Re: Meeting, Charging & Documents

[Carol Orange <]

Hi Wilfried,

Thanks for your input and for raising some interesting questions. I've
attempted to address them below. 
 
"Wilfried Woeber, UniVie/ACOnet" writes:

>>   Dear all !
>>   
>>   ACOnet has in the past strongly supported the idea of flat fees - mostly for
>>   simplicity reasons, but also because we do not have a "down-stream" food-chain
>>   to pass on any variable expenses. Thus we still propose to continue with Model 1.
>>   
>>   If the current system of self-determination of category is somewhat flawed,
>>   then I'd rely on the input from the NCC to advise on necessary changes. 
>>   We might even want to make that a regular activity of the NCC before the
>>   discussion of the fee strucure or semi-annually, either as part of the regular
>>   reporting or in private to the LIRs. 
>> 
>>   If we feel that there is a need for just a couple of more categories (few!),
>>   then let's try to fix that.
>> 
>>   On top of that, we would be happy to pay an additional fee for things that are
>>   requested only once in a while and can be clearly attributed to a certin project,
>>   like assigning an AS# or having a *fixed* surcharge for an address request
>>   exceeding a certain size (or the assignment window - although that is probably
>>   already too dynamic for small and/or startup registries. Also, requesting
>>   another /16 (or whatever size is appropriate) to carry a one-time charge is
>>   conceivable, because that can be planned avead in due time and can be accounted
>>   for in local budget forecasts.
>>   
>>   We are *strongly* against Model 2 because it involves too much administrative
>>   overhead *on both sides*, but can probably live with Model 3.
>>   
>>   Trying to review Models 2 and 3, I think we need more input on why there would
>>   still be a different basic charge? I'm currently inclined to propose to our
>>   management to down-grade ACOnet's category to small, if Model 2 (or 3) should
>>   be adopted.

This is an interesting issue which that has now been addressed by Daniel in 
a separate mail.

>>   And, given the amount of time I could devote to analyzing the address space
>>   cost table in Model 3, I think I need more input to understand why a chunk of
>>   addresses allocated at the beginning of 1997 would be charged at 548.72 in Q1
>>   compared to a mere 208.95 by Q4 - that's less than 50% ?

The figures in Table 5 of ripe-143 show the quarterly charge for allocated
address space. That means that for a /16 allocated in quarter 1994-1, one
pays 42 ECU, 30 ECU, 22 ECU and 16 ECU in quarter 1997-1, 1997-2, 1997-3,
and 1997-4 respectively. What you see (and quote above) is a drop in charges 
for address space as it ages. This comes about in part because of economy
of scale - the revenue to be earned is generated from an increasing amount
of address space which allows the quarterly charge for address space to 
drop throughout the year. In addition however, we have introduced a sliding 
fee to reflect that the level of work generated by address space also 
decreases with age.

>>   In general, if we propose to go to Model 3, I think we need more input on the
>>   rationale of the figures involved, and why addresses are going to be tagged
>>   with grossly different "values".

The economic justification is stated above. The mathematical details are in
the appendix of ripe-143, but are unfortunately only readable in the
postscript version of that document. 

>>   Thanks for your consideration.

And for yours!

Greetings,

- -- Carol

------- End of Forwarded Message