Folks, after the discussion at in Prague I recirculate on the list a message I sent to Daniel and Rob with y ideas on the funding issue. Greetings to everybody from a much warmer climate than in Prague ;-) Blasco Forwarded message:
From bonito Thu Jan 21 12:23:53 1993 Subject: Re: NCC Funding To: Daniel.Karrenberg@RIPE.NET, k13@nikhef.nl (Rob Blokzijl) Date: Thu, 21 Jan 93 12:23:53 MET Reply-To: In-Reply-To: <9301201639.AA00678@jolly.nis.garr.it>; from "Daniel Karrenberg" at Jan 20, 93 5:18 pm Organization: GARR Network Information Service X-Mailer: ELM [version 2.3 PL11]
Daniel, Rob,
Rob and I have been asked by many people (among them the current funders of the NCC) to come up with ideas for funding the NCC in 1994 and maybe to provide addtitional funding in 1993.
This is related to the strategic decisions which must be made about charging for the registry functions. Below is a *very* rough first draft representing out thinking in this. I circulate it for discussion during the agenda point registry funding at Praha local-ir meeting. Please do not take this as prejudicating anything. It is just some very basic thoughts.
Comments welocme.
I was already stimulated to think about this issue by the NSF announcement. Your (well done) paper has unblocked my natural attitutive against writing. I feel myself involved in the problem as "representing" a service provider as well as RIPE deputy chairman.
I generally agree with what is said in your document about feasible and unfeasible ways of funding the NCC. I think you should spend some effort trying to "model" funding criterias in the overall european networking scene, but only SOME EFFORT. Our major concern should be to exactly define rules to fund NCC, leaving to service providers the task of defining their own rules to charge customer organizations or even final users. It is evident that there are very different requirements and constraints put on a national research network than on a commercial network: the first is supported by the government and is able to give services for free.
I think that:
RIPE Network Management Database has not yet reached the status of a strongly needed service, so it could be a risk to charge service providers on this base. I think we should continue the effort in order to have major backbones (EBONE, EMPB, GIX, ...) making use of the database: it is then a unavoidable tool!
A small uncorrectness:
The community represented in the database is limited to Eu- ropean organisations with Internet connectivity. *stats* -------------------------- not true, it contains also networks with connect: LOCAL
I think that the
European Regional Internet Registry despite it is a young service, it is already a strongly needed service: without unique IP numbers no one can be connected to Internet either now or in the future. By the way the NCC should start some activity as soon as possible about the registration of the successors of 4-byte IP addresses whatever they will be.
I think it is feasible to define ways of finance the NCC which are based on the IP registry
RIPE Support General Coordination should be financed through basic funding mechanisms (i.e. RARE, the CEC trough RARE, the CEC directly): the importance of RIPE has reached a sufficient general consensus, I think.
A note:
The NCC also partecipates in global activities representing RIPE ------------ I think this is a task of the RIPE chairman and deputy chairmans, not of the NCC.
Now about charging mechanisms:
European Internet Service Providers ... Charging the service providers could be achieved in the same way as above through a database registration charge and with the same drawbacks Also the use of the registry service could be billed, with similar difficulties.
The big benefit of funding via the service providers is that the number of entities to bill is relatively small and -even more importantly- there is a chance to come to a consensus about the charging model. On the other hand the wider Euro- pean user community will be funding the NCC services from which they benefit via the providers. So the users having a direct benefit pay, albeit indirectly.
I think that a reasonable and acceptable way to charge should be based on services actually given, so - a service provider can charge a customer organization on the base of the address space, because a larger address space can connect more hosts and can generate more traffic - a registry service can charge a service provider on the base of the effort needed to make and maintain a registration
The funding model I have in mind is the following: 1- any service provider is registered by RIPE-NCC and pays an annual fee (subscription + some advance payment for a foreseen number of actual network registration) 2- any "RIPE registered" service provider can submit registration requests during the year, as well as updates of the database concerning routing privileges. This registration are tipically about connected networks/block of networks. They have a steady benefit from the RIPE registration. 3- any national "non-provider registry" is also registered by RIPE-NCC and pays a much smaller annual fee 4- non-provider registries can tipically submit a number of new "unconnected" network registration. They have a "once-only" benefit from the RIPE registration.
The RIPE-NCC issues an yearly bill a) to service providers computed as follows: inetnum entries with connect: <service providers name> are counted (note this is NOT the number of networks) and then multiplied by a database-management tariff and then discounted by the advance payment already done. b) to non-provider regitries as follows: inetnum entries with country: <non-provider country>, connect: LOCAL and changed: <current year> are counted and multiplied by a registration tariff.
Service providers can then bill, if they need so, their customers on the base of the address space used or any other rule. Non-provider registeries can bill requestors per single request/assignment made or as they like, but I think that it is time to state that an official address cannot be given for free.
All my best, see you in Praha. ---------- ---------- Antonio_Blasco Bonito E-Mail: bonito@nis.garr.it GARR - Network Information Service c=it;a=garr;p=garr;o=nis;s=bonito c/o CNUCE - Istituto del CNR Tel: +39 (50) 593246 Via S. Maria, 36 Telex: 500371 CNUCE I 56126 PISA Italy Fax: +39 (50) 904052 ---------- ----------
-- ---------- ---------- Antonio_Blasco Bonito E-Mail: bonito@nis.garr.it GARR - Network Information Service c=it;a=garr;p=garr;o=nis;s=bonito c/o CNUCE - Istituto del CNR Tel: +39 (50) 593246 Via S. Maria, 36 Telex: 500371 CNUCE I 56126 PISA Italy Fax: +39 (50) 904052 ---------- ----------