[arin-ppml] Curious about consensus

Frank Bulk frnkblk at iname.com
Tue Apr 26 23:46:59 EDT 2011

The difference between addresses and vehicles & houses is that the latter
can be seen and touched (as they say, possession is nine-tenths of the law).
The seller could be easily misrepresenting their netblock and really not
have any rights at all.  So the buyer will either do their own due
diligence, require the seller to sign an LRSA, or have a broker manage that


-----Original Message-----
From: arin-ppml-bounces at arin.net [mailto:arin-ppml-bounces at arin.net] On
Behalf Of Ted Mittelstaedt
Sent: Tuesday, April 26, 2011 7:50 PM
To: arin-ppml at arin.net
Subject: Re: [arin-ppml] Curious about consensus

On 4/26/2011 5:01 PM, Leo Bicknell wrote:
> In a message written on Tue, Apr 26, 2011 at 07:26:37PM -0400, Milton L
Mueller wrote:
>>> 2008-2 required a transferor to sign an RSA or LRSA _before_ they could
>>> release the block to ARIN for transfer.  The presumption is that the
>>> receiver wanted a legitimate, recognized by ARIN block and thus the
>>> transferor had incentive to sign an RSA or LRSA first.
>> But this is precisely my point. Everything you say here and below
>> seems to be founded on the assumption that a transferor requires
>> ARIN's pre-approval. Nortel has just blown that assumption out of
>> the water. The fact that it is under bankruptcy court may strengthen
>> its resolve not to be bothered with a transfer agreement involving
>> ARIN, but I see no reason why _any_ legacy holder couldn't do the
>> same thing.
> I think my point is that it is in the buyers hands, as I stated
> before.  It is my _best guess_ that most buyers will want blocks
> recognized by ARIN for a number of reasons, and thus will demand
> transferors follow whatever process ARIN has set up.

I have to disagree, Leo, and agree somewhat with Milton.

Over the years I've bought a number of assortment of vehicles, cars
and motorcycles, with missing titles.  In all cases the seller didn't
give a rat's ass about spending the money to file for lost title and
obtain a title before selling.  They just wanted the vehicle gone and
if they sold me a car for $500 less than what they could have got by
spending the $100 with the DMV to obtain the title and sell it like
they should have, then good for me - but they still didn't give a damn.

All I cared about was the VIN wasn't on the stolen vehicle list.  Once
I got possession I then took care of the paperwork.

These Legacy IP address sales will, I think, mostly go the same way.
The buyer wants the addresses, the seller doesn't give a damn about
the paperwork, they just want cash.  All the buyer is going to do is
look at the current WHOIS listing for the block and if is seems to
match who the seller claims they are, then they are going to get it
before the seller sells it to someone else.

If the seller has 2 buyers, one who will pay only if they get an RSA
in advance, and the other who will pay 20% less but the seller doesn't
have to lift a finger, most sellers are going to go with the buyer
who doesn't require them to do anything.

Geeze, haven't you ever bought or sold a house in a hot market?  When my 
wife and I bought ours the seller waved an as-is clause at me and I 
signed it immediately - thus shutting out 3 other buyers who had been 
looking at the property.  Likely by the time those other 3 ended up
buying their homes, the price in the market jumped another $20K, 4
times more than I had to spend to fix all the little gotchas that
the seller had covered up with the as-is clause.

You buy ANYTHING used and you assume risk.  The buyers in these deal
WANT to assume risk, they are HAPPY to do whatever legal pretzels are
needed to get the IP numbers usable.  The sellers are the losers
here because if they were at all competent they would be using those
IPs already, to make money.

> It is because of this I am most curious about Microsoft's motives
> in this case.  They appear to have paid a huge sum for something
> they could have gotten much cheaper directly from ARIN, and done
> it in a way that may not have gotten ARIN's recognition.  Of course
> the bankruptcy court just wants cash, and doesn't care about any
> of these things.

But they couldn't have got it cheaper because they obviously don't
meet utilization requirements, not now, and not anytime within the
next year.

How much do you want to bet that a year from now most of those Nortel
numbers will still not be in use?

> Bankrupticy courts have sold things that don't exist, just to get
> cash for creditors.  Caveat Emptor, as they say.  Nothing on that
> side disturbs me here, other than I hope we don't end up with a lot
> of companies in bankruptcy!

I don't think most of these transfers are going to involve bankruptcies.

> ARIN has always been voluntary though.  See the thread about the
> "routing police" for some parallels.  ARIN's "authority" rests with
> the fact that most ISP's use the ARIN database to determine if they
> should route blocks.  There's a rogue few who don't, but 98% is
> "good enough" for a functional Internet.  Transfers are no different,
> if buyers (in particular) demand the use of an ARIN process for
> their own protect, the transfer process and whatever rules will be
> followed.  If buyers don't demand it, or a significant number of
> sellers don't participate it will be a failure.  ARIN can't litigate
> 10,000 different sales outside the process.

The buyer has an ongoing incentive to play nice with ARIN.  Most buyers
will just sign the standard RSA and start paying the fees.  Most Legacy 
blocks are gonna be small and ARIN is big and so in any argument ARIN is 
going to get it's way, the fees to litigate will be high enough to
discourage 9,990 of those different sales that want to go outside the
process to stay in the process.

The Nortel MS deal was special because both those companies were large
so it was a battle of the giants in that arena.  ARIN is big and has
money too - but costs to litigate would probably have been in the 
million dollar range.  ARIN made the pragmatic decision that we save a
million and we get a legacy block that wasn't paying us jack crap in
the first place under LRSA, so that the next time those IPs are sold,
that they can't pull our pants down with this "transfer into LRSA"

Many of these Legacy IP number block "sales" may be structured so that
the seller doesn't get a big chunk of money all at once - but instead
the seller gets a little bit of money then gets a residual for the next
100 years or whatever.  That is a lot cheaper for the buyer and it 
protects them from the seller misrepresenting a lot better.


> This is also one data point.  One does not make a trend in any
> direction.  I don't view the Microsoft-Nortel deal as a omen, but
> rather as a first concrete transaction to look closely at and make sure
> what the community wants to happen does happen.
> For the record, I was against all paid transfers from day one.  In my
> time on the AC an via all of the discussions my distaste for them has
> mellowed a bit, but I am still opposed.  I don't think ARIN has gotten
> the policy right in any of the drafts or policies, and the result will
> be more chaos.  Further, I think IPv6 will be deployed quickly (faster
> than most, it would appear from talking to my contemporaries), and thus
> this is all a huge distraction.  Still, if we're going to do them we
> need to use whatever community goodwill ARIN has as leverage towards an
> ARIN process, and I don't think the current policy does that at all.
> _______________________________________________
> You are receiving this message because you are subscribed to
> the ARIN Public Policy Mailing List (ARIN-PPML at arin.net).
> Unsubscribe or manage your mailing list subscription at:
> http://lists.arin.net/mailman/listinfo/arin-ppml
> Please contact info at arin.net if you experience any issues.

You are receiving this message because you are subscribed to
the ARIN Public Policy Mailing List (ARIN-PPML at arin.net).
Unsubscribe or manage your mailing list subscription at:
Please contact info at arin.net if you experience any issues.

More information about the ARIN-PPML mailing list