ARIN X Public Policy Meeting
Eugene, Oregon, US
October 30 - 31, 2002
Minutes of the Public Policy Meeting, Day 1
Wednesday Oct. 30, 2002
John Curran, ARIN Board of Trustees Chairman introduced Richard Jimmerson, ARIN Director of Operations.
Richard Jimmerson opened the first day of the ARIN X Public Policy meeting at 9:00 AM (PST) welcoming all those in attendance and making some announcements. The announcements included:
- Richard announced that ARIN President Ray Plzak was representing ARIN at the ICANN meeting in Shanghai, China.
- Thanks to NANOG for their help in this, the first NANOG and ARIN back-to-back meetings. This was done to encourage greater operator input
- Thanks to ARIN X sponsors - University of Oregon, Sprint and Preferred Communications Inc., NW.
- Meeting Room Layout and hours of operation -- ARIN Learning Center and RS Help Desk, Terminal Room
- Thursday evening social, Monster Mash Bash
- 152 ARIN X registrants, representing 27 states and 10 countries
- Over 90 people attending both NANOG and ARIN
- Over 30 first time ARIN attendees
Richard also introduced the ARIN Board of Trustees, the ARIN Advisory Council, and ICANN ASO AC members in attendance and reviewed the meeting agenda. There were approximately 145 people in attendance. The agenda was adopted without change and the meeting proceeded.
RIR Statistics and Activities
Presentation: | PDF | PPT |
Speaker: Leslie Nobile
Leslie Nobile, ARIN Director of Registration Services, presented an update of joint Regional Internet Registry (RIR) activities, efforts and registration statistics.
There were no questions from the floor.
RIR and Emerging RIR Updates
APNIC:
Presentation: | PDF | PPT |Anne Lord, APNIC, presented a report on activities in the APNIC region, including the results of its member survey, help desk statistics, and current and future project status. Details of APNIC 14, which took place in Kitakyushu, Japan, from September 3-6 2002 were also announced.
There were no questions from the floor.
RIPE NCC:
Presentation: | PDF | PPT |Leo Vegoda, RIPE NCC, presented a report on activities in the RIPE NCC region, including updates on the development of a secure member-only web portal, the addition of PGP signatures to all outgoing hostmaster e-mail, and a new and easier process for membership.
There were no questions from the floor.
LACNIC:
Presentation: | PDF | PPT |Richard Jimmerson presented on behalf of the LACNIC staff. The presentation included status reports on the transition of services to LACNIC necessary for its functioning as an RIR.
There were no questions from the floor.
AfriNIC:
Presentation: | PDF | PPT |Ayitey Bulley, Interim AfriNIC Secretariat, presented an update on AfriNIC. The status on meeting ICANN's criteria for establishment as a new RIR was presented, and details of the agreement with RIPE NCC were explained. Ayitey also thanked ARIN for their letter of support that was submitted to the Interim Board of Directors in four languages.
There were no questions from the floor.
Comments:
A question from the floor was asked about publishing RIR meetings on a single calendar, noting when they overlap. The speaker asked for comprehensive calendar information. Richard Jimmerson pointed out that the information is available on the various RIR websites and on the http://aso.icann.org website.
ASO AC Election Process
Presentation: | PDF | PPT |
Speaker: Susan Hamlin
Susan Hamlin, ARIN Director of Member Services, explained the ASO AC election process by highlighting the voting and election procedures. She concluded by noting the biographies and statements of support for the candidates are available on website and that the person elected will represent the ARIN region for a three year term.
Cathy Wittbrodt was thanked and received a round of applause in recognition of her dedicated service over three years as a member of the ASO AC from the ARIN region.
ASO AC Candidates
- Kyle Hamilton - not present, the URL where biography could be found was offered
- Tanya Hinman - present, introduced herself and offered a statement on her qualifications
- Uchenna Ibekwe - not present, the URL where biography could be found was offered
- Mark McFadden - present, introduced himself and offered a statement on his qualifications
A summary of each candidate's biographical information was posted on the screen, along with the URL to more detailed information and statements of support. Richard Jimmerson encouraged all attendees to visit the URL of each candidate prior to casting their vote.
ICANN Address Supporting Organization Address Council Update
Presentation: | PDF | PPT |
Speaker: Cathy Wittbrodt
ASO AC member Cathy Wittbrodt presented a status report on the activities of the ASO AC.
Cathy reported:
- Takashi Arano and Wilfried Woeber were reelected for the 2003-2006 term
- ARIN has been providing Secretariat support this year and in 2003 the Secretariat support moves to APNIC
- Standard ICANN Board of Directors selection process being documented
- Working with Mark McFadden on update of RFC 2050
- Working closely with RIRs on ICANN reform process
- Encourages discussion at the nrr-blueprint@arin.net
Internet Resource Policy Evaluation Process
Speaker: Richard Jimmerson
Richard explained the Internet Resource Policy Evaluation Process in detail.
New policy proposals were posted to the ARIN public policy mailing list for discussion
more than 30 days prior to the start of this meeting with all text and archive material available at ARIN website (http://www.arin.net/policy/proposal_archive.html).
Reducing the Minimum Allocation/Assignment Size
Moderators: Alec Peterson and Bill Darte
The text and links to mailing list discussion on these proposals can be found at
http://www.arin.net/policy/2002_3.html; http://www.arin.net/policy/2002_7.html, and http://www.arin.net/policy/2002_9.html
Policy proposals 2002-3, 2002-7, and 2002-9 were discussed as a combined agenda item because of the similarities in the proposals. However, equal time was allotted on the agenda for each policy proposal.
Alec Peterson and Bill Darte, ARIN Advisory Council members, presented the general rationale of these policy proposals and described the issues that were previously raised on the Public Policy Mailing List. It was noted that this was not the first time the minimum allocation size has been discussed. The minimum allocation size had been previously lowered from a /19 to a /20. There was also a policy proposal submitted for consideration in 2001 to lower the minimum allocation size for multihomed networks, however it was not favorably considered.
Rationale behind 2002-3 and 2002-7:
- Multihomed networks require AS number
- Database and maintenance fee relationship already exists
- No additional routing announcements needed
- Existing ARIN policies discriminate against small businesses
- Provider independence is a primary goal
- ISPs charge too much for IP addresses
- Address space should be easier to acquire
Issues:
- Why /24 and not /32?
- Does multi-homing matter?
- Where would the address space come from? Old swamp space? New block?
- Aggregation - not possible at /20 level, with /24 allocations aggregation will only be possible at /24 in these allocations. If ARIN decides to issue longer allocations, partitioning of network would result.
- Routing table impact?
- How does one grow beyond /24? People want to avoid renumbering, because it's hard. If one grows beyond /24, they'd have to renumber.
- Could this be a one- or two-year temporary policy?
General Comments:
- You might want to consider announcing only exactly what you have, then you can mitigate a lot of the tools that extra traffic takes advantage of.
- The question of whether or not pricing should be specified in a policy was raised, and John Curran pointed out that when there's pricing strictly for the purpose of policy implication - then it should be discussed here.
- The concern was raised that if the minimum was changed to less than a /24, it would impact routability in a negative way
- It was stated that providers have historically followed ARIN minimum allocation size and that whatever is decided is going to be accepted, as long as its reasonable, as the "default" filtering limit.
- A suggestion was made that these micro-allocations and micro-assignments be made from "swamp" space to organizations that need to be multi-homed. It was pointed out that while ARIN has recovered space, it has not been put in the pool from which address delegations are made. It was later brought up that making these micro-delegations from a new pool of addresses would be easier to deal with than if they were delegated from swamp space.
- A suggestion was made that a specific range be set aside for these micro-delegations to be made from, and that providers could then choose to filter on them.
- It was noted that ARIN currently has a micro-assignment policy for critical infrastructure organizations
- It was commented on that the policies as written were very vague and inconsistently used the terms "allocation" and "assignment." Requests were made that the AC refine the language and then put them out for reconsideration.
- Questions were raised about what levels the minimum size should be changed to: /21 to /24 and higher were all discussed.
General Consensus:
- Who would support a change to the minimum allocation and/or assignment size?
Yes? 33 No? 38 - Who would be in favor of changing the minimum allocation size to /24?
Yes? 1 - Who would be in favor of changing the minimum assignment size to /24?
Yes? 0 - How many people feel if a change is made it should be made for both allocations and assignments?
Yes? 14 - How many believe that changes should be made for either allocation or assignment independently?
Yes? 27 - How many believe a change should be made to the minimum assignment size?
Yes? 39 No? 23 - How many believe a change should be made to the minimum allocation size?
Yes? 13 No? 48
It was noted that the discussions concerning these proposals, along with mailing list discussion, would be used by the ARIN Advisory Council to determine consensus.
Policy Proposal 2002-2: Experimental Internet Resource Allocations
Presentation: | PDF | PPT |
Author and Moderator: Philip Smith
The text and links to mailing list discussion on this proposal can be found at
http://www.arin.net/policy/2002_2.html
This is a proposal on experimental Internet resource allocation authored by Geoff Huston and Philip Smith. The objective of this proposal is to allow ARIN to make allocations or assignments on a temporary basis to entities conducting experiments. There is currently no ARIN policy that addresses this situation. This policy was previously approved in the APNIC 14 meeting and the RIPE 43 meeting.
General Comments:
- Clearly define that experimental requests shouldn't be able to be filled by traditional requests
- Further definition of language "commercial purposes"
- What if there was an experimental allocation to see if we could support multiple GSM phones, and they're going to report back to the community at the end of a year. Would that be considered "commercial purposes?"
- A question about the requirement in the policy for "public disclosure" was also raised, for example, if a company making an experimental request may want to work on a new architecture, they may want to work on it on a proprietary basis, and so they may not want to - for competitive reasons - disclose where they're going.
- Would cost-recovery be considered noncommercial?
General Consensus:
- How many people are in favor of policy proposal 2002-2 as written?
Yes? 49 No? 4 - How many people think the "commercial use" clause should be clarified?
Yes? 32 Left as is? 15 - Disclosure clause needs to be modified to make optional?
Yes? 0 No? 30
RFC 2050 Working Group
Presentation: | PDF | PPT |
Working Group Chair and Moderator: Mark McFadden
The group has been in operation for about a year now. Addressing policy is pretty complex, but it dates back several years. Is there actually something needed to replace the old RFC? Landscape has been changed dramatically, and updating would be difficult (many more constituents and other registries). The Working Group's objective was to address the issues relating to the relevance of RFC 2050, propose a method for replacing it with new document(s). Group activity has been coordinated through the mailing list, available on ARIN website. RFC 2050 inventory has been published on the list, and discussions have been held at three RIRs - decided to have RIRs work cooperatively rather than have a single RIR do the work. A series of documents has been proposed, and small editorial teams proposed for the documents.
I think this is really important, especially given the situation with RIRs and ICANN. It seems important to tell you where we are and invite you to participate - either as someone who's just aware of what's going on or as an active participant. If you want to contact me via e-mail, my e-mail address is mcf@uwm.edu. Mark also encouraged attendees to subscribe to the public RFC 2050 mailing list. He noted that he will be posting an update of all working group activities on the ARIN website.
There were no questions from the floor.
Policy Proposal 2002-6: Aggregation Requests
Presentation: | PDF | PPT |
Author and Moderator: Bill Woodcock
The text and links to mailing list discussion on this proposal can be found at
http://www.arin.net/policy/2002_6.html
This policy is meant for organizations that have a number of small unaggregatable blocks. Organizations should be able to return them to ARIN and receive a replacement aggregatable block without too much justification.
General Comments:
- The suggestion was made that the policy should be modified to specifically apply to only portable blocks
- It was also suggested that if the organization didn't have 70% of utilization or more of the returned space then they could get a contiguous allocation of the same size, but not a larger size - with /20 being a minimum size
- Bill Woodcock pointed out that the intention is not for someone to be able to get more than they had
- Routability was raised as an issue again, in terms of the size of the allocation that could be received
- A question of whether organizations would be allowed to barter between themselves about returning blocks and receiving aggregatable blocks, or if the allocations would be left up to ARIN staff. It was explained that while it was not specified, it was intended to be left to the ARIN staff.
- Anne Lord, with APNIC, added that they have a similar policy in effect and they have not had great demand for it, as renumbering seems to be a deterrent, She also pointed out that when giving out space to those that returned it under their policy, they were given an allocation up to the next bit boundary
- A suggestion for language clarification was made: the policy as proposed stated that the end-user is allowed to select space. The suggestion was that a change be made to the text that ARIN will select the space. Specific language suggested "Shall receive at their option," rather than "shall be allowed to select"
- It was also suggested that a period of twelve months be added to the policy as the period organizations would have to renumber
- A limit on the size of an allocation received was also suggested, with up to a /20 being suggested
General Consensus:
- Support policy as is with changes (limit on amount allocated, etc.)?
Yes? 8 - Support limitation that returning up to a /20 gets an organization a /20 in return
Yes? 0 - Support policy as written?
Yes? 49 No? 0
Policy Proposal 2002-5: Amnesty Requests
Presentation: | PDF | PPT |
Author and Moderator: Bill Woodcock
The text and links to mailing list discussion on this proposal can be found at
http://www.arin.net/policy/2002_5.html
This proposal shares much in common with policy proposal 2002-6. The rationale of this policy is that if your organization has a block and doesn't need a block that large, and it's still the power of 2, your organization should be able to turn it back in to ARIN and get something smaller without having to justify it.
General Comments:
- Like the last policy, this one should be considered to be modified with the same additions:
- 12 month limit to renumber
- Applies to portable addresses only
- Any smaller size down to a /24
- ARIN should inform the people using this policy (if adopted), that the addresses they receive may not be routable
- Discussion focused on what should happen if they are or are not currently paying fees, and what fees they should pay under this policy. It was brought up that fees help ensure people maintain their records
- It was also mentioned that many organizations who are legacy space holders shouldn't be charged if they aren't currently paying
General Consensus:
- Should ARIN, in theory, have an amnesty policy with a /24 minimum limit?
Yes? 51 No? 2 - Should the ARIN Advisory Council look at other limits besides a /24?
Yes to /24 limit? 17
Yes to larger limit (either /20 or whatever ARIN minimum allocation size is)? 11
Community Learning and Education Working Group (CLEW):
Presentation: | PDF | PPT |
Working Group Chair and Moderator: Bill Darte
Bill Darte reported on the group's history, its recent charter revision and the general successes and failures of CLEW.
Bill's recommendation was that with the lack of involvement in the group, and the fact that other avenues existed to present ideas for training, the CLEW working group could be disbanded.
General Comments:
- ARIN should have at least one person responsible for training
- Other RIRs have extensive training schedules. It would be really interesting to see how ARIN staff responds to requests. Are there really many training requirements for this region?
- Other RIRs don't have working groups like CLEW because it is not perceived as being needed
- Richard Jimmerson noted that RIPE NCC and APNIC training courses focus primarily on training customers to follow their procedures. He further commented that ARIN has put much work into simplifying its procedures and that there is a telephone help-desk available 12 hours each work day to provide immediate, interactive support. He also noted that ARIN did conduct some in-person training this year in support of the database conversion and have also developed a CBT.
General Consensus:
- Should the CLEW working group disband?
Yes, passed unanimously
Bill Darte was thanked for his leadership and work associated with CLEW.
ICANN Reform
Moderator: John Curran
Information on this topic can be found at
http://www.arin.net/library/ICANN/
Summary:
- RIRs concerned about ICANN reform process
- chronology of RIR/ICANN interaction explained
- reforms under consideration
- Stressed lack of ICANN response to RIR input
- Described proposed creation of a Numbering Resource Registry (NRR) http://www.arin.net/nrr-blueprint/
- Relation between NRR and RIRs/ASO/ICANN
- Recognition of ICANN/Department of Commerce Memorandum of Understanding
- Description of the NRR Blueprint
General Comments:
- Are there any similar initiatives in the protocol community?
Leslie Daigle, Chair of the IAB stated that the IAB has expressed concern over the ICANN reform process. They do not have a very different experience than the RIRs in dealing with ICANN. In addition, they also feel a need to move on and deal with issues, with or without ICANN. - What breaks when ICANN breaks? What has it brought to us and what do we lose when it goes away?
- Concerns about needing an independent nonbiased oversight organization to the RIRs were raised
- Questions were raised about why the RIRs developed the NRR-Blueprint without first going to its membership or the community it represents. John Curran pointed out that the blueprint was not the end of the matter, but was in fact aimed at developing a dialogue in the community and should be considered only a first draft. The nrr-blueprint@arin.net mailing list was pointed out as the place for this discussion to take place. He mentioned that the RIRs requested feedback from the community on the issue of ICANN Evolution and Reform each time a joint statement was released prior to the publishing of the NRR Blueprint. He noted that limited comments were received.
- The lightweight nature of the NRR was commented on as being a good thing
- Support, in the nature of an e-mail to the nrr-blueprint@arin.net mailing list, from the ccTLD community was cited
- NRR was mentioned as a better solution than having the IANA function move to the ITU
- The question was raised about what role the U.S. Department of Commerce would have with the NRR. John Curran explained that the RIRs had met with Commerce before they published the NRR blueprint.
- John Curran also added that ARIN and the other RIR's goal is to contribute to the stability of the Internet regardless of the status of ICANN.
It was asked if this session could be deferred until Public Policy Meeting, Day 2. There were no objections.
The meeting was adjourned at 5:05 PM (PST)
Minutes of the Public Policy Meeting - Day 2
Thursday, Oct. 31, 2002
Richard Jimmerson called the meeting to order and reviewed the day's agenda.
Mark McFadden was announced as the winner of the ASO AC election. There were approximately 130 people in attendance.
Database Implementation Working Group (DBWG)
Presentation: | PDF | PPT |
Working Group Chair and Moderator: Ginny Listman
There were several topics discussed.
Post-conversion report
Moderator: Ginny Listman
Ginny Listman, ARIN Director of Engineering, presented a report providing statistics on use of the new features of the database.
General Comments:
- It was asked what the community's feelings were about how rejected templates should be marked. It was suggested that the return e-mail should include either "Approved" or "Rejected" in the subject line, along with the appropriate ticket number. Richard Jimmerson agreed that ARIN staff would take this as direction from the community.
Discussion of WHOIS output
Moderator: Ginny Listman
Ginny asked for feedback on what the process should be to make recommendations and implement changes on the format of WHOIS results.
General Comments:
- The suggestion was made that rather than continuing development of WHOIS and RWhois, ARIN should investigate open directory standards such as LDAP. Mark Kosters pointed out that VeriSign had investigated some other directory services and that in testing they could be as fast as WHOIS is currently. It was recommended that a specific proposal regarding this be posted to the dbwg@arin.net mailing list.
- Suggestion for option to list two (2) separate address fields in WHOIS output, rather than using a delimiter such as a comma or semi-colon.
- Suggestion to include a flag to produce a full postal address for network queries.
- Suggestion was made that each page of results needs to specify an acceptable use policy for the data displayed. There were concerns that without a mechanism for enforcement the AUP would be useless. The recommendation was made that this should be proposed as a separate policy proposal.
- Frustration was raised with the need to use wildcards in command-line WHOIS queries where there was no previous need. The question was asked if this was for searches on the person's own network or someone else's. The response was that it was for searches on their own network. Richard Jimmerson pointed out that new tools to see your organization's reassignments would be discussed later in the day.
- A question was asked why a statement about portable or nonportable was not appearing on new allocations in WHOIS output.
- ARIN staff responded that the net type field is being used to denote portability status and that it would be possible to put a nonportable statement next to the net type in WHOIS output.
- There was also a suggestion that portability status be made a field in the database.
- Richard Jimmerson said that the annotation could be added next to the net type field.
- Leslie Nobile added that the designation was a feature of the old ISP request template, not an automated feature of WHOIS.
Stale POC Data
Moderator: Ginny Listman
Ginny presented information about the progress of identifying stale POC data in the database and determining what further work needs to be done.
Policy Proposal 2002-8 Privatizing POC Information
Presentation: | PDF | PPT |
Moderator: Sanford George
Sanford George, ARIN Advisory Council member, presented this proposal on behalf of the original author.
The text and links to mailing list discussion on this proposal can be found at
http://www.arin.net/policy/2002_8.html
General Comments:
- Concerns expressed about balancing the need to protect internally from spam and providing personally accountable contact information
- A suggestion for providing different access to the database based on need or other qualifications. It was pointed out that this would need to be a specific proposal. Authentication of users was also pointed out as something that would need to be addressed
- One point of contact needs to remain visible
- Ginny Listman stated that the POCs would still be visible in WHOIS, but not displayed with the associated organization or resource
- There was some discussion regarding role vs. person and that this is unrelated to the discussion of the privatization of POCs. Background was provided on the types of POCs available.
General Consensus:
- How many accept this proposal as written?
Yes? 53 No? 0
Policy Proposal 2002-1 Lame Delegations in IN-ADDR.ARPA
Presentation: | PDF | PPT |
Moderator: Ron da Silva
The text and links to mailing list discussion on this proposal can be found at
http://www.arin.net/policy/2002_1.html
Click here for a text file containing background information, prepared by ARIN staff, relating to this proposal.
Ron da Silva, ARIN Advisory Council member, moderated the discussion.
General Comments:
- The proposal as written needs language clarification. It also needs to talk about how to contact the organizations. There is an ongoing project by NANOG, so it's a question of economics. What's the value of this work, and how's that going to affect our fees. I haven't been able to discern the economic value of this.
- It's OK to take the delegation away, but you should keep a record of what was there in case someone comes after you.
- There should be more than one measuring point to establish lameness
- Suggestion that ARIN publish something on the website, to explain not only why it's bad to be lame, but also to ask them to contact ARIN to do something about it.
General Consensus:
- Should ARIN contact organizations about lameness?
Yes? 60 No? 1 - How should they contact the organization? Approval in increasing levels of escalation from e-mail, to Route lookup for ASN, to telephone call, to postal mail?
- Only E-mail? 0
- Only E-mail and Route lookup for ASN? 1
- E-mail, ASN lookup and Telephone? 8
- All the steps outlined above and including postal mail? 46
- If contact is not made, what steps should ARIN take?
- Mark and store - 0
- Remove delegation - 50
- Do nothing - 0
- Accept proposal as is?
Yes? 6 - Accept it with suggested changes?
Yes? 39
Questions from Richard Jimmerson prior to opening the floor to next proposal:
- How many people here are attending an ARIN meeting for the first time? 40
- Of those 40 people, how many attended the NANOG meeting - about 40
- How many would not have come to the ARIN meeting if it had not been for the NANOG meeting earlier this week? about 40
- Of those of you who are here for the first time, either because of the NANOG meeting or another reason, how many would attend in the future? How many would plan to come back to an ARIN meeting without a NANOG meeting? 22
- How many of you that are attending for the first time are more likely to come to an ARIN meeting if it was back to back with a NANOG meeting? 21
Policy Proposal 2002-4 Bulk Copies of ARIN's WHOIS
Presentation: | PDF | PPT |
Moderator: Lea Roberts
The text and links to mailing list discussion on this proposal can be found at
http://www.arin.net/policy/2002_4.html
Lea Roberts, ARIN Advisory Council member, moderated the discussion.
Richard Jimmerson provided some background on the topic. There was a policy discussion (2001-7) at ARIN VIII, where the community stated that ARIN may provide bulk copies of ARIN WHOIS data minus POC information, and ARIN has been doing that ever since. Since then, many organizations have come to ARIN staff saying that the data minus the POC information is useless to them. We've also had many organizations come to ARIN who are interested in having ARIN provide bulk copies of ARIN data similar to what is provided by RIPE NCC and APNIC which includes POC information.
General Comments:
- RIPE NCC has a slightly more complex framework, with three types of AUP. First they have to sign the AUP and meet certain conditions, and it doesn't imply any bulk copies. The second is a project based AUP, and it may or may not contain contact data. The 3rd type is a mirror data AUP, where they get bulk copies of the RIPE database on a continuing basis. However there are certain conditions, like no telemarketing and no redistribution.
- RIPE NCC - Proxy AUP for script harvesting, Project AUP - may or may not contain POC info, Mirroring AUP - includes POC, no bulk distribution, includes POC info
- APNIC makes bulk copies available, and mirroring, by using the same AUP that the requestor has to sign, and it states that they will be using it for Internet experimental purposes, but not spam-type purposes.
- Question was raised if it would be possible to insert "honey pot" data in the database to monitor for proper usage (i.e. no spam). It was noted that “honey pot" data would need to be different for each copy of the distributed WHOIS in order to track if someone is misusing the data.
- Two levels of access proposed. Speaker was asked to formulate a policy proposal.
- Perhaps a time limit for accessibility, there should be a way to make clear that the information they have is not their property and can not be resold.
General Consensus:
- Accept proposal as is?
Yes? 0 - Accept proposal with change that private POC info be kept out of bulk?
Yes? 27 - Should there also be a mirroring service?
Yes? 3 No? 5
Next Steps for ARIN Database
Presentation: | PDF | PPT |
Speaker: Cathy Murphy
Cathy Murphy, ARIN Principal Engineer, discussed the engineering plans for the near future. There were no comments or questions from the floor.
Early Registration Transfer (ERX)
Presentation: | PDF | PPT |
Speaker: Ginny Listman
Ginny presented an overview of the transactions that have taken place and the future plans.
General Comments:
- A question was posed about pointing people to the correct registry for any record in WHOIS. Richard Jimmerson pointed out that after the ERX project is completed, one of the next projects is for a joint WHOIS.
Inserted Agenda Item- Getting Consensus on ARIN Minimum Allocation Size Policies
Moderators: Alec Peterson and Bill Darte
The Advisory Council met the previous evening and determined that they needed further clarification in order to consider each policy separately.
Questions about 2002-3
- Accept policy as written? 2
- Reject? 3
- Accept some changes? 36
Questions 2002-7
- Accept policy as written? 0
- Reject? 13
- In favor of not approving as written, but try to make appropriate changes? 31
Questions 2002-9
- Accept policy as written? 2
- Reject? 24
- Approve with appropriate changes? 4
General Questions:
- Change minimum assignment size? 16
- Change minimum allocation size? 0
- Change minimum size for both? 12
- Should multihoming w/ an ASN be a requirement for allocation? 39
- Should multihoming alone be a requirement? 1
Minimum assignment size be changed to:
- /21 1
- /22 3
- /23 0
- /24 14
- /longer 0
Minimum allocation/assignment size to:
- /21 0
- /22 1
- /23 0
- /24 6
- /longer 0
Justification criteria:
- Should we enact a special policy for micro-assignment w/ multihoming as requirement?
Yes? 10 - Multihoming requirement for micro-assignments/allocations?
Yes? 3 - Utilization needs to be considered in assignments or allocations?
Yes? 12 - Utilization needs to be criteria for any assignment only?
Yes? 0 - Should a micro-allocation policy implemented, should it be interim 2-year policy?
Temporary? 16
Permanent? 13
Special fee considerations:
-
Should special fee consideration as a part of this policy?
Yes? 36 No? 0
Subsequent allocations:
- Micro-assignments/allocations should include reservations for subsequent delegations?
Yes? 6 No? 9
Routing Table Measurement and Analysis (RTMA) Working Group
Working Group Chair and Speaker: Cathy Wittbrodt
Philip Smith presented a routing table update.
Cathy Wittbrodt presented findings of work being conducted by graduate students at UCLA to show the amount of time between the allocation of a prefix and its appearance in the global routing table.
General Comments:
- You should only inject into the routing system things you're actually using. The rest of it sort of sits there. Is that what this means? Cathy Wittbrodt responded that it means there are routing table entries that are parts of other fragments.
- Maybe we can tune the size of the minimum allocation unit if we have empirical data that says people aren't using everything we're giving them. Cathy responded that she believed that a lot of the fragments total the allocated space but we should look into that.
- How many of these were actually covered with the larger aggregate announcement with a bunch of /24's in them, rather than all /24's? Cathy responded that was what we called identical + fragmented. Follow up question was if there are any distributions - over 60% of the allocations no longer advertised - the allocations were gone, and they're not being used. Cathy responded that they may be used, but they're not in the routing table. From where we were looking, we could no longer see them in the routing table.
Proposal for Transfer of 6Bone Address Management Responsibilities to RIRs
Moderator: Mark Kosters
The text of this proposal and information on how to provide input until December 31, 2002 is available at: http://www.arin.net/announcements/08202002.html
Mark Kosters, ARIN Advisory Council member, pointed out that while this is not a formal ARIN policy proposal; comments have been requested on this document in all three RIR regions and the 6bone community. There have been no comments on the PPML mailing list, and comments from the 6bone list have focused on cost.
General Comments:
- Question regarding the phrase “low as possible" -- $500 is a lot more difficult than, say, $5.00. I'd like to see something more specific.
- I think this is kind of like the experimental allocations, except this isn't a limited time experiment that will hopefully phase out. So it looks like we've allocated a bunch of addresses to an organization that will then sub-allocate them. If someone is retiring, that's their problem, why is this our problem? They can come to the registry.
- This space shouldn't be treated differently than any other
- The purpose of this proposal is to shut down 6bone, not to produce something that's longer for the RIRs. The RFC which created the allocation ... the point is to shut down 6bone and this is a mechanism to do so - to normalize the assignments that have been used in 6bone and to shut it down
- There are more than 120 prefixes out there for 6bone, and the idea that you have to have an ID that basically takes that away, has that been socialized in the 6bone community yet?
- The question was raised whether a concrete date should be put in the policy. It was pointed out that while which date should be put in is not known, it might be something to consider for the policy
- The question of what would be the consequence of the registries NOT doing this was raised. It was explained that the RIRs developed this proposal and that buy-in still had to occur with the 6bone community. The possibility of RIR involvement creating a more orderly demise for the 6bone experiment was raised as well.
- There is a lot of address space. I'm not so sure that registries should be in the business of trying pressure people to do things. If people want to continue to use 6bone forever, why should we have anything to say about that? For people who are connected to 6bone, there are things they can do to move away from it. I really don't understand why we're trying to force people to do things one way or another. A response was made that the IETF is going to terminate the original allocation that they granted. The registries aren't trying to take it away
- The original plan was that this (the 6bone) would be a short-term use of experimental space.
Mark Kosters pointed out that this will all be taken as feedback, and a summarization of the comments will be produced following the end of the comment period on December 31, 2002.
Break - The audience engaged in a round of applause on the news of ICANN's approval of LACNIC as the fourth RIR.
Adjustment of Maintenance Fees
Presentation: | PDF | PPT |
Moderator: Lee Howard
Lee Howard, ARIN Treasurer, proposed an increase of the maintenance fee from $30 to $120.
General comments:
- Go for it!
- What is the cost that needs to be recovered? Lee responded that the difficulty is the invoice/payment process
- What about legacy space holders?
- Would like some focus on the collection side, addressed independently
- Domain side has economy of scale that doesn't exist on the RIR side
- What about a lower fee, but institute a penalty fee for those who don't pay? It was pointed out that this would actually increase costs for ARIN.
- Need expressed for some sort of cost/benefit analysis
- Cost recovery is only part of reason for fee, also used to help ensure data maintenance
- Proposal of flat annual fee (i.e. $100) per organization, rather than per object
- Suggestion of a third part collection agency. It was explained that collection companies will not normally go after $30 invoices
- Question whether small or large fees actually encourage maintenance of data
- The goal is to get people who are not members into the fold. Fee shows that they have responsibility to keep database maintained. Approximately what it will take to recover ARIN cost. Not cost prohibitive and fair.
- Those that are paying the fees would have to be paying the way of those that aren't paying the cost.
- What about a hybrid? Like $100/year and $20/object?
General Consensus:
- Support fixed rate maintenance service fee (i.e. $100 or so)?
Yes? 30 - Support hybrid fee (i.e. $100 year, $5 or $10 object)?
Yes? 16 - Increase in per-object maintenance fee to $120?
Yes? 0 - Establish fixed annual maintenance fee of $100 to $120?
yes, majority supported
General Comments:
- Question about swamp addresses being involved in ERX. It was explained that RIPE NCC will only charge fees if they want to become members and if people want to access services through members, RIPE won't charge for that.
- The question of what happens in with the ERX project if contact information doesn't exist for a record. It was explained that the registration was being transferred based on the country code of the resource. If the country code is outside the ARIN region, it would be included on the slate of resources to be transferred.
- If we start saying that the legacy holders are free of paying particular fees, then why aren't they all? Where do you stop?
- I think it's something that if anyone that has legacy space is going to be charged, then everyone who has legacy space needs to be charged. I think the decision this year to waive the transfer fee was a terrific incentive to get people to clean up their records. Part of that that's puzzling for me is why someone who is currently not paying for a legacy space would then on an ongoing basis be charged for maintaining that space. I think we need to address equal treatment and incentives.
- I have a legacy.com and a legacy.org and I have not been exempted from paying fees for them ever since fees were enacted. However, I think whether we do it as a per-org or whatever, I don't think it should necessarily be something where we charge them the same as someone who's getting a new allocation, but I think we need a fee structure that's fair for everyone, and a financial incentive for people in the swamp to maintain their records or possibly lose their space in the swamp in ARIN's database.
- How about ARIN not try to charge for swamp space, just publish a list of blocks that are not paid up and if people don't want to deal with these people, that's up to them.
- One of the fallouts of this whole discussion about ICANN is what the authorization record trail is for ARIN and the other RIRs to allocate address space. Right now that goes through ICANN. If we say we're going to start exerting our control over legacy space, our authority might be questioned.
- I have some confusion about how Scott and Bill Woodcock can suggest that there are historical precedents that you can't charge for formerly-owned address space - the argument that there is a registry of last resort doesn't hold water, because everyone charges these days for domain space. I think that since ARIN wants to avoid lawsuits, should we try to charge for legacy space that a transaction based charge, as painful as it would be for ARIN staff, would be the fairest way to do it.
- The problem is not a transaction fee, the problem is the slippery slope of going down the road and eventually charging legacy block holders the same as ARIN members. Legacy holders are not, mostly, large ISPs. There are people with legacy address blocks and if they start getting billed at the same rate that ARIN members get billed, they'll have no way of paying those fees - it's not an option. What service are they receiving from ARIN? There hasn't been any maintenance of the records.
- The question was raised of what ARIN's legal obligation, if any, to maintain these historical records was. John Curran replied that legal obligation is an interesting challenge - there are a number of things we have to do. We need to foster the support and development of the Internet and if we were to pull the records, someone might be able to say that we aren't following our charter. We don't have a contract, as I understand it, that says we have to continue carrying those records, but that could change if we signed a contract with ICANN.
- An inquiry was made on the general thoughts of those at the meeting to have the ARIN staff make an ongoing verification for multi-homing. Further discussion revealed that this would be very difficult to accomplish with any amount of certainty.
Richard Jimmerson encouraged all attendees to complete the meeting survey which he stated was now available on the website. He again expressed thanks to the meeting sponsors: University of Oregon, Sprint, and PCI - who have done a wonderful job. Richard gave a special thanks to the NANOG Program Committee for the ease of coordination over the last six months.
The meeting was adjourned at 5:04 PM (PST).