems with that, however, moving up to the front would move it
out from underneath the disclaimers at the head of the section.  I think
that a lot of folks would then have an issue with it.

 
> For each of your justification paragraphs, I think you should have a
> summary bullet, e.g. "must be applicable to IPv6", "must be good for the
> long term".

Would section headings do?
 
>>    These are not
>>    incompatible specifically because we truly intend short-term
>>    improvements to be completely localized and temporary.  As the long-
>>    term solution is rolled out and gains traction, the short-term
>>    improvements should be of less benefit and can subsequently be
>>    withdrawn.
> 
> Ha ha ha.  "Incremental" isn't just about new things coming in, it's
> about old things going out as well.  The text should note that the
> short-term tweaks will be with us for a long time, but that they are
> necessary steps.  When you get down to specifics later, explain why
> having the short-term tweaks hang around isn't a terrible problem.

You can laugh all you like, but I do suspect that the operational pain of
AIS is going to motivate operators to retract it as quickly as possible.
Added text as you suggest.

> Brian's right, renum-needs-work is not a solution, it's a discussion.  I
> would mention it in the "Rationale" as something that helps make ILNP
> more feasible in the long term, but not up here where you're talking
> about "solutions".

Already done.

>>    We recommended ILNP because we find it to be a clean solution for the
>>    architecture.  It separates location from identity in a clear,
>>    straightforward way that is consistent with the remainder of the
>>    Internet architecture and makes both first-class citizens.  Unlike
>>    the many map-and-encap proposals, there are no complications due to
>>    tunneling, indirection, or semantics that shift over the lifetime of
>>    a packets delivery.
> 
> "Clean" etc. doesn't say much -- you should be less subjective.  In what
> ways is it cleaner?  Why specifically are tunneling and indirection
> considered problems?

Heh.  This dovetails nicely into other conversations that I've been having
recently.  Unfortunately, architectural complexity seems to be very
subjective.  Suggestions for text?

 
>>    We recommend further work on automating renumbering because even with
> 
> Here's where you mention renum-needs-work as a focal point for this
> important effort, but skip the last sentence -- it's unnecessary since
> you are no longer presenting it as a solution.

? Brian asked me to add it, now you want me to take it out.

> Finally, I recommend a new section: A bulleted list of issues the IETF
> needs to consider in order to turn this recommendation into real
> engineering for real networks.  I think that would help the IETF figure
> out what to do with this recommendation.  I'll try to contribute to that
> next but as you have seen I don't have a lot of time these days.

I welcome the text.  We should be careful to suggest work that should be
done and not how it should be done.

> Oh, and I appreciate all the work you're doing.

Thanks.  Glad someone does.  ;-)

Tony


_______________________________________________
rrg mailing list
[email protected]
http://www.irtf.org/mailman/listinfo/rrg

Reply via email to