Re: IDPR as a Proposed Standard

[Noel Chiappa <[e-mail]>]

	Yakov:


> As you correctly pointed out, the new version is different from the previous
> one. I don't think that comparing the table of contents is that indicative of
> the differences, but just the difference sizes (as you pointed out 7 K
> characters) should give you a hint about the possible scope of differences.

	Since you think that looking through the table of contents isn't good
enough, I looked through a substantial amount of the two versions in two
window mode. (I gave up before doing the whole thing since I was glazing over
and lacked the time.)
	Since you don't explicitly point out major differences, I assume
that either a) you didn't bother to do this same comparison yourself, or
b) there aren't any. Once again, this is a guess, and if it's wrong I'd
like to hear otherwise.
	From my audit of a substantial amount of the text, it appeared to me
that the bulk of the changes were either a) editorial, to make the text
clearer, or b) changes in structure to make clearer the characteristics and
reasons for the choice between Link State and Destination Vector.
	(The latter is my new expansion of the acronym DV, since you appear to
feel that 'distance vector' maligns EI-N mechanisms, which you call 'path
vector', a nice descriptive term for those mechanisms, actually).


>  1) It was posted as an Internet Draft WITHOUT any prior
>		 review by the working group.
>  2) Specifically, it WAS NOT posted to the WG mailing list prior
>	   to its appearance as an Internet Draft.
>  3) It appears as an Internet Draft with no advance notice to the WG.

	I'm not aware that any of these are necessary. Many of the
WG's in the Internet area make new drafts of documents available to
WG members precisely by posting them to the I-D directory. This is
exaactly what happened here.


>	The issue is that there are several key design issues in IDPR,
>	where choices were made quite a while ago, and the choices
>	turned out to be ill-justified from a technical point of view.

	Now we come to the heart of the issue. I realize that the reason you
may be wasting time raising these pointless procedural issues is that you
are unhappy with some of the basic design choices in IDPR, and that it has
been suggested to you (by me among others) that the time to do that was
some time ago, and that it's too late to really expect to do so now.
	I thus propose to discuss some of these on the IDPR WG mailing list,
even though a) I don't feel that it is necessary to do so, and b) I'm not an
active memebr of the WG, and c) this debate has happened before, most notably
in 1988 when this group was called the Open Routing WG. It was settled at that
time to general satisfaction, although I admit not everyone agreed.
Unfortunately, the mail archives from that period cannot be found (I asked for
them some time ago on another matter).
	Anyone on the IESG or IAB who cares about this matter should
get on that list.

	Noel