Re: I-D ACTION:draft-ietf-dnsop-dontpublish-unreachable-00.txt

[Philip Hazel <ph10@cus.cam.ac.uk>]

On Mon, 10 Sep 2001, Robert Elz wrote:

> Further, the way it is written would simply encourage cantankerous
> people like me to do exactly what it says I must not do.   What exactly
> is anyone going to do about it?  Tickle me with a feather?   Tell the
> world my servers don't conform to the BCP?   Why would I care?
>
> A much better approach, rather than attempting to create some totally
> unenforceable rules, would be to simply explain the problems, and advise
> people how to avoid them, and why they should.

I appreciate the comments that have been made on this list, as I'm just
an MTA hacker with not much expertise in networking issues.

Randy Bush warned me that if my draft was too "prohibitative" it would
get the backs up of "cantankerous people". I decided to push it anyway,
to test the hypotheses. Thanks, Robert, for proving Randy right!

I have worked through my draft, taking into account the comments made.
What I have done is to further separate the cases of truly private
addresses (RFC 1918 et al) from the much more fuzzy situation concerning
public addresses that are inaccessible.

In the former case, I've retained the MUST NOT, as it seems that there
is support for this position (which in any case just strengthens what
RFC 1918 already says). In the second case, I've pointed out some
examples of legitimate fuzzy public/private boundaries, but tried to say
that if there does happen to be a clear boundary for some service, the
BCP is not to advertise services that cannot be accessed.

Now, I'm new to all this draft management stuff. What is the right thing
to do? Should I just submit the -01 draft right away, or would people
like me to post it (or the substantive parts of it) to the list for
further discussion here before I submit it?

Philip

-- 
Philip Hazel            University of Cambridge Computing Service,
ph10@cus.cam.ac.uk      Cambridge, England. Phone: +44 1223 334714.