ASO General Assembly Meeting - 4 April 2001

The Address Supporting Organization
Address Council Second General Assembly

Second ASO Address Council General Assembly

4 April 2001

San Francisco, USA

Draft Minutes

Meeting Chair: Hans Petter Holen (Chair, Address Council)

1. Meeting opening and welcome

Hans Petter Holen, the chair of the Address Council, opens the second ASO General Assembly meeting and welcoms the attendees.

1. Address Council Operational Update

Hans Petter Holen provided a short report of the ASO and AC operations and pointed out the goals for the current year.

He introduces the new members of the Address Council and gives a short status report of the Emerging RIRs, LacNIC and AfriNIC. He also thanked the APNIC for running the ASO Secretariat in 2000 and points out that this task has been taken over by the RIPE NCC for 2001.

2. Presentation of the regional open policy forums (RIRs)

The RIRs presented the work and processes of the open regional forums. These presentations can be found at:

http://aso.icann.org/meetings/ga-2/presentations/open-policy/
http://aso.icann.org/meetings/ga-2/presentations/apnic/
http://aso.icann.org/meetings/ga-2/presentations/arin/

3. McFadden/Holmes report (Mark McFadden)

Mark McFadden summarises the McFadden/Holmes report.

Mirjam pointed out that the bibliography written by Mark McFadden has been seen as a useful document and and has been available on the RIPE NCC web site since its publication at the following location:

http://www.ripe.net/ripe/wg/lir/Ad_Hoc_Committee_on_Addressing_Ad_Hoc_Effort_Bibliography.html

Questions: Scott Bradner: Mark explained in detail the history of the ad hoc committee. Unfortunately it was set up as a bulletin board, which is a way not many people may have been familiar with.
He acknowledges the work Mark McFadden and Tony Holmes have done with their report.

One reason behind the creation of the ASO was to discuss global policy. There are obviously also regional issues (fees, etc). He pointed out that one has to be careful that the work of the global policy development does not conflict with the the bottom up procedures.
The MoU defines the procedure clearly: policies should first be discussed in the regional bottom-up processes. He does not believe that the ASO is the place to generate the policies, they should be developed through one of the regional processes.

Mark believes that the ad hoc group came out of a disagreement about the success of the bottom up process.

Scott stresses that it is threatening if people who come from outside the IP communities come and tell them how to do things. Suggestions should not be ignored, but there is a different understanding on what the problem is and what the aim of the policies are. Mark quotes Scott: we've got to get used to them and they've got to used to them.

David Huberman cautions the ASO to respect the fact and reality that there are regional differences.

Mark believes that the intersection of their policies is increasing.

Linda Melvin, representing the US GSM Association, mentioned that she attended this ARIN meeting and agrees that the two communities need to get used to each other.

She reports that the groups she represents would like to work together with the IP community and that they would like to work together with the IP community. She further commented that especially after this meeting that it is a very impressive community.

Raimundo Beca, who was a member of the Ad hoc editorial group found the work of the editorial group disappointing. He was often under the impression that there were private discussions and the report showed that this was apparently true. It was concluded earlier that the proper place to discuss global policies is in the open RIR policy forums.

Wilfried Woeber felt frustration and pessimism in people involved in the ad hoc group and asked for feedback on this perception. He feels that it is not as pessimistic as Mark seems to believe. It is certainly appropriate, however, to review the procedures. Both ICANN and the ASO are new kids on the block. One of the side effects of the ad hoc group is the effort to review allocation policies and procedures and to define what is policy and what is procedure. These are often confused. He recommends that everyone take a look at the policy matrix document.

4. Paul Mylotte: Address Space Exhaustion

Scott Marcus wonders if Paul Mylotte has prepared any estimates on how much address space could be reclaimed? How much is allocated, how much is announced, reachable etc.

Paul Mylotte responds that he had not looked into this, but welcomes this feedback.

Scott mentioned that he will not go into the credibility of the market projections of the phone industry. However, if these are accurate projections, what could the RIRs do?

Paul thinks one option would be for the RIRs to make the allocation policies tighter.

Scott points out to Paul that he did not take the 'panic factor' into account. Once people believe that IPv4 addresses will soon be exhausted there will be a big rush which will cause the address to be depleted even faster.

HPH wonders if enough is done to support the deployment of IPv6.

Scott wonders if the RIRs can do anything about the deployment of IPv6.

Mark thinks that one way the RIRs could influence the deployment of v6 is to liberalise the allocation of IPv4 address space. HE pointed out that there are people who seriously suggest this.

5. Scott Marcus: Address Statistics

Scott summarises the presentation given at the ARIN meeting. He then concentrates on the presentation given by Paul Mylotte and on the McFadden/Holmes report. He notes that he receives different results and tries to analyse the reasons. He points out that Paul Mylotte had shown an assumed growth of 30%, 50%, or 80% per year, but that measured growth was less than 10% per year. It might be misleading to show multiple ranges of growth, when the lowest figure is alsready higher than reality. Scott also noted that the shape of the overall historical graph since 1991 does not appear to be exponential. The notion of annual growth at a fixed percentage rate appears to be inconsistent with the known data. In general, the projected rate of IPv4 consumption in the McFadden/Holmes report does not appear to be consistent with the usage rate observed in reality. He added however, that growth in the APNIC region since 1997 has been substantial, and may possibly be exponential (thanks to Takashi Arano for pointing this out).

He further concentrated on the part of the projection that was based solely on marketing data and points out that in most cases the sources of the marketing data were not identified which makes it difficult to validate them.

Scott also disputed the claim made in the report that marketing data is essential to an understanding of consumption trends. Marketing data may be essential to the extent that the underlying trends are _changing_. Where the underlying trends are relatively stable, however, the historical data may be sufficient for purposes of projection (this is the method used by Scott Marcus for his recent projections of AS number depletion). He believes that marketing data will generally contain inaccuracies and wishful thinking, unlike the historical data.

He ended his analysis by noting that aside the specific disagreements about this particular report, he believes that it is healthy for the Internet community to see research and analysis from multiple independent sources. He agrees that RIR allocation data should be available to independent researchers, and he knows that the RIRs are working to make it available within the coming month or two.

Cathy Wittbrodt reminds the audience of a case where a small company thought they would have a lot of customers in the future. In the end this turned out to be true. However, they requested the allocation of a /8 which they still would not have used to date. Marketing figures are always difficult to use, but we do have real data, e.g. the deployment of cable, palm pilots, DSL etc. which we can use for forecasts.

John Brown asked if the data Scott presented is available. Scott noted that the RIRs are in the final stages of cleaning up the data, it will then be publicly accessible on the RIR web sites.

John Brown reported from his experience in the DSL world: there are business level and consumer level customers. In the consumer world addresses are usually dynamically assigned. In the business customer world they use a /30 and NAT.

Others had different experiences and were shocked how fast the address space goes.

Mark McFadden suggested that the RIRs should spend enough resources to produce these statistics. Often those kind of statistics show different things (e.g. Geoff Huston's and the RIRs show different things). This debate needs to go on. When will this community get together again? We need to get the people who work on these kinds of things.

Scott believes that physical meetings are not always necessary. He suggested to use the ARIN RTMA mailing list or to set up a special list for that purpose.

Paul Mylotte clarified that his predictions of 30, 50, 80% are not based on any technlogy, they are pure postulations.

Scott Marcus suggested in that case to take the current trend as lowest curve, otherwise it is misleading.

Tony Holmes noted that they were in big trouble in the telecommunication world, because their forecasts were based on historic data and not looking into the future.

Scott Marcus commented that he was not advocating that policy makers should not take possible future trends into account. different things (e.g. Geoff Huston's and the RIRs show different things). This debate needs to go on. When will this community get together again? We need to get the people who work on these kinds of things.

Scott believes that physical meetings are not always necessary. He suggested to use the ARIN RTMA mailing list or to set up a special list for that purpose.

Paul Mylotte clarified that his predictions of 30, 50, 80% are not based on any technlogy, they are pure postulations.

Scott Marcus suggested in that case to take the current trend as lowest curve, otherwise it is misleading.

Tony Holmes noted that they were in big trouble in the telecommunication world, because their forecasts were based on historic data and not looking into the future.

Scott Marcus commented that he was not advocating that policy makers should not take possible future trends into account.

6. Progress Report on ASO nominees for the ICANN board of Directors (ASO Secretariat)

Mirjam Kuehne in her capacity as representative of the ASO Secretariat gave a progress report on the ASO nominees for the ICANN board of Directors. A Call for nominations was issued on 3 January 2001 and was closed on 5 March 2001. The Address Council will elect one individual before Ken Fockler's term will end in October 2001.

Candidates were invited to present themselves to the audience.

Meeting close.