[arin-ppml] [arin-council] Returning Legacy Space to IANA. Attorney client communcation.

Jo Rhett jrhett at svcolo.com
Fri Jun 12 13:39:15 EDT 2009

On Jun 11, 2009, at 5:27 PM, Martin Hannigan wrote:

>> On Jun 9, 2009, at 5:49 PM, Martin Hannigan wrote:
>>> Most large service providers in the region don't believe we're going
>>> to get a fair shake in the end. Hopefully, the AC and BoT realize
>>> that.

> On Thu, Jun 11, 2009 at 2:57 AM, Jo Rhett<jrhett at svcolo.com> wrote:
>> Martin, I know you are clueful.    So I'm assuming that this  
>> response can't
>> possibly be what it means on its face.
>> Can you name a large service provider in the ARIN region who  
>> doesn't have
>> miles and miles (on any yardstick of your choice) of extra lenient  
>> space
>> they are paying nothing for?    That the people who have gotten the  
>> most for
>> free are afraid that they might lose some of that advantage isn't
>> surprising.
> You think someone who has and is using a /8 is stressing over the $18K
> yearly fee? Cost is probably the least interesting argument.

I wasn't thinking of cost, so yes I agree.

> I can't define excess space because I don't know the specifics as to
> why space that someone may have isn't being utilized. Wearing
> sunglasses at night blinds you to most everything around you except
> the bright lights and here we have what appears to be excess space as
> that light.

Leaving the direct insult aside, it appears you are trying to say that  
I don't know what the problem they fear is.  Yes, that is true.  Could  
you skip the insult and instead be specific about what these providers  
don't think they are getting a fair shake on?

>> That this could be seen as a Bad Thing surprises me.
> See 2009-3 thread. There were some fairly interesting fairness
> comments that are relevant to the participants in this region.

I have looked in several threaded archives and it's not clear to me  
what the issue is.  Could you explain or provide a specific pointer?

> It's not about advantage until the allocation policies become slanted
> towards percentages of fulfillment. If my competitors can fulfill 100%
> of their needs and I can only fill 10%, that's not fair. I don't know
> how anyone thinks it could be fair. If I'm being obtuse, I apologize.

There are lots of benchmarks for fairness.  I believe the large  
providers have good ground in IPv6 rollout, most specifically that  
they can absolutely prevent the small providers from using IPv6  
effectively until the large providers do.   Since the large providers  
are holding the strings the small providers dance on... let's talk  
about fair?  There's nothing fair there.

And honestly, I am looking forward with pleasure to the day when  
providers with legions of legacy space run out.  Why?  Because I'm  
sick and SICK AND SICK of refusing IP allocation requests that have  
insufficient justification, only to have the customer complain through  
management that I'm not doing my job, and offer as proof that Cogent  
or someone else gave them a /24 just for asking, no justification  
required.  (note that these are never "almost" cases -- this is  
usually someone with a half cabinet and 3 hosts in it, etc)

You want fairness?  Let's make the environment truly fair:  before  
Cogent or any other provider with legacy space can get any more  
allocations, they have to demonstrate 80% usage of ALL of their space,  
including all legacy blocks.  *THAT* would be fair.  But we both know  
that ARIN has no stomach for this, and it won't happen.   Fair?  No.   
Certainly a large provider advantage.

So yeah, Martin -- I'm just not seeing anything at all that really  
hurts large providers.   The only thing I could maybe see is that  
large providers hit the wall first, and frankly even if none of these  
policies pass they're going to hit the wall first anyway.

Now if they went and applied strict ARIN allocation policy to ALL of  
their space, I'll bet they have years of space already in their  

Jo Rhett
senior geek

Silicon Valley Colocation
Support Phone: 408-400-0550

More information about the ARIN-PPML mailing list