I have no problem with the general idea of this draft (reserving
an address block to use in documentation, examples, etc), but the
address block chosen is the wrong one.
If an address block is to be picked for this, then it should be one
that is, on the face if it, not a valid address block, not just
a piece carved out of the regular address space, and which looks
just like everyone else's address (even if an expert can tell that
it is one that shouldn't be used).
I'd have thought the example learned from Sun's use of 192.9.200
as its documentation example would have made that clear - that was
equivalent to a reserved address (reserved by Sun, as one of their
addresses, rather than by IANA, but that's an irrelevant difference).
192.9.200 looks like a valid address, so lots of the world simply
copied it to their nets. I suspect there are probably still large
numbers of nets numbered from that /24.
I don't have a particular alternative in mind, but just to throw
out a suggestion, something like 0000:DEAD::/48 (or /32 if needed)
would be a good example prefix to use. It is already in the reserved
space, so doesn't need to be treated specially, though it should be
documented as being available for use in examples, and only in examples.
Or perhaps even 0000:FFFF:DEAD::/48 if you want to put something closer
to the end of an address "block". Or 00FF:DEAD::/48 - you get the picture.
kre
--------------------------------------------------------------------
IETF IPng Working Group Mailing List
IPng Home Page: http://playground.sun.com/ipng
FTP archive: ftp://playground.sun.com/pub/ipng
Direct all administrative requests to [EMAIL PROTECTED]
--------------------------------------------------------------------
