[arin-ppml] Taking back UNUSED addresses.

Joe Maimon jmaimon at chl.com
Tue Sep 16 20:42:45 EDT 2008



Berse, Ryan E wrote:
> When looking at these numbers, you need to keep in mind that just
> because a particular block of space is not advertised to the (public)
> Internet does not mean it is unused.  Organizations also use their
> registered address space on B2B links between their own network and
> those of partners/clients/vendors/etc.  I would bet such use accounts
> for a significant chunk of the allocated but unadvertised space.
> Attempting to reclaim such space would probably meet with fierce
> resistance from the registrants (especially if legacy) and their legal
> departments.  
> 
> Ryan
> 


With which exact legal leg to stand on? What contract do they have with 
whom that prevents ARIN from writing down in its book that some other 
organization is now associated with a set of numbers?

Suppose it was my book. Suppose I published a blog where I randomly 
associated ranges of 32 bit numbers to organizations and friends of 
mine. Suppose I even took "donations" to run this blog. What cause of 
action would you have to stop me from doing so with some arbitrary 
subset of those 32 bit numbers?

And suppose ARIN is indeed immune to those whom it does not have a 
contractual relationship. Who else would you go after? The organization 
now trying to use those numbers on this shared network called the internet?

What legal leg does an organization have to stand on when another 
organization claims the use of some set of numbers they feel somehow 
"belong" to them?

Here is a fairly non-malicious example.

Suppose org a has swamp space x/8, they dont advertise it to the world, 
they use it only on b2b links. Suppose org b decides to also use x/8 on 
their b2b links. Suppose org a wants to stop org b from doing so. What 
legal theory would they use?

When a real hijacking of active ip space happens, what legal wheel 
turning occurs then? I havent heard of anything much.

Numbers dont belong to anyone. The use of a set of numbers on a shared 
network is _tracked_ centrally *by consensus*, because of the mutual 
common benefit that brings.

This consensus is lot weaker for non-advertised space, precisely because 
the benefits consensus brings there is some fraction as compared to 
advertised.

Networks operate according to or enter into contract with this central 
tracking _voluntarily_, and most do so with varying amounts of exceptions.

Before too long, things may degenerate to the point that this becomes 
more than armchair wrangling.

Joe



More information about the ARIN-PPML mailing list