Re: draft-hain-templin-ipv6-localcomm-03.txt

[Fred Templin <ftemplin@iprg.nokia.com>]

Erik Nordmark wrote:

> > If you are still seeing the document as advocating a particular solution
> > alternative, please send specific text change suggestions as that is not
> > the document's intention.
> >    
> Fred,
>
> The question in my mind is not what one would change in the
> document but instead what parts one could retain.
> The problem is that the document was originally written as an argument
> for why local addressing in general, and site-local addressing
> in particular, was required.
The above statement is false. The document in question came about
by combining two earlier works of the co-authors. The earlier works
are still in the I-D repository - see for yourself:

http://www.ietf.org/internet-drafts/draft-hain-ipv6-sitelocal-01.txt
http://www.ietf.org/internet-drafts/draft-templin-lsareqts-00.txt

> Doing incremental edits to change the document to have a completely different
> perspective isn't likely to be efficient. Better to start over with a blank
> slate and import the part of the existing document that make sense with that
> new perspective. (Just to make it clear, I think the document includes
> many perceived important aspects desirable properties for active sites, but
> the history of the documents colors things in ways that makes the document
> confusing and not very succinct.)
>
> Another high-order bit which I perhaps didn't make sufficiently explicit in
> my note is that the document is completely silent on the desirable
> properties of global communication for these active sites.
> I think we can assume that the importance of global communication
> is greater than zero; those sites which don't need any global communication
> aren't part of the Internet even though they use IP technology (and IPv4
> is fine for them as long as long as they have only a few hundered million nodes
> or so :-)
Global communications are out of scope for the document in question;
a quick read of the document title makes this abundantly clear. Of course,
support for global communications is vitally important, but that is a topic
for another document.

> Thus I think it makes sense to describe the desirable properties for global
> communication as well even though different individual sites will place
> different relative importance on local and global communication;
> in some cases local will be much more important and in other cases
> global might be the more important one.
If you want a document on global communications goals for sites,
by all means go ahead and write one. We already have a document
on local communications goals for sites.

Fred
ftemplin@iprg.nokia.com


--------------------------------------------------------------------
IETF IPv6 working group mailing list
ipv6@ietf.org
Administrative Requests: https://www1.ietf.org/mailman/listinfo/ipv6
--------------------------------------------------------------------