[address-policy-wg] Transferring InterNIC space from ARIN Database to other RIR datab ases

On Wednesday, Jul 30, 2003, at 22:42 Europe/Dublin, Hans Petter Holen 
wrote:

> Hmm, we did form a special task force for this matter to work with the
> RIPE NCC with the process etc. (References in Leos message)
>
> As chair of this working group I would really like to know what more we
> can do to interact more closesly with the community - please feel free 
> to
> suggest to the list or approach me personally with suggestions !

Hans Petter,

Thanks for your encouraging words.  I've been away for a couple of weeks
and am still mining the e-mail mountain.  During second half of July, 
I've
had some off-list exchange with Leo, and found him really eager to be 
helpful.
I still have some sense of the "blind man and elephant" problem, in both
directions.

As I'm not involved in a LIR, I missed the formation of the task force.
In addition, I couldn't travel to Barcelona.  In any case, a TF 
involving
LIR-WG and RIPE-NCC is likely (as seems FWIS to be actually the case) to
focus on managing the production line rather than on managing customer
expectations. It's worth bearing in mind that ERX customers are likely,
for the same historical reasons by which they have become ERX customers,
to be involved directly with the RIR rather than via their ISP as LIR.

[ Hmm, the emergence of the Address Policy WG
   seems to be just what I needed. ]

It seems to me that a useful view of the ERX project and its progress
would be a set of graphs derived from a table like the one below; 
needless
to say, my numbers can't be related to reality except by co-incidence.

Reporting Period (Month or Qtr)	2003.01	2003.02 ...
Estimated work (contact count)	   8000    8251 ... (estimate may evolve)
Estimated work (network count)	    215	    216 ...
Completed work (contact count)	      0	     57 ...
Completed work (network count)	      0	     16 ...

The graphic would show clearly, for two different measures of the work
involved, both the estimated and (cumulative) completed work at each
reporting instant. The gap between the estimated and completed work
would show how much was left to be done.  Hopefully, the curves would
converge!

I haven't been able to find such a simple overview in the material I've
located on the web, or in the nice Member Update which Leo has kindly
had sent to me.  I think presentation of this kind of information would
help customers anticipate ERX progress in the foreseeable, rather than
indefinite, future.

It would also be useful to know in what order the different /8 blocks
are to be dealt with, what criteria have led to choosing this order,
and how soon each /8 is expected to be finished, at least as planned
at present.  I expect that the plan may evolve in the light of
experience, and that some blocks of space will prove more troublesome
than others.  There may be scope for taking advantage of the usual
80/20 result/effort ratio.

I expect that the information I would like to see presented must
already be in use to manage the project, and so should not require
significant extra effort to prepare.

It's probably a good idea for me to say that I'm just looking for
information, not trying to paint anyone into a corner.

[ As you can probably guess,
   I'm really only interested in 137.43/16,
   8-) ]

I hope these suggestions help.

I expect to come to Amsterdam for the first week in September,
and look forward to seeing some nice graphs!

Best regards,

Niall O'Reilly

PS. FWIS: From where I sit.