You are here: Home > Participate > Join a Discussion > Mailman Archives
<<< Chronological >>> Author Index    Subject Index <<< Threads >>>

RE: Berlin ASO meeting minutes

  • To: "'Mirjam Kuehne'" < >
  • From: Jean-Michel Becar < >
  • Date: Fri, 4 Jun 1999 16:15:11 +0200
  • Cc: John M Meredith < >

Dear all,

ETSI is ready to go with a facilator: John Meredith.
So you can keep in touch with him at the following
Wish John a sucessfull mediation. Jean-Michel B�car becar@localhost http://www.etsi.org E.T.S.I. FAS Systems Project Manager Tel: +(33) (0)4 92 94 43 15 Fax: +(33) (0)4 92 38 52 15 > -----Original Message----- > From: Mirjam Kuehne [
] > Sent: Wednesday, June 02, 1999 20:04 > Subject: Berlin ASO meeting minutes > > > > Dear all, > > Please find below the minutes of the open ASO meeting held in Berlin, > Germany last week. > > Mirjam Kuehne > RIPE NCC > > > > Minutes of open ASO meeting 25. May 1999 in Berlin > ************************************************** > > Daniel Karrenberg summarises the current state of the ASO discussion: > The Regional Internet Registries (RIRs) are working on a proposal. > In the meantime a proposal has been submitted to ICANN by CIX, > EuroISPA and eCOM-LAC. Recently a number of open ASO meetings have > been held: > - at 27. January 1999 in Amsterdam in conjunction with the > RIPE Meeting > - at 02. March 1999 in Singapore in conjunction with the ICANN meeting > - at 17. March 1999 in Minneapolis in conjunction with the IETF > - at 12. April 1999 in Atlanta in conjunction with the ARIN > members meeting > > Mark McFadden from CIX mentions the idea of using a facilitator for > the process, this had been discussed at the RIPE Meeting in Vienna. > > Keith Mitchell confirms that, thinks its a good idea, however, no > individual has been identified yet. He summarises the function > description of such a facilitator: > - the person should be reasonably independent > - 50% of the time of one individual is needed > > Daniel adds that this person would devote time to drive the > process, to organise meetings, to draft documents etc. > > Bridget Cosgrave from ETSI offers a staff member for this function. > > Rob Blokzijl stresses that the process is and will be open to anyone, > not just the membership of the RIRs, this is the way the RIRs operate > currently. > > Rob Hall points out that there is a document out there (the proposal > CIX and others submitted), why produce a new one. > > Daniel explains that going into detailed documents from the start is > not a good idea in a process that needs buy-in from many > constituencies, writing legal documents should not be rushed, first > agreement on a set of principles is needed. He further suggests to > then write a document that will describe how the ASO will work > de-facto. Only after this has been agreed, detailed legal documents > should be written. The de-facto operations document should > describe at > least: > - how substantial policies are developed and passed to ICANN > - how this process is managed (for instance by an address council) > - how ICANN board nominations are done > > Rob H. thinks we should move forward, because ICANN exists and the > concept of an ASO is there to be filled. > > Keith stresses that there are much less open/contentious issues in the > address arena, there is no reason to rush. > > Esther Dyson says that ICANN was hoping to get a ASO proposal from the > entire community for the ICANN meeting in August. A proposal that is > on the table then will be on the agenda. > > Bridget asks what the crux actually is in the discussion among the two > groups/communities. > > Rob H. explains that the RIRs think that only RIRs should be on the > ASO, ISP Associations think there are other constituencies that should > be part of it, RIRs should only operate/implement the policies. > > Rob B. clarifies that the registry processes and policy making are > open, no-one has to be a member of whatever organisation. > > Keith points out that there are various possibilities to elect ICANN > board members : > - they could be purely appointees from RIR boards > - or they could be elected from the RIR membership > - or they could be elected from regional open forum participants > - or some combination of the above > > Daniel is not confident in going into such details at this stage; > these details should be sorted out on a regional level. > > In the group there is apparently some misunderstanding regarding the > actual functions of the ASO and address council. Daniel describes the > current structure in Europe: > - RIPE is open forum that makes recommendations and decisions > about policies > - the RIPE NCC is a membership organisation; the members > decide about the > activities and budget of the RIPE NCC > - very close relationship between RIPE and the RIPE NCC > > These constituencies want to be represented. RIPE NCC members and RIPE > participants are worried that they would have to attend different > meetings on different levels to protect their interests. > > Daniel also clarifies that the members of the RIPE NCC, almost all of > them ISPs, tell the RIPE NCC how to spend their money, but there is a > much wider forum who is involved in the policy making (RIPE) > > Bridget asks what the problem is with allowing people that do not want > to be represented by RIPE etc. to be represented by some other > organisation on the ASO. > > It could not be clarified who these other organisations would be. > > John Klensin describes some of the issues related with IP > address space. > There is currently nothing broken in the address allocation process. > There is nothing one could vote on wrt. address allocation policy, > it basically follows laws of physics (e.g. routing, IP > address scarcity). > > The suggestion is made to close the discussion here and to start > working on a proposal together with the facilitator. > > Rob H. repeats that not only the existing 3 RIRs should be on the ASO. > > Daniel stresses again that we should not take steps too quickly, > we must get the principles right first before we worry about > the details. > > John mentions that most likely there will be more than 3 RIRs, and the > existing RIRs support that, they are open to this process. > > Rob H. also doesn't like the idea that people are forced to go through > the RIRs. > > At this point Esther asks if people in the group think that there is a > difference between the American situation and the European and if so, > this should also get on the table. > > A few people in the group apparently think that there is a difference, > it is felt that ARINs processes are not sufficiently open. > > It is mentioned that ARIN was going to change this towards open > policy meetings and in general more open and transparent processes; > this had been decided at ARINs last member meeting. > > Barbara Dooley thinks this has only been proposed, it has not been > decided yet. > > John tries to find out again what is the problem actually is people > are trying to fix. If it is the ARIN problem, the discussions > regarding the ASO are probably not the right place to do this. > > Rob H. thinks that policies should be made on a global level, not > regionally, all experts should come together in one place. > > John is not sure if the policies necessarily need or should be the > same in all regions. They are sufficiently different and there might > be reasons to develop different policies, because there are different > circumstances in the regions. > > Barbara stresses again that ARIN is not open enough. > > Bridget feels that people are trying to importing the ARIN problem to > the ICANN level. > > A gentleman from FR Telekom thinks that RIPE should be > formalised more, > it needs to be a legal organisation in order to have enough power. > > Daniel replies that this question has come up regularly inside RIPE, > and so far the participants seem to be confident with the current > situation. > > Mark says that in other regions it is not working like that; the goal > is to define a global ASO/ICANN with global representation, maybe the > European solution does not work everywhere. > > Daniel responds that he is not suggesting to use the same solutions > in North America, it should be left to each region to define their > processes, and there are movements to open up the processes in > other regions. > We should not create a global structure to fix a regional problem. > > Mark states that also another problem that at first appeared to be an > American problem later turned out to be a global problem (the > creation of > new gTLDs), and the RIPE TLD-WG realised that. > > Keith disagrees. He says, speaking for a ccTLD registry, this was > never seen as a global problem. > > John thinks there is one hypothetical difference between different > regions: the number of very small ISPs related to the total number of > ISPs is very high in North America, therefor one can for instance not > implement the same allocation policies as the RIPE NCC. If this > situation would appear in Europe the RIPE NCC's policies might have > to be adapted. A single international policy would create problems. > > Daniel explains that this is how policy development was handled over > the last years: making sure the polices are coordinated as much as > possible among all regions and defining where the differences are. > A framework has to be set to allow regions to define their own > procedures, for instance wrt. the election of ICANN board members, > this should not be defined on a global level, but should be left to > every region to find the right process. > > Daniel stresses again what the question for (at least the RIPE) > regional constituency is: If there is a global level to it, do I need > to be represented on that as well. If the answer is 'Yes', people will > have a problem with it, because they will have to make sure they are > represented on all levels in order to protect their interests. > > If someone comes and says that he or she does not feel represented, > Daniel would like to ask how one can make that person feel more > represented. To mention the specific issue with the EuroISPA, the > problem is that the RIPE NCC membership overlaps with EuroISPA's > membership, and the RIPE NCC's members are saying that they are happy > to be represented by the RIR process (the RIPE NCC and RIPE > in this case) > when it comes to address space issues. > > Keith stresses that all meetings where these issues are discussed are > open. > > Rob H. thinks that only 'techies' go to these meetings, not the CEOs > of the organisation and they might have different opinions or > concerns. > > Keith believes that this distinction is meaningless. > > Daniel gives an example how the RIPE NCC process work: at one of the > previous RIPE NCC Annual General Meeting some members proposed that > the RIPE NCC should also become a lobbying organisation. However, > consensus was reached that the RIPE NCC should not be doing this, this > should be left to the trade associations. Addressing issues should be > dealt with in a neutral and impartial environment, it should not be > confused by issues where not all members necessarily agree upon. > > Rob H. thinks that CENTER is funded, staffed etc. by the NCC. > > Daniel clarifies that this is not the case: > - the RIPE TLD-WG was created to discuss general TLD issues > - they came to the conclusion that there is some work that needs to be > done by RIPE in an open forum, some work should be done by > the ccTLDs > directly > - ccTLDs decided to establish a separate group to work on these items > - they then asked the RIPE NCC to help facilitating this effort > - the RIPE NCC provided office space and administrative support > - financially the group is independent (the funding is > provided by ccTLDs) > and the project will move out of the RIPE NCC offices shortly > - the constituencies are different, that is the reason why it was > appropriate to form a separate group > > Rob H. was asked what the actual problems wrt. address space are that > need to be addressed urgently and what his problems are to participate > in the regional processes. As an answer he asks the RIRs why they are > so afraid of having a more open process. > > Rob B.: We are open, we cannot be more open than we are! > > Keith thinks that the discussion is trying to bang a > triangular problem > into a circular solution. Various people are not here. It is > difficult to > discuss their issues. The RIPE NCC and RIPE has been working with ARIN > to make their process more open and hopes their members have > been, too. > Isn't this a way forward? > > Mark asks, why not recognise that there are other > possibilities than the > European model. One approach to move forward is to find a facilitator > to find convergence between different points of view. Another > approach is to use a facilitator to start from scratch and to document > the status quo as Daniel suggested. > > Daniel answers that this is not what he meant. He repeats the proposed > process: agree on principles and then concentrate on a de-facto > document; don't worry about legal language right now. The ASO should > have a regional structure. It needs to be further defined what this > exactly means: which details need to be discussed on a global level, > which details can be left to the regions? > > John says that he is personally very reluctant to see a ASO that is > only comprised of RIRs. But because of the technical issues > related to the > address allocation, the RIRs should have a strong majority. > > He also has another concern and that is the direction and the scope of > these constituencies and SOs in ICANN. Is his organisation required to > be part of all these groups to protect its interests? This is not what > he wants! > > Rob B. would be horrified if the ICANN structure would lead to the > fact that many people would not be able to attend the RIPE Meetings > anymore on a regular basis. > > Mark says that in his organisation there are people that fear that > they cannot influence the address policies > > David Randy Conrad explains that also historically the RIRs didn't set > policy in a vacuum, IANA would also listen to other sources, > e.g. the IETF. Decisions were made based on technical > considerations, not > political ones. He would be very worried if this were going to > change. There might be other 'technical advisory councils' to advise > ICANN. The IETF might not necessarily be the appropriate forum in the > future. > > Barbara has not yet heard clearly what the objection would be to allow > regions to define their own mechanisms to participate in the ASO > outside the existing RIR structure. > > Keith thinks the following steps have to be done to move forward: > facilitation has to be used to extract principles and to define where > the parties differ. In addition to that all individual > regions have to go and > solve their individual problems. > > A person from eCOM-LAC reports that his organisation is currently > working together with the academic community in Latin America in order > to form a LatiNIC. He also mentions that he is not happy with the > ARIN services. He believes that commercial organisations (ISPs?) must > have their own constituency in the ICANN/ASO. In the end customers > of ISPs are paying for the next generation of IP. > > Nii Quaynor from Ghana reports that the organisations/ISPs in Africa > seem to be satisfied with the services provided by the RIPE NCC. He is > also not aware of anyone in the other part of Africa that is served by > ARIN that has complained about services provided by ARIN, also they > seem to be happy with the status quo. However, Africa is working > towards its own RIR in order to have more influence. He mentions that > they get a lot of support from the existing RIRs. > > A gentleman from Japan emphasises that it is very difficult to > effectively participate in many meetings, because of travel expenses > and the time spent; it will certainly be difficult to attend even more > meetings. He admits that his organisation is not totally happy with > the current APNIC services, but still wants to be represented by the > RIRs on the ASO. > > John says that ARIN is as anxious to 'get rid of' Africa and Latin > America as those are 'to get out'. These matters will be handled > differently in each region. We should not create mechanisms to solve > problems that might never occur. > > Daniel explains that an AfriNIC is expected to be formed and that > the RIPE NCC supports these efforts. > > David asks if the RIPE NCC would force someone in the region to > go to AfriNIC for services and not to the RIPE NCC anymore. > > Daniel replies that the RIPE NCC would strongly encourage that > organisation and that the NCC would also try to help the AfriNIC to > provide such services that the ISPs in the region would want > to receive > services directly from them. > > Keith thinks that this is typically something that should be discussed > on a global level. > > Keith summarises the discussion: it has been agreed that there should > be some regional structure, there is however no agreement if this > should necessarily be exclusively. There is agreement on the idea to > get a facilitator to work on the further process and to move forward. > ETSIs offer is noted. Keith takes an action to go back to the RIRs > boards to discuss this, it needs to be agreed by all RIRs. > > Because the RIPE/RIPE NCC structure is a working mechanism, someone > suggests a possible way forward: an assignment with a clear deadline > to the RIRs to submit a proposal. This may then only be on probation, > the process will then have to be reviewed in two years time for > instance, because there is the concern that not all stake > holders agree > with this structure. However, the RIR structure is in place and > working right now, therefor the RIRs should be allowed to start. > > Mark notes that this would then not be based on consensus, and that > ICANN is looking for consensus. > > Esther explains that ICANN is not recognising groups but bylaws. > She also does not think that the process needs to be rush at > this point. > > Izumi from APIA says that his organisation has no clear point of > view at this moment. One problem is that the interest of the ISPs in > the Asian region is not as high as it should be, because of the > economic situation, they basically hope that someone will > take care of it. > > Esther stresses that the model should be flexible enough to change, > but it should not purely be a placeholder. > > Barbara seconds Keith' proposal to move forward. > > Keith adds that the proposal needs to be ready for the ICANN meeting > in August; there will be additional open ASO meetings at INET in June > and at the IETF in July. > > Esther confirms that the deadline for submission of a proposal is end > of July, beginning of August. > > Daniel believes that a de-facto document should be ready at that > point. It will be however too soon for details such as bylaws and > articles. > The following issues need to be addressed in the de-facto document: > 1. how is policy developed > 2. how is the process managed > 3. how are ICANN board nominations done > 4. what will be defined at the ASO level, what at the regional level > > Keith and Daniel agreed to update Kim Hubbard from ARIN and > Paul Wilson from the APNIC about the discussion and discuss > the facilitator > idea further among the RIR boards. > > > > >

  • Post To The List:
<<< Chronological >>> Author    Subject <<< Threads >>>