- cooperation-wg - mailman.ripe.net

Key elements of the transition of IANA stewardship
by Richard Hill 10 Sep '14

10 Sep '14

NOTE: this message has been cross-posted individually to the RIR mailing lists dealing with this issue. Here are some things that appear to me to be key elements for the transition of the IANA stewardship. At present, the "RIRs are an interested and affected party of the IANA contract because IANA holds ultimate responsibility for allocated and unallocated IPv4, IPv6 and Autonomous System Number address spaces. IANA delegates IP and ASN address blocks to the RIRs on a needs-based approach according to global policies agreed by all the Regional Internet Registries (RIRs). The global policy development process is described in the ICANN Address Supporting Organization (ASO) memorandum of understanding. ICANN and the Number Resource Organization (NRO) signed this MoU in 2004. The NRO is an unincorporated organization created in 2003 as a coordination mechanism for the RIRs. The NRO fulfills the role, responsibilities and functions of the ASO as defined within the ICANN Bylaws." (This is a citation from the document at: https://blog.apnic.net/wp-content/uploads/2014/09/NRO-RELATIONSHIP-IANA.pdf ) The Mou between ICANN and the NRO is at: https://aso.icann.org/about-the-aso/aso-memorandum-of-understanding/ That MoU specifies how ICANN's ASO is constituted and provides that "Under this agreement the ICANN Board will ratify proposed global policies in accordance with the Global Policy Development Process, using review procedures as determined by ICANN." Thus, the ICANN Board has ultimate responsibility for IP address policies. This is consistent with the ICANN Bylaws. Up to now, under the IANA functions contract with the NTIA, ICANN was clearly bound to defer to the RIRs for what concerns IP address policies. If there is no contract between ICANN and some external entity, then ICANN would have unrestricted ultimate authority over IP addresses. That is, the ICANN Board could, if it considered it appropriate, override RIR policies. This puts too much power in the ICANN Board which, under ICANN's current structure, is not accountable to any external entity. It would appear desirable that the IANA functions should be contracted for by the served communities, that is, by the NRO/RIRs for what concerns IP addresses. And indeed the draft proposal presented by APNIC on 8 September 2014 envisages a Service Level Agreement between ICANN and NRO and also a non-binding Affirmation of Committments between ICANN and NRO. These proposals are found in the slide show contained in the web page at: http://blog.apnic.net/2014/09/08/iana-session-apnic-38-a-discussion-propos al/ But a non-binding Affirmation of Committments, coupled with a Service Level Agreement, are not the equivalent of a contract for the IANA functions. Thus, it would appear more appropriate to adopt the same approach that has been adopted by the IETF regarding protocol parameters, namely a Memorandum of Understanding (which appears to be a contract) between the NRO and ICANN for the IP addresses. The Memorandum of Understanding between ICANN and the IETF is at: https://www.icann.org/resources/unthemed-pages/ietf-icann-mou-2000-03-01-e n and http://tools.ietf.org/html/rfc2860 That text could easily be modified to refer to NRO and IP addresses instead of IETF and protocol parameters. However, item 4 of the text should be modified so that it reads as follows: "4. Agreed technical work items. ICANN agrees, notwithstanding any provisions in its Bylaws or other corporate documents that might be construed differently, that during the term of this MOU it shall cause IANA to comply ..." In addition, consideration should be given to adding a choice of law clause and a dispute resolution clause, presumably referring to arbitration rather than to national courts. There was extensive discussion (but no agreement) on the IANA Transition mailing list regarding whether or not the fact that a US court could, in theory, order ICANN/IANA to do something contrary to agreed community policies is an issue and, if so, whether anything should be proposed to deal with that issue, such as proposing that the entity that performs the IANA function should have immunity of jurisdiction, or that the entity should have redundant sites in more than one jurisdiction. If there is support for dealing with that issue, then some text could be added. Finally, and in addition to the above, article I.1 of the ICANN Bylaws should be modified to make it clear that it is the NRO that has the overall responsibility for the coordination and allocation and assignment of IP addresses. That change is needed because, at present, article I.1 of the ICANN Bylaws implies that ICANN has the overall responsibility for the coordination and allocation and assignment of IP addresses. Best, Richard

1 0

RIPE NCC Report on IGF 2014
by Chris Buckridge 05 Sep '14

05 Sep '14

Dear colleagues, The ninth annual Internet Governance Forum (IGF) concluded on Friday, 5 September, in Istanbul, Turkey. Over the course of the week, RIPE NCC staff and RIPE community members contributed to many of the workshops and main sessions. These sessions looked at topics like the impact of carrier-grade NAT technology on users and policymakers, the IANA oversight transition process, cybersecurity and capacity building. You can find more information about the week’s discussions, and the RIPE NCC’s involvement, on the RIPE NCC website: https://www.ripe.net/internet-coordination/news/ripe-ncc-report-on-igf-2014 Best regards, Chris Buckridge Senior External Relations Officer RIPE NCC

1 0

Update on IANA Oversight Discussions: ICG Meeting and IETF 90
by Chris Buckridge 30 Jul '14

30 Jul '14

Dear colleagues, In recent weeks, two significant community discussions regarding the transition of oversight of the IANA functions have taken place. The following summaries are provided as feedback to the RIPE community - an online version of these summaries, containing links to more detailed reports and information, is available at: https://www.ripe.net/internet-coordination/news/industry-developments/updat… Best regards, Chris Buckridge Senior External Relations Officer, RIPE NCC ——— First Meeting of the ICG The IANA Stewardship Transition Coordination Group (ICG) held its first meeting in London on 17-18 July 2014. This group has been established to coordinate input from the various relevant communities and IANA stakeholders, moving towards development of a global transition plan. Discussions and outcomes of the meeting are documented on the ICANN website. Adiel Akplogan and Paul Wilson, the CEOs of AFRINIC and APNIC respectively, took part in the meeting as representatives of the Number Resource Organization (NRO), and Hartmut Glaser, NRO Number Council member from the LACNIC region, represented the Address Supporting Organization (ASO). In addition, Daniel Karrenberg, RIPE NCC Chief Scientist, participated as one of two representatives of the Root Server System Advisory Committee (RSSAC). Key points from this meeting: • ICG members strongly asserted the group’s independence from ICANN. • The constructive atmosphere in the meeting resulted in the successful definition of a draft charter [PDF] and organisational principles for ICG. • The draft charter acknowledges that parts of the proposal will be developed in parallel across three main pillars (names, numbers and protocol parameters), coordinated by the respective operational communities. • The draft charter also makes it clear that the ICG does not expect to create substantive content in the proposals itself. Should serious further work be needed on the proposals, the ICG expects to send them back to the respective communities to do that work. • The group agreed to the request from the ICANN Government Advisory Committee (GAC) for five seats (rather than the proposed two). • The ICG recognised that there is a very tight schedule for developing a global proposal to the NTIA for transition of IANA oversight, and called on the various communities to produce input to a global proposal by the end of 2014. • There were around 100 remote observers for both days of the meeting. • The next meeting of the ICG is being planned for 6 September 2014 in Istanbul, Turkey. IETF 90 IANAPLAN BoF Session On 24 July 2014, during the Internet Engineering Task Force (IETF) meeting in Toronto, a dedicated Birds-of-a-Feather (BoF) session was held to discuss the IANA oversight transition. Approximately 150 people participated on-site, with more participating remotely. The session was chaired by Andrew Sullivan and Marc Blanchet. Session materials, including minutes and presentations, are available from the IETF website. Key points from this session: • Alissa Cooper reported on the first meeting of the ICG, and noted that the IETF community will take the lead in drafting a proposal regarding the protocol parameters. • Participants strongly supported the notion that the current arrangements (both contractual and operational) should not change significantly as a result of the transition process. • It was agreed that there should be reviews to ensure that: • Current procedures and relationships are sufficiently documented • The transition of oversight will not require significant changes of the current arrangements • In terms of IETF process, there was agreement that, while the Internet Architecture Board (IAB) should take a leading role, there should be a dedicated Working Group formed to act as sounding board and confirm rough consensus on the proposed transitioning plans. This working group could act as a focal point for other stakeholders to relay input back to the IETF.

2 1

IANA Stewardship Coordination Group Meeting
by Chris Buckridge 17 Jul '14

17 Jul '14

Dear colleagues, As noted previously, a Coordination Group has been formed to facilitate development of a global proposal to the NTIA for its transition out of the IANA oversight role. This Coordination Group meets for the first time today and tomorrow (17-18 July) in London: https://www.icann.org/news/announcement-2014-07-16-en Please note that a Virtual Meeting Room has been set up to allow the community to follow the meeting remotely: https://icann.adobeconnect.com/coordinationgroup/ An agenda for the meeting is available here: https://www.icann.org/en/system/files/files/agenda-stewardship-coordination… Best regards, Chris Buckridge, RIPE NCC

1 0

ICANN 50: Talking IANA and Accountability
by Chris Buckridge 01 Jul '14

01 Jul '14

Dear colleagues, The RIPE NCC has published a brief summary of discussions at last week’s ICANN 50 meeting on issues including the IANA oversight transition and ICANN accountability. The news item also includes links to the transcripts and audio archives of these sessions: https://www.ripe.net/internet-coordination/news/industry-developments/icann… Best regards, Chris Buckridge, RIPE NCC

2 1

Updated IANA Oversight Transition Process
by Chris Buckridge 16 Jun '14

16 Jun '14

Dear colleagues, Following a public comment period during which more than 50 submissions were received, ICANN has posted an updated description of the “Process to Develop the Proposal and Next Steps” in relation to the transition of the NTIA’s stewardship of the IANA functions: https://www.icann.org/news/announcement-2014-06-06-en The page details all of the changes made to the initial draft, which was published in April. A couple of key points: - The title "Coordination Group” has replaced the original "Steering Group" - The number of participants has increased from 22 participants to 25, including the addition of numerous stakeholder groups and an IANA expert liaison - The number of representatives from the Address Supporting Organization (ASO) has decreased from two to one - Representatives are to be selected by their communities, not by the ICANN Board and Government Advisory Committee (GAC) - Coordination Group members are to be announced by 2 July (but preferably before or during the ICANN 50 meeting, which runs from 22-26 June), with an expected first public meeting to be held in mid-July The number of Internet number resource community representatives on the Coordination Group totals three - two from the Number Resource Organization (NRO), and one from the Address Supporting Organization (ASO). More information on the RIPE community contribution to the development of a proposal regarding transition of the NTIA’s oversight of IANA, including summaries of discussions at recent RIPE community events, is available at: https://ripe.net/iana-oversight-transition Best regards, Chris Buckridge Senior External Relations Officer, RIPE NCC

1 0

Open Call to join IGF Best Practices Forums Preparatory Process
by Filiz Yilmaz 12 Jun '14

12 Jun '14

Dear all, Please note this call: http://www.intgovforum.org/cms/open-call-to-join-igf-best-practices-forums-… Kind regards Filiz

1 0

Minutes of the Coop-WG meeting during RIPE 68 in Warsaw - Updated
by Alain Van Gaever 11 Jun '14

11 Jun '14

Dear All, Following modifications proposed by Jim and Daniel, please find below the updated minutes of the Coop-WG meeting in Warsaw In case you have any further comments/updates please let us know Alain ------- Beginning of Minutes ------- Cooperation Working Group Draft Agenda Thursday, 15 May, 11:00 – 12:30 A. Administrative Matters The co-chairs opened the meeting. New co-chairs Meredith Whittaker and Alain Van Gaever introduced themselves and briefly outlined their hopes for the working group, specifically ensuring that the content is relevant and of practical use to the RIPE community. The minutes from RIPE 67 were approved, as was the agenda for RIPE 68. B. Content Blocking • B1. A Technical Overview of Content Blocking Methods – Pier Carlo Chiodi, Olaf Kolkman Olaf Kolkman gave an overview of the work done by Pier Carlo Chiodi on content blocking on the Internet and other work in the same area. Outlining a number of strategies employed by those wishing to block content, he noted that blocking is done most effectively at the end-point (or origin of the content), and that blocking in the network involves a variety of trade-offs. Lars-Johan Liman noted that even “legitimate” blocking has collateral damage, and pointed to the example of the hotel network, where the interception of traffic prevents the use of DNSSEC. He suggested that in such situations it is best if the effects of the blocking are stated upfront for users. Andrei Robachevsky recalled a paper produced by ISOC (prepared around the time of the SOCA/PIPA/ACTA legislative proposals), and noted that security and hidden costs are all important, and public policy makers need to be made aware of these negative impacts. Olaf pointed out that users will do whatever they can to get to blocked content, and that this can also have potential negative effects (including a greater viral footprint or exposing backdoors to computer systems). Alexander Isavnin asked participants in the room who thought their countries were doing blocking, and whether they thought it was being abused. Richard Barnes noted that these issues reinforce the importance of an end-to-end strategy, and agreed that the hotel network is a good example of how blocking can break important security elements like DNSSEC. Jim Reid noted that blocking access to specific content can mean that broader services are blocked. He had advised on a case where one of the questions asked was "what else would be affected if we prevented port 80 access for this particular domain?”. The only person who could provide a definitive answer would be the administrator of the web server at that IP address and they would be unlikely to cooperate with anyone intending to blocking traffic for that address. • B2. Telex: A Proposal For Circumventing Censorship in the Network – Eric Wustrow Eric Wustrow outlined the Telex project, which has been developed to circumvent content blocking measures. The system provides a means of connecting users to blocked content via a mechanism that is invisible to the censoring technology. He encouraged ISPs to contribute to the work with advice and prototype deployment assistance. Robert Kisteleki noted that PGP key distribution may be a bottleneck in the Telex system. Eric noted that the paper discusses some of these issues, including preventing censors from distributing “bad” public keys. He suggested that having a central Telex entity that is known and trusted will be important to mitigate these risks. He also noted that getting information into areas subjected to censorship is often less difficult than might be supposed. • B3. Web Censorship Circumvention: Challenges and Opportunities – Walid Al-Saqaf Walid Al-Saqaf outlined his project to map URL filtering via crowdsourcing, which is developing longitudinal data, while allowing contributing users to access blocked websites via his own servers. He presented a range of data and analysis obtained from this work, including content blocking methods and strategies. He noted the need to make people more aware of the range of circumvention tools and solutions available, and that speed, security and anonymity are all important to users. He also described his plans for the future, including open-sourcing his own project and cooperating with similar projects. Andrei Robachevsky asked whether the project looked at which means of blocking were most common. Walid stated that he has done some analysis of this, using the packet headers - the more data he can get, the better this analysis will be. Meredith Whittaker noted that the Open Observatory of Network Interference (OONI) project, coordinated by the Tor team, is also doing work in this area and is generating public data. Alexander Isavnin suggested that a RIPE task force might be a useful vehicle for RIPE community members interested in this issue. Walid agreed that the technical community, and particularly its relationship to civil society actors in this space, needs to be further explored and understood. C. IANA Transition Chris Buckridge and Paul Rendek of the RIPE NCC presented background information on the U.S. Government’s announcement of its intention to transition out of its IANA functions oversight role. They noted that any proposal for a future model of IANA administration needs to come from a global, multi-stakeholder development process, and that RIPE and the RIPE NCC are key IANA stakeholders. They suggested that the RIPE community’s discussion of these issues should be centred around the Cooperation Working Group, with the RIPE NCC assisting in facilitating input to that process from regional events and voices. Rob Blokzijl warned that any process involving ICANN will necessarily be complicated and political. On a technical point, he noted that future RIPE NCC presentations on this subject should include the RIPE NCC’s reverse DNS interactions with IANA, which may be more regular than the number resource requests already noted. He further noted that the RIPE NCC’s direct interactions with the NTIA itself were non-existent, and stressed that the processes developed by RIPE and the other RIR communities already met the requirements laid out by the NTIA as necessary for oversight of the IANA functions. Danniel Karrenberg recalled that this is not the first time this discussion has taken place, and that the RIPE community has historically taken an active part in the discussion around the formation of ICANN and the organisation and delivery of IANA services. He stressed the legitimacy and credibility of the RIR community processes in policy-making and argued against over-complicating the situation. He also noted that the IANA is three distinct groups of functions (number resources, the DNS root zone and protocol parameters), and the RIPE community discussions should focus primarily on the number resource functions; if difficulties in defining governance processes for the DNS root zone threaten to derail the oversight transition process, the community should be explicitly prepared to propose unbundling those functions and taking oversight of the numbering functions. Malcolm Hutty disagreed with the perception that NTIA oversight was not important, and stressed that this oversight has protected the policy-making relationship with ICANN itself. He noted that policy regarding the DNS is determined by the ICANN community and imposed on registrars, meaning that users essentially have to submit to ICANN policies. RIPE and the RIR communities determine their own policies with regard to Internet number management, but it may be conceivable that ICANN would decide it wants to set these policies in future and impose them on the RIPE community. He argued that a credible external oversight function must be retained to prevent this. Nurani Nimpuno argued that the RIR communities should take ownership of this issue, as custodians of the Internet number resources, and that the community members should be contributing to the broader discussion, while maintaining a focus on the numbering functions. She also stressed that the communities should be pro-active in defining terms like “multi-stakeholder” and “openness” which appear in the NTIA requirements. Jim Reid agreed with Malcolm Hutty on the importance of preventing ICANN mission-creep, and on the need to tightly define the relationship between the IANA operator and the RIR communities. He also warned that achieving consensus on a community proposal may be difficult, and suggested that there should be a fallback position to allow for RIPE and the RIPE NCC to make a meaningful contribution to the global discussion. Jari Arkko noted the evolution that has occurred in how the IETF and IAB manage oversight of the protocol parameters and their relationship to IANA. He agreed that the RIPE community needs to take ownership of this and take charge of what needs to change or not change. Olaf Kolkman, also an active participant in the IETF, noted the efforts in the IETF to align on a principle-based approach, with the most important principle being that the IETF controls its own destiny. Ensuring that people are empowered to participate in these discussions will help the debate going forward, and the community needs to provide guidance - developing a set of principles may be a good first step. Olaf suggested RFC 6220 as a good starting point. Salam Yamout provided some perspective from the government side, particularly in the Arab world - notably the perception that the United States has control of the Internet, and governments’ strong focus on DNS-related issues. She noted that governments concerns centre primarily around ICANN. Phil Rushton urged the community to be aware of events in other forums, including the UN, WSIS and the ITU - while there is not the need for everyone to be directly involved, we need to be aware of what governments are thinking and the where they still need to be convinced by the RIR communities. Daniel Karrenberg argued that the community does not need another level of oversight for protection, and noted that the RIR communities already have solid agreements in place with ICANN, which ICANN cannot unilaterally change. He stressed the importance (and his optimism) of achieving community consensus on a proposal. He suggested that the RIPE community should avoid being drawn into the discussions about governance and oversight of DNS root zone management and stick to its purview of Internet number resources. Paul Wilson noted that the IANA functions comprise three quite different areas (numbers, protocol parameters and DNS), and that only one of these (the DNS) is controversial - the IAB has stated its readiness to take responsibility for the protocol parameters, and the RIR communities should also, in the very near future, be ready to make such a statement regarding the number functions. He and Adiel Akplogan agreed that strengthening the RIR processes, ensuring that they are consistent, clear, accessible and well documented, is vital. Sandy Murphy warned that the outcome of this process may impact our current model of Internet governance, and stressed the need for the RIR communities to have their voice heard. She also asked about the ICANN consultation timeline, specifically the call for comments on its proposed process, and whether this process is now set. Paul Rendek noted that there is expected to be more information on ICANN’s planning in time of the ICANN 50 Meeting, which takes place in London in June. Chris Buckridge also noted that all relevant information, including links to the relevant ICANN web, is posted on the ripe.net website. D. Interconnection • D1. The Internet, the Internets, and Splinternets – Peter Koch Peter Koch discussed the proposals coming from Germany for establishing a separate “German” Internet. Randy Bush noted the experience of the Saudi industry, after the regulator decided that no traffic between two Saudi users should leave the country, and pointed out that IXPs are a key element facilitating this. There was a question as to whether the German-only email system used the DNS (which would generate its own cross-border traffic). Peter noted that the one he mentioned uses special domains, but DNS leakage was of less concern than the actual content of the messages. Alain Van Gaever asked about the rate of take-up. Peter didn’t have figures to hand, but noted that there are incentives, it is early in the deployment, and the operators are targeting users of existing email services. Brian Nisbet said that while he can see what’s being attempted, it never succeeds and generally breaks things that the users want to do. Peter stressed that walking away from the discussion probably isn’t the right strategy, and that users often learn what they want from marketing campaigns. Expanding on the question of what users want, Meredith Whittaker noted that users want security, and the technical community needs to be a public voice stating that this is not the way to achieve that. Marco Davids noted an initative in the Netherlands using a closed user group in BGP - this doesn't combat traffic monitoring, but rather helps mitigate DDOS attacks, and if users (such as banks) are under attack the group can be closed to users outside the Netherlands. Olaf Kolkman asked whether any EU research funding had been channeled to this, and whether the project could lead to some sort of European standardisation. Peter noted that the work is based on IETF standards. Jean Jacque Sahel noted that the European Commission has publicly said that this is all a very bad idea. Jaap Akkerhuis recalled a proposal from Italy to establish a trusted network for digital mail. Olaf Kolkman and Peter Koch, summarising some of the discussion’s key points, suggested that we are moving intelligence to the core of the network, and the core is represented by big players. The technical community cannot just dismiss these initiatives - there are some laudable goals behind them, and we need to engage in the discussion. • D2. Interconnection: Russia, the EU, and Internet Cooperation and Governance – Igor Milashevskiy Igor Milashevskiy, representing the Russian government in its first RIPE Meeting, shared some perspectives on that government’s view of the Internet and related public policy. The Russian government sees the Internet as a driver of development, with the Russian Internet market the biggest in Europe - 68 million users, more than 56 million people use Internet every day, including a significant percentage outside big cities. The Russian language is also the second largest in Internet, there are slightly fewer than five million .ru domains and more than 800,000 .рф domains. Speaking from personal perspective, he noted that the RIPE NCC is a reference organisation in the Internet space, and the target is to restore trust and confidence to the Internet environment, and develop international tools for preventing improper use of the Internet. Ho noted that the main actor in the Intenet is the user, and if those users have certain rights in the offline world - access to information, privacy, secrecy of communication and freedom of opinion - we have to protect those rights online. The process to do this has just begun, and the Russian government believes there are no rights without duties, no freedom without responsibility. He also suggested that the role of governments in Internet governance needs to be recognised. NETmundial was a good and innovative attempt to include all stakeholders, but the outcome document doesn’t reflect all the contributions. Ciprian Nica, participating remotely, asked how and who should define the proper purposes of using the Internet. Igor explained that the Internet is a universal tool, and can be used for a wide range of purposes, but that its primary purpose should be to make users’ lives richer. Desiree Milosevich asked if there could be some elaboration on the Russian government’s issues with the NETmundial statement. Igor noted that this is in the public record of the Russian statements. Paul Rendek extended his thanks to Igor and the Russian government for their increased willingness to engage with the RIPE community and RIPE NCC, and welcomed Igor's attendance at the RIPE Meeting. E. Making the Internet a Little Bit Safer Cryptographically - Randy Bush Randy Bush discussed the development of an open public architecture for hardware security modules. The goal is a design (not a product) that is scalable, composable and assured. He stressed that the project needs people to audit the code. Aaron Kaplan asked where to get a development board. Randy noted that the boards are available for 170 USD. Aaron also noted that it would be a good idea to have the testing procedure online and publicly available. Eric Wuster agreed that this is good work and asked why go the FPGA route rather than using a small embedded chip. Randy noted that some of the applications need speed, particularly some of the encryption stuff. Regarding chips though, Randy noted that the FPGA Verilog was first done in Python, meaning there is a Python version and a Verilog version. Eric also asked about how the project is sourcing hardware random number generators. Randy replied that they are currently investigating this issue. F. Policy Radar • F1. RIPE NCC Updates, including NETmundial and IGF Developments – Chris Buckridge Chris Buckridge gave an update on the wide range of Internet governance events taking place in the coming months, including the ITU Plenipotentiary 2014 that is scheduled to take place in October. He highlighted the links between many of these events, particularly in terms of broader strategies. He noted suggestions from earlier sessions that the RIPE NCC provide more targeted information for the community, and reported that the RIPE NCC is investigating the best method for doing this. Phil Rushton noted that the technical community can have significant impact on events at the Plenipotentiary via their input to Member State delegations. He also noted that while NETmundial produced a good outcome for the multi-stakeholder model, other forums, such as the CSTD Working Group on Enhanced Cooperation, saw much less support for multi-stakeholder processes and governance. • F2. Co-chair Updates and Working Group Initiatives Meredith Whittaker closed the session by reiterating the goals of co-chairs, particularly the need to engage people who are affected by Internet governance and public policy issues, but don't currently take an active interest. She noted options such as producing white papers, using RIPE Meeting time for more workshop-style events, and other ideas for the working group to serve as a RIPE community “brains trust” for those involved in public policy discussions. Nurani Nimpuno noted her support for the co-chairs’ approach and stressed the need to bring discussions back to specific issues and make the topics practical for RIPE community participants. The co-chairs closed the meeting. --------------------- End of Minutes ---------------------------------

1 0

Minutes of the Coop-WG meeting during RIPE 68 in Warsaw - Part 2
by Alain Van Gaever 10 Jun '14

10 Jun '14

Dear all, The text in my mail yesterday was the summary of the IANA discussion - the full minutes of the working groups session are now below. My apologies for the confusion! Alain > > ------- Beginning of Minutes ------- > > Cooperation Working Group Draft Agenda > > Thursday, 15 May, 11:00 – 12:30 > > A. Administrative Matters > > The co-chairs opened the meeting. New co-chairs Meredith Whittaker and > Alain Van Gaever introduced themselves and briefly outlined their hopes for > the working group, specifically ensuring that the content is relevant and > of practical use to the RIPE community. > > The minutes from RIPE 67 were approved, as was the agenda for RIPE 68. > > > B. Content Blocking > > • B1. A Technical Overview of Content Blocking Methods – Pier Carlo > Chiodi, Olaf Kolkman > > Olaf Kolkman gave an overview of the work done by Pier Carlo Chiodi on > content blocking on the Internet and other work in the same area. Outlining > a number of strategies employed by those wishing to block content, he noted > that blocking is done most effectively at the end-point (or origin of the > content), and that blocking in the network involves a variety of trade-offs. > > Lars-Johan Liman noted that even “legitimate” blocking has collateral > damage, and pointed to the example of the hotel network, where the > interception of traffic prevents the use of DNSSEC. He suggested that in > such situations it is best if the effects of the blocking are stated > upfront for users. > > Andrei Robachevsky recalled a paper produced by ISOC (prepared around the > time of the SOCA/PIPA/ACTA legislative proposals), and noted that security > and hidden costs are all important, and public policy makers need to be > made aware of these negative impacts. Olaf pointed out that users will do > whatever they can to get to blocked content, and that this can also have > potential negative effects (including a greater viral footprint or exposing > backdoors to computer systems). > > Alexander Isavnin asked participants in the room who thought their > countries were doing blocking, and whether they thought it was being abused. > > Richard Barnes noted that these issues reinforce the importance of an > end-to-end strategy, and agreed that the hotel network is a good example of > how blocking can break important security elements like DNSSEC. > > Jim Reid noted that blocking access to specific content can mean that > broader services are blocked. He noted a case that he had provided advice > on, where one of the questions asked was "what would happen if we switched > off port 80 access on this particular domain?” - the only person who knows > is the webmaster of the domain, but that person is unlikely to happily > cooperate in the blocking of their domain. > > • B2. Telex: A Proposal For Circumventing Censorship in the Network – Eric > Wustrow > > Eric Wustrow outlined the Telex project, which has been developed to > circumvent content blocking measures. The system provides a means of > connecting users to blocked content via a mechanism that is invisible to > the censoring technology. He encouraged ISPs to contribute to the work with > advice and prototype deployment assistance. > > Robert Kisteleki noted that PGP key distribution may be a bottleneck in > the Telex system. Eric noted that the paper discusses some of these issues, > including preventing censors from distributing “bad” public keys. He > suggested that having a central Telex entity that is known and trusted will > be important to mitigate these risks. He also noted that getting > information into areas subjected to censorship is often less difficult than > might be supposed. > > • B3. Web Censorship Circumvention: Challenges and Opportunities – Walid > Al-Saqaf > > Walid Al-Saqaf outlined his project to map URL filtering via > crowdsourcing, which is developing longitudinal data, while allowing > contributing users to access blocked websites via his own servers. He > presented a range of data and analysis obtained from this work, including > content blocking methods and strategies. He noted the need to make people > more aware of the range of circumvention tools and solutions available, and > that speed, security and anonymity are all important to users. He also > described his plans for the future, including open-sourcing his own project > and cooperating with similar projects. > > Andrei Robachevsky asked whether the project looked at which means of > blocking were most common. Walid stated that he has done some analysis of > this, using the packet headers - the more data he can get, the better this > analysis will be. > > Meredith Whittaker noted that the Open Observatory of Network Interference > (OONI) project, coordinated by the Tor team, is also doing work in this > area and is generating public data. > > Alexander Isavnin suggested that a RIPE task force might be a useful > vehicle for RIPE community members interested in this issue. Walid agreed > that the technical community, and particularly its relationship to civil > society actors in this space, needs to be further explored and understood. > > C. IANA Transition > > Chris Buckridge and Paul Rendek of the RIPE NCC presented background > information on the U.S. Government’s announcement of its intention to > transition out of its IANA functions oversight role. They noted that any > proposal for a future model of IANA administration needs to come from a > global, multi-stakeholder development process, and that RIPE and the RIPE > NCC are key IANA stakeholders. They suggested that the RIPE community’s > discussion of these issues should be centred around the Cooperation Working > Group, with the RIPE NCC assisting in facilitating input to that process > from regional events and voices. > > Rob Blokzijl warned that any process involving ICANN will necessarily be > complicated and political. On a technical point, he noted that future RIPE > NCC presentations on this subject should include the RIPE NCC’s reverse DNS > interactions with IANA, which may be more regular than the number resource > requests already noted. He further noted that the RIPE NCC’s direct > interactions with the NTIA itself were non-existent, and stressed that the > processes developed by RIPE and the other RIR communities already met the > requirements laid out by the NTIA as necessary for oversight of the IANA > functions. > > Danniel Karrenberg recalled that this is not the first time this > discussion has taken place, and that the RIPE community has been vocal in > its preference for the US government stepping away from this oversight role > since the late 1990s. He stressed the success of the RIR community > processes in policy-making and argued against over-complicating the > situation. He also noted that the IANA is three distinct groups of > functions (number resources, the DNS root zone and protocol parameters), > and the RIPE community discussions should focus primarily on the number > resource functions; if difficulties in defining governance processes for > the DNS root zone threaten to derail the oversight transition process, the > community should be explicitly prepared to propose unbundling those > functions and taking oversight of the numbering functions. > > Malcolm Hutty disagreed with the perception that NTIA oversight was not > important, and stressed that this oversight has protected the policy-making > relationship with ICANN itself. He noted that policy regarding the DNS is > determined by the ICANN community and imposed on registrars, meaning that > users essentially have to submit to ICANN policies. RIPE and the RIR > communities determine their own policies with regard to Internet number > management, but it may be conceivable that ICANN would decide it wants to > set these policies in future and impose them on the RIPE community. He > argued that a credible external oversight function must be retained to > prevent this. > > Nurani Nimpuno argued that the RIR communities should take ownership of > this issue, as custodians of the Internet number resources, and that the > community members should be contributing to the broader discussion, while > maintaining a focus on the numbering functions. She also stressed that the > communities should be pro-active in defining terms like “multi-stakeholder” > and “openness” which appear in the NTIA requirements. > > Jim Reid agreed with Malcolm Hutty on the importance of preventing ICANN > mission-creep, and on the need to tightly define the relationship between > the IANA operator and the RIR communities. He also warned that achieving > consensus on a community proposal may be difficult, and suggested that > there should be a fallback position to allow for RIPE and the RIPE NCC to > make a meaningful contribution to the global discussion. > > Jari Arkko noted the evolution that has occurred in how the IETF and IAB > manage oversight of the protocol parameters and their relationship to IANA. > He agreed that the RIPE community needs to take ownership of this and take > charge of what needs to change or not change. > > Olaf Kolkman, also an active participant in the IETF, noted the efforts in > the IETF to align on a principle-based approach, with the most important > principle being that the IETF controls its own destiny. Ensuring that > people are empowered to participate in these discussions will help the > debate going forward, and the community needs to provide guidance - > developing a set of principles may be a good first step. Olaf suggested RFC > 6220 as a good starting point. > > Salam Yamout provided some perspective from the government side, > particularly in the Arab world - notably the perception that the United > States has control of the Internet, and governments’ strong focus on > DNS-related issues. She noted that governments concerns centre primarily > around ICANN. > > Phil Rushton urged the community to be aware of events in other forums, > including the UN, WSIS and the ITU - while there is not the need for > everyone to be directly involved, we need to be aware of what governments > are thinking and the where they still need to be convinced by the RIR > communities. > > Daniel Karrenberg argued that the community does not need another level of > oversight for protection, and noted that the RIR communities already have > solid agreements in place with ICANN, which ICANN cannot unilaterally > change. He stressed the importance (and his optimism) of achieving > community consensus on a proposal. He again suggested that the RIR > communities should make every effort to unlink the number-related IANA > functions from the DNS. > > Paul Wilson noted that the IANA functions comprise three quite different > areas (numbers, protocol parameters and DNS), and that only one of these > (the DNS) is controversial - the IAB has stated its readiness to take > responsibility for the protocol parameters, and the RIR communities should > also, in the very near future, be ready to make such a statement regarding > the number functions. He and Adiel Akplogan agreed that strengthening the > RIR processes, ensuring that they are consistent, clear, accessible and > well documented, is vital. > > Sandy Murphy warned that the outcome of this process may impact our > current model of Internet governance, and stressed the need for the RIR > communities to have their voice heard. She also asked about the ICANN > consultation timeline, specifically the call for comments on its proposed > process, and whether this process is now set. Paul Rendek noted that there > is expected to be more information on ICANN’s planning in time of the ICANN > 50 Meeting, which takes place in London in June. > > Chris Buckridge also noted that all relevant information, including links > to the relevant ICANN web, is posted on the ripe.net website. > > D. Interconnection > > • D1. The Internet, the Internets, and Splinternets – Peter Koch > > Peter Koch discussed the proposals coming from Germany for establishing a > separate “German” Internet. > > Randy Bush noted the experience of the Saudi industry, after the regulator > decided that no traffic between two Saudi users should leave the country, > and pointed out that IXPs are a key element facilitating this. > > There was a question as to whether the German-only email system used the > DNS (which would generate its own cross-border traffic). Peter noted that > the one he mentioned uses special domains, but DNS leakage was of less > concern than the actual content of the messages. > > Alain Van Gaever asked about the rate of take-up. Peter didn’t have > figures to hand, but noted that there are incentives, it is early in the > deployment, and the operators are targeting users of existing email > services. > > Brian Nisbet said that while he can see what’s being attempted, it never > succeeds and generally breaks things that the users want to do. Peter > stressed that walking away from the discussion probably isn’t the right > strategy, and that users often learn what they want from marketing > campaigns. Expanding on the question of what users want, Meredith Whittaker > noted that users want security, and the technical community needs to be a > public voice stating that this is not the way to achieve that. > > Marco Davids noted an initative in the Netherlands using a closed user > group in BGP - this doesn't combat traffic monitoring, but rather helps > mitigate DDOS attacks, and if users (such as banks) are under attack the > group can be closed to users outside the Netherlands. > > Olaf Kolkman asked whether any EU research funding had been channeled to > this, and whether the project could lead to some sort of European > standardisation. Peter noted that the work is based on IETF standards. Jean > Jacque Sahel noted that the European Commission has publicly said that this > is all a very bad idea. > > Jaap Akkerhuis recalled a proposal from Italy to establish a trusted > network for digital mail. > > Olaf Kolkman and Peter Koch, summarising some of the discussion’s key > points, suggested that we are moving intelligence to the core of the > network, and the core is represented by big players. The technical > community cannot just dismiss these initiatives - there are some laudable > goals behind them, and we need to engage in the discussion. > > > • D2. Interconnection: Russia, the EU, and Internet Cooperation and > Governance – Igor Milashevskiy > > Igor Milashevskiy, representing the Russian government in its first RIPE > Meeting, shared some perspectives on that government’s view of the Internet > and related public policy. The Russian government sees the Internet as a > driver of development, with the Russian Internet market the biggest in > Europe - 68 million users, more than 56 million people use Internet every > day, including a significant percentage outside big cities. The Russian > language is also the second largest in Internet, there are slightly fewer > than five million .ru domains and more than 800,000 .рф domains. > > Speaking from personal perspective, he noted that the RIPE NCC is a > reference organisation in the Internet space, and the target is to restore > trust and confidence to the Internet environment, and develop international > tools for preventing improper use of the Internet. > > Ho noted that the main actor in the Intenet is the user, and if those > users have certain rights in the offline world - access to information, > privacy, secrecy of communication and freedom of opinion - we have to > protect those rights online. The process to do this has just begun, and the > Russian government believes there are no rights without duties, no freedom > without responsibility. > > He also suggested that the role of governments in Internet governance > needs to be recognised. NETmundial was a good and innovative attempt to > include all stakeholders, but the outcome document doesn’t reflect all the > contributions. > > Ciprian Nica, participating remotely, asked how and who should define the > proper purposes of using the Internet. Igor explained that the Internet is > a universal tool, and can be used for a wide range of purposes, but that > its primary purpose should be to make users’ lives richer. > > Desiree Milosevich asked if there could be some elaboration on the Russian > government’s issues with the NETmundial statement. Igor noted that this is > in the public record of the Russian statements. > > Paul Rendek extended his thanks to Igor and the Russian government for > their increased willingness to engage with the RIPE community and RIPE NCC, > and welcomed Igor's attendance at the RIPE Meeting. > > > E. Making the Internet a Little Bit Safer Cryptographically - Randy Bush > > Randy Bush discussed the development of an open public architecture for > hardware security modules. The goal is a design (not a product) that is > scalable, composable and assured. He stressed that the project needs people > to audit the code. > > Aaron Kaplan asked where to get a development board. Randy noted that the > boards are available for 170 USD. Aaron also noted that it would be a good > idea to have the testing procedure online and publicly available. > > Eric Wuster agreed that this is good work and asked why go the FPGA route > rather than using a small embedded chip. Randy noted that some of the > applications need speed, particularly some of the encryption stuff. > Regarding chips though, Randy noted that the FPGA Verilog was first done in > Python, meaning there is a Python version and a Verilog version. > > Eric also asked about how the project is sourcing hardware random number > generators. Randy replied that they are currently investigating this issue. > > F. Policy Radar > > • F1. RIPE NCC Updates, including NETmundial and IGF Developments – Chris > Buckridge > > Chris Buckridge gave an update on the wide range of Internet governance > events taking place in the coming months, including the ITU Plenipotentiary > 2014 that is scheduled to take place in October. He highlighted the links > between many of these events, particularly in terms of broader strategies. > He noted suggestions from earlier sessions that the RIPE NCC provide more > targeted information for the community, and reported that the RIPE NCC is > investigating the best method for doing this. > > Phil Rushton noted that the technical community can have significant > impact on events at the Plenipotentiary via their input to Member State > delegations. He also noted that while NETmundial produced a good outcome > for the multi-stakeholder model, other forums, such as the CSTD Working > Group on Enhanced Cooperation, saw much less support for multi-stakeholder > processes and governance. > > • F2. Co-chair Updates and Working Group Initiatives > > Meredith Whittaker closed the session by reiterating the goals of > co-chairs, particularly the need to engage people who are affected by > Internet governance and public policy issues, but don't currently take an > active interest. She noted options such as producing white papers, using > RIPE Meeting time for more workshop-style events, and other ideas for the > working group to serve as a RIPE community “brains trust” for those > involved in public policy discussions. > > Nurani Nimpuno noted her support for the co-chairs’ approach and stressed > the need to bring discussions back to specific issues and make the topics > practical for RIPE community participants. > > The co-chairs closed the meeting. --------------------- End of Minutes --------------------------------- On Wed, Jun 4, 2014 at 7:58 PM, Alain Van Gaever <avangaev(a)gmail.com> wrote: > Dear all, > > Please find below the minutes of the Coop-WG meeting during RIPE 68 in > Warsaw. > WG members are welcome to suggest further edits or changes. > > Hope to see you all in London for RIPE-69 ! > > Meredith, Maria & Alain > Co-Chairs of the Coop-WG > > > PS Thanks for Chris for producing the notes! > > ---- beginning of minutes ---- > > Location: Warsaw, Poland > > Attendance: Approximately 150 > > - Minutes of the session > - Full transcript of the session > <https://ripe68.ripe.net/archives/steno/39/> > - Video of the session and information slide-pack > <https://ripe68.ripe.net/archives/video/216/> > > Major points coming out of the discussion: > > 1. The RIR communities need to assert their ownership of issues > regarding the distribution and registration of Internet number resources. > This transition is an opportunity to more solidly formalise that ownership, > with minimal change to the existing policy-making and operational processes. > 2. While the IANA functions (as a bundle) present a number of complex > issues, identifying a future model for the IANA Internet number registry > functions should be straightforward. The policy-making and operational > processes relating to the IANA Internet number registry functions are solid > and have been in place for many years, and they have never included an > explicit oversight role for the NTIA. > 3. The Regional Internet Registries must ensure that their processes > and policies are clearly defined, well documented, transparent and > accessible. > 4. While the RIPE NCC will facilitate discussion of these issues > throughout the service region (including at community regional events), the > RIPE Cooperation Working Group will serve as the central venue for RIPE > community discussion and development of any proposal relating to the future > of the IANA functions. > > > > Some additional points made during the discussion: > > - Speakers reported on the progress of the Internet Architecture Board > (IAB) and the Internet Engineering Task Force (IETF) in establishing their > ownership and authority over protocol parameter registries maintained by > IANA. > - Several speakers noted the dangers arising from this process, > particularly the potential for an outcome that does not solidly define and > protect the community-driven, bottom-up control and development of IANA > policy. > - Several speakers noted the interest that governments throughout the > world are taking in this process and that government voices will be a > factor in the final outcome. > - Several speakers stressed the importance of reaching RIPE community > consensus on a proposal or position, with this process potentially seen as > a test-case for bottom-up policy making. > - Several speakers argued strongly that any RIR proposal should aim to > separate the IANA number management functions from oversight of the other > IANA functions. > > > > --------- End of minutes ------ >

3 2

IANA discussions: ENOG 7
by Chris Buckridge 06 Jun '14

06 Jun '14

Dear colleagues, Following on from Alain’s email, I would like to note a couple of items relating to discussions of the IANA oversight transition in the RIPE community. The Eurasian Network Operators' Group (ENOG) held its seventh meeting in Moscow on 26-27 May, the week after RIPE 68. There was a discussion of the IANA oversight transition at that meeting, and a brief summary of that discussion is provided below: > ENOG 7, 27 May 2014 > Attendance: Approximately 150 > > Major points coming out of the discussion: > - Additional ways in which the RIPE NCC interacts with the IANA include reverse DNS and running the K-root name server. > - It is important to note that the oversight function will not transition to a government-based body. > - The technical community has not had any issues with the U.S. government's conduct of its oversight of these functions for the past 20+ years - any new model should not complicate the current system and processes. > - The RIPE NCC is not interested in DNS operations and should not be put under that umbrella. > - The community should ensure that operators are protected from possible commercial interests, given that some of ICANN’s activities, such as the registration of domain names, are profit-making. As noted in the RIPE 68 session, the RIPE Cooperation Working Group is the official venue for community discussion of this issue. For this reason, the RIPE NCC will ensure that any discussion in other RIPE community forums is fed back to this list. Summaries of these discussions will also be posted on the RIPE website: http://ripe.net/iana-discussions Any comments or responses to this process or on the substance of the summaries should be posted to this mailing list. Best regards, Chris Buckridge Senior External Relations Officer, RIPE NCC

4 5