[ppml] Policy Proposal 2004-3 Global Addresses for Private Network Inter -Connectivity

Chip Mefford cpm at well.com
Wed Feb 16 14:45:18 EST 2005


-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hi Randy;

Randy Bush wrote:
|>>this is arin, not some advice to the hacker group.  use of 1918
|>>space is dangerous and can cause nasty surprises to pop up
|>>years later.  see
|>>
|>>    <http://rip.psg.com/~randy/040226.apnic-nats.pdf>.
|>
|>Agreed in principal.
|>
|>But the core issue remains. There are some of us out here
|>who are tiny, and must get by with (-also, another generally bad
|>idea, CIDRs- of) /26 or /28s or less.
|>
|>What choice have we?
|
| somehow you seem to only receive or read the top ten lines of
| messages.  check your mta and mua.

No, believe it or not, I actually do read the whole thing.
But I only reply to the bits I have some issue with, be
that comment, controversy, or whatever.

| i did not propose to remove rfc 1918, forbid use of 1918 address
| space, ...

No, of course not. I'm sorry if I somehow implied that you did.
Wasn't my intention.

| what i said was that we, arin, should not "encourage" it.

It's a bit funny for me to say "we, arin" but I suppose
in some sense, I am too. I have been to meetings, and
I did pay attention.

That said, non rfc-1918 address space (hows that for backwards
thinking?) isn't the easiest thing for us small guys to have allocated
to us. We are dependent on our upstreams, who often in turn are
dependent on their upstreams who love to consider their allocations
as inventory that is saleable. I don't really blame them.

We, ARIN, are completely cool about direct allocation. But I don't
need a /20, and certainly cannot afford one. And even if I could,
it would be a waste. (like a lot of the /8s, and for that matter
a number of the /16s as well).
We, ARIN (I'm starting to like that) may feel that we shouldn't
encourage the use of rfc-1918 for internet-worked networks, but
it is a fact of life, and will remain so.
We, ARIN, already encourage ISPs to use netblocks as
small as /29, (and smaller) and do encourge SWIPing those blocks, all of
which makes sense. However, by recognising netblocks above /24, we are
almost implying those downstream are going to be using rfc-1918
addresses, and as such, perhaps we should be suggesting or at least
actively participating in how those networks can be best utilized
for the good of all.

I'm rambling, and I apologize for that.
I'll go back to lurking now.

Thanks for your time/input.

| randy
|
|

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.0 (GNU/Linux)

iD8DBQFCE6LO0STXFHxUucwRArtVAJ9fhNg36oOnBA/IyCZxhTXQkLaS8wCfcH8a
noZytZ9VlsnYaCqqp+6P2is=
=XGlk
-----END PGP SIGNATURE-----



More information about the ARIN-PPML mailing list