Date: Fri, 14 Jun 2002 11:39:28 -0400
From: Steve Blake <[EMAIL PROTECTED]>
Message-ID: <[EMAIL PROTECTED]>
| They are out-of-the-box, for those (the majority of users, probably) who
| can't be bothered to configure a NUSLA.
OK, but then all the "problems" that people keep complaining about
remain.
| AS numbers have an existing infrastructure for allocation, and they are
| unique.
Yes, but they're not allocated to all sites normally - I have never
been at a site that had its own AS# I think, even back in the old
days when I was assisting the first version of AARnet, there was no
AS# (for the net which linked all of Aus...) - that's changed of
course, but most end user sites currently don't need an AS#, so
allocating them for this would vastly change the dynamics (and 32 bits
might not be enough).
| The set of nets that would ever care about true uniqueness probably
| intersects heavily with the set that would own an AS# anyway (although the
| first set might be the null-set for all I know).
Possibly, but that assumes that the NUSLA part remains, and that stuff
seems to have been written off previously (for whatever reason).
| Note that draft-hain-ipv6-pi-addr-02.txt addresses are a perfectly suitable
| alternative to my proposal (at least in some cases), but they aren't
| zero-cost to acquire either.
Ah yes, Tony's draft has some very interesting ideas, and is almost
certainly worthy of much more discussion that it seems to have elicited
(for which I apologise to him as one of those who never found the time
to comment, though I have read it as new versions appeared).
Unfortunately, while it might make a nice alternative to provider based
addressing as a global address method (for some) if it can be made to
work (which largely depends upon the routing people managing to support
the address space at a fine enough resolution to make it feasible to use),
it isn't an alternative to site locals at all.
Addresses there are still constrained - I don't get allocated one which
I get to keep forever, no matter what, which is what SL addresses give me
(with or without some higher layer identifier embedded in them).
So, as an alternative to SL, they don't work, regardless of how good
they may be as global addresses. Of course, they also weren't designed
(I believe) as an alternative to SL, so this is probably no surprise.
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]
--------------------------------------------------------------------
