Re: last minute review of draft-ietf-ipv6-addr-arch-v4-03.txt

[Bob Hinden <[e-mail]>]

Thomas,

At 05:57 PM 05/16/2005, Thomas Narten wrote:
>Bob,
>
> > This raises the question I have had for a while, is when doing updates to
> > existing standard what to put in the IANA section.  It seems to me it
> > should be what changes should be made to the IANA registries, not a
> > complete copy.  This is very different from the technical content of the
> > document.
>
>Yep. And I don't have a one-size-fits all answer. I've seen some
>documents just repeat the old section, but as you note that reads
>funny if IANA has already done the work.
>
>And since this document is obsoleting RFC 3513, it seems odd to point
>back to it for definitive text.
>
>I'm not sure the IESG has thought in a lot of detail about the
>significance of obsoleting a document that contains an IANA
>considerations section.
>
>But, having said that, when trying to find the IANA considerations
>related to a name space, it can be rather interesting sometimes just
>trying to track down the pointers. So, I prefer seeing at least some
>pointers to other documents than nothing at all.

I agree with what you are saying.  The problem here is that even RFC3513 
isn't consistent with what is on the IANA page 
(http://www.iana.org/assignments/ipv6-address-space)  It was, of course, 
changed the RFC3879,  RFC-carpenter-obsolete-1888-01.txt, and 
RFC-huston-ip6-iana-registry-05.txt.  Some of the history for each entry 
can be deduced from the references on the IANA page, but I agree it is not 
the complete history.  It would be nice if the IANA kept a publicly 
accessible log of each change.  That might helpful in the future.

>My wording was an attempt to provide at least the start of a
>reference, while still making it clear that this document isn't
>changing any of the rules that are already in place.

I understand.

> > Instead of what you suggested, I think something like the following would
> > be better:
>
> >     No other IANA IPv6 address registries need to be changed based on
> >     this document.
>
>For me, having a reference to other documents is really the purpose of
>adding something to this section. The above doesn't really add much
>beyond complete silence.

It does at least say nothing else was changed :-)

> > A few reasons for this include the references to 3513 and 3307 doesn't
> > capture the complete history of the current IANA registries and this
> > document will obsolete 3513 and this might make the note confusing.
>
>Another possibility is to include some (but perhaps not  the entire
>table) from RFC 3513, but point out that the new text is being
>included for completeness and doesn't establish any new policy. That
>might be a better the referring to an obsoleted document.

As noted above, the format was changed in 
RFC-huston-ip6-iana-registry-05.txt so that might well cause more 
confusion.  We wouldn't want the IANA to change it back to the old 
format.  :-(
>What I was trying to achieve was a balance between repeating all the
>history (which is sometimes a bit convoluted) with at least a pointer
>on where to get started for the details. But maybe we shouldn't even
>go here... :-(

I agree, there isn't any simple solution.  I am starting to think it best 
to not say anything else about the history here short of trying to capture 
the whole history.  What do you think?

Bob



--------------------------------------------------------------------
IETF IPv6 working group mailing list
[e-mail]
Administrative Requests: https://www1.ietf.org/mailman/listinfo/ipv6
--------------------------------------------------------------------