IP: Next Generation Area Directors: o Scott Bradner: sob@harvard.edu o Allison Mankin: mankin@cmf.nrl.navy.mil Area Summary reported by Allison Mankin/NRL The IPNG Area co-Directors gave a plenary presentation on their plan for the IPng decision process and introduced the IPng directorate. Frank Solensky gave an introduction to the ALE BOF and the three IPng proposals gave status reports during the IPng plenary session; Peter Ford gave a status report on TUBA, Steve Deering gave an overview of SIPP, and Rob Ullmann gave an overview of TP/IX (also known as CATNIP). Address Lifetime Expectations BOF (ALE) Phill Gross gave an update to the presentation he and Dennis Ferguson prepared for INET '93 describing the growth of the Internet (in terms of both assigned addresses and connected networks) and presented some recommendations for increasing the efficiency of how IP addresses are deployed. A lively discussion ensued. The working group will be formed, combining resources with CIDR deployment. The emphasis will be on the measurement and projections, evaluating the potential impact of recommendations rather than formulating recommendations itself. There is also a pressing need to collect more information; all known projections are based on incomplete data. P. Internet Protocol Working Group (PIP) and Simple Internet Protocol Working Group (SIP) The PIP and SIP Working Groups have combined their efforts and the working groups will be merged into a new working group called Simple Internet Protocol Plus (SIPP). The two working groups met in two combined sessions co-chaired by Steve Deering, Paul Francis, and Bob Hinden. At the first session Steve Deering presented an overview of the SIP/PIP Merger. This included the motivation behind the merger, benefits of the merger, and described the new features of SIPP. The purpose of the merger is to keep the simplicity and transition features of SIP and to benefit from the advanced routing capabilities of Pip---while making them easier to use and to understand. Following this Paul Francis presented the SIPP routing and addressing. This included a description of address sequences and how they are used for mobility, provider selection, and extended addressing. Ramesh Govindan presented detailed examples of these usages of SIPP address sequences. A overview of the new IPAE draft was given by Bob Gilligan. He gave a short overview of IPAE, and discussed and resolved several open issues. Bill Simpson presented the current state of his work on SIPP neighbor discovery. It focuses on a ``where are you'' and ``I am Here'' functions with optional extensions for additional functionality. During the second session Rob Coltun presented his proposal for a version of OSPF for SIPP. The group concluded that he should focus on just extending OSPF to support 64-bit addresses and defer the work to add additional levels of hierarchy. The latter work should be presented to the Open Shortest Path First IGP Working Group (OSPF). Sue Thompson presented her proposal for DNS changes to support SIPP. The group concluded that this was the correct approach for SIPP DNS records. Jim Bound presented his thoughts on the changes required to the Dynamic Host Configuration Protocol. There was an extended discussion which resulted in general agreement that auto configuration was a key part of any IPng. Paul Francis presented a proposal for provider based address assignment. After an interesting discussion, the group agreed to proceed with this approach. TCP/UDP Over CLNP-Addressed Networks Working Group (TUBA) Dave Marlow reported that the CLNP Multicast work has made progress in ISO. Changes exist in addressing, CLNP, ES-IS and the network service definition. Group addressing is a full standard, other changes are in ballot at this time. Ross Callon discussed the revised NSAP Addressing Guidelines document and took an action item to make the document somewhat less ``backbone-centric.'' CLNP mobility was discussed. Mark Knopper briefly described CDPD, a specification for cellular mobile data service from a consortium of cellular carriers. It uses CLNP as the primary protocol, and provides IP service using IP-over-CLNP encapsulation. The mobility protocol is similar to ongoing work in the IP Routing for Wireless/Mobile Hosts Working Group (MOBILEIP). The group discussed whether or not it should be proactive, or wait for the MOBILEIP Working Group to settle. Yakov Rekhter and Dave Piscitello agreed to recast the mobile IP document in terms of CLNP and publish it as an Internet-Draft. Yakov Rekhter described his work on a method for transparently adding options to CLNP. It codes which options are required to be processed by routers and/or hosts, even when the option is otherwise unrecognized. He also described work on strong versus weak QoS forwarding. Dave Katz spoke about the outcome of the Extensions to OSI for use in the Internet BOF (OSIEXTND) that was held in Amsterdam. The net effect of IESG policies is that the work will progress within the TUBA Working Group. Dave Katz then presented an extension to the standard dynamic NSAP address assignment function, which would allow the end system to suggest a system ID for itself. Peter Ford presented his draft document on the Dual Stack Transition plan. It is an ``inside out'' approach that begins with infrastructure deployment. It was pointed out that this transition framework needs to be completed as soon as possible. TCP/UDP Over CLNP-Addressed Networks Working Group (TUBA) and TP/IX Working Group (TPIX) The TUBA Working Group met in joint session with the the TPIX Working Group. TPIX then continued on to a separate session in the same room. Robert Ullmann presented his new proposal Common Architecture For The New Internet (CATNIP). The new proposal is based on RFC 1475. CATNIP is designed to use header compression by including a flow cache ID or "handle" in its header. It also uses a NSAP style of addressing. The joint meeting was held to explore commonality between CATNIP and TUBA proposals. The group came up with the following list of milestones: o Submit the CATNIP proposal as an Internet-Draft o Rewrite the TPIX Working Group charter to realign it with the new proposal o Possibly rename the TPIX Working Group to the CATNIP Working Group