[IMR] IMR87-12.TXT Westine [Page 1]  ~ DECEMBER 1987 INTERNET MONTHLY REPORTS ------------------------ The purpose of these reports is to communicate to the Internet Research Group the accomplishments, milestones reached, or problems discovered by the participating organizations. This report is for research use only, and is not for public distribution. Each organization is expected to submit a 1/2 page report on the first business day of the month describing the previous month's activities. These reports should be submitted via network mail to Ann Westine (Westine@ISI.EDU) or Karen Roubicek (Roubicek@NNSC.NSF.NET). BBN LABORATORIES AND BBN COMMUNICATIONS CORPORATION --------------------------------------------------- WIDEBAND NETWORK Connectivity was established between a Wideband Butterfly Gateway installed at the Naval Ocean Systems Center and the BSAT packet switch located at M/A-COM Government Systems, Inc. These two San Diego, CA sites are connected via a T1-rate terrestrial communication circuit. Difficulties with the installation of this circuit had delayed the establishment of the NOSC-Wideband connection, which can now provide Internet access for hosts to be installed on a NOSC Ethernet. Westine [Page 1]  Internet Monthly Report December 1987 The Wideband Network supported a multi-site demonstration of the Cronus Distributed Operating System which was presented at the Rome Air Development Center on December 9. Wide area communication through the Wideband network was a prominent part of the presentation, which was given as part a RADC Technology Exchange meeting. BSAT Release 6.3 was distributed to the Wideband Network sites this month. This release increases the number of requests for future satellite channel capacity that can be sent within a single satellite channel burst. It also removes a previously existing limitation on the number of static multicast group members that can be interfaced to a given BSAT. SATNET Although the SATNET performance was generally good, there were two major problems in the month of December. The Goonhilly SIMP began adversely affecting both satellite channels causing a loss of connectivity with Europe. The PSP terminal was reset and a tripped breaker was restored. We also experienced high error rates on the phone line to DCEC. Because of a misunderstanding, the local phone company began working on the line to CSS. This resulted in periodic losses of connectivity to the Arpanet. Service was restored to CSS and the problem with the line to DCEC was resolved. Even with these problems, the availability of the channel still averaged 98% from tests run by ISI. A patch has been tested in the SATNET testbed which approximately doubles the throughput on the two satellite channels. The tests were made with the SIMP message generator and will be verified in the operational SATNET. There still appears to be a throughput bottle neck at the host interface of the SIMP, progress is being made to eliminate it as well. We plan to distribute software with the new channel code at the SATNET meeting in January. INTERNET RESEARCH & DEVELOPMENT During December we tested a patch for the LSI-11 Gateways to allow them to send more datagrams into the Arpanet by using a different handling type in the 1822 AHIP message when the RFNM limit (8) was reached. This had the hoped for result of greatly reducing the number datagrams dropped by the gateways, in some cases by two orders of magnitude! It had the unexpected result (in the Arpanet under the New End-End protocol) for X.25 hosts of causing the PSN's to multiple identical (from the hosts point of view) X.25 virtual circuits to be opened to the hosts. This caused problems for some X.25 hosts which were not prepared to accept duplicate virtual circuits. Westine [Page 2]  Internet Monthly Report December 1987 We believe the patch results in a very significant improvement to overall Internet performance and we would like to see it installed in the LSI-11 gateways, a patch written for the Butterfly Gateways, and similar changes be made to all gateways. We think it is worth while the for X.25 hosts to change to be able accept multiple virtual circuits. Overall service would be much improved. Please send me your comments (hinden@bbn.com). Bob Hinden (Hinden@BBN.COM) ISI --- Internet Concepts Project Jon Postel and Paul Mockapetris attended the 2nd TCP/IP Interoperability conference in Crystal City, VA., Dec 1-4. Paul organized a session on Domain Practices. Paul Mockapetris visited BBN Dec 17-18 to attend the Internet Architecture Task Force meeting. Paul distributed a new version of root nameserver to TOPS-20 sites, to implement distribution of SOAs for negative caching and to fix problem related to distribution of address information for root servers. Two RFCs were published this month. RFC 1036: Horton, M., and R. Adams, "Standard for Interchange of USENET Messages, December 1987. RFC 1037: Greenburg, B., and S. Keene, "NFILE - A File Access Protocol, December 1987. Ann Westine (Westine@ISI.EDU) Multimedia Conferencing Project We are still awaiting delivery of the Concept Communications IMAGE 30 video codec. In the meantime, we are planning out the necessary changes to the packet video software required to accommodate the new codec, and implementing those which can be applied to the current experimental codec as well. Steve Casner (Casner@ISI.EDU) Westine [Page 3]  Internet Monthly Report December 1987 Brian Hung is working on an assembly language program for the NEC uPD7720 signal processing chip to implement echo cancelling function. The purpose of this is to eliminate the echos generated at the local 4-to-2 wire telephone hybrid. Brian Hung (Hung@ISI.EDU) NSFNET Project Bob Braden and Annette DeSchon completed the first release of the NSFNET statistics gathering package NNStat. A beta-test version of the acquisition program statspy was first installed on a Sun at Michigan by Hans-Werner Braun, so we could verify the ability to collect data from it remotely. Word of the program spread rapidly, and a number of sites requested the first release. We set up a mailing list of interested people, which now numbers over a dozen. We have been kept busy fixing bugs and enhancing the capabilities of the package. This resulted in a second release on December 31, and another scheduled for early next month. Bob Braden attended the Second TCP/IP Interoperability Workshop in Crystal City, Virginia, chaired a session on higher-level protocols, and gave a talk entitled "Internet End-to-End Protocols -- Progress and Prospects". Bob Braden worked on a joint document with Bob Cooper, Peter Kirstein, and Roland Rosner of the UK, proposing to link the academic networks in the US and the UK. In particular, the proposal will lease a 64kbit transatlantic circuit linking NSFNET in the US to JANET, the UK academic X.25 network. This circuit is to carry IP encapsulated in X.25. At the UK end, there will be a MicroVax running a clone of the UCL application-level gateway; that is, it will relay between SMTP and Grey-Book mail, between FTP and NIFTP/FTAM file transfers, and between Telnet and XXX terminal service. The plan is to connect to JANET at the University of London Computer Center and to NSFNET at JVNC. Bob Braden & Annette DeSchon (DeSchon@ISI.EDU, Braden@ISI.EDU) Supercomputer and Workstation Communication Project Alan Katz continued working on using GNU Emacs as a split editor. Most, if not all of the functions to do this are being written in GNU Emacs Lisp. Alan Katz (Katz@ISI.EDU) Westine [Page 4]  Internet Monthly Report December 1987 During December, the packet-echoing and packet-dropping feature of PVP was tested for correctness in a three-way (i.e. more than two-way) connection. During testing, a new mode of packet-disordering was discovered in the Wideband network. In an effort to investigate this, an unrelated bug was found in PVP having to do with passing virtual address pointers between nodes. D.J. Walden (Walden@ISI.EDU) MIT-LCS ------- No internet related progress to report. Lixia Zhang (Lixia@XX.LCS.MIT.EDU) MITRE Corporation ----------------- Nothing to report this month. Ann Whitaker (Whitaker@Gateway.Mitre.Org) SRI --- Internet Research I attended IETF Open Routing Working Group meeting on December 3-4, at Proteon, Inc., Westborough, Mass. The Meeting was devoted to the continuing preparation of the document "Requirements for Inter-Autonomous Systems Routing". Zaw-Sing Su UCL --- UCL is involved in a number of OSI acitivities. These are seen as an important indirect component of the DARPA activity, and are reported at intervals. This is a summary of the ongoing activity. There is ongoing work to develop an OSI infrastructure at UCL. This means provision of OSI communication protocols and tools to facilitate development of OSI applications. It is hoped that during 1988, these experimental systems will be used as a basis for Westine [Page 5]  Internet Monthly Report December 1987 pilot OSI services and for research into OSI applications. Most work has been on the ISODE package, developed by M.T. Rose and D.E. Cass at Northrop Research and Technology Center. UCL acts as the European distribution point for this package. The basic OSI services are support of the Transport and Session layers. The ISODE package was implemented with Transport class 0 over TCP/IP, and work at UCL has extended this to operate over X.25. A particular problem has been to provide a fully asynchronous OSI application interface over synchronous X.25 interfaces. The disadvantages of synchronous network interfaces were painfully emphasised by this work. An OSI support environment at UCL is available on the majority of the UNIX systems at UCL. This means SUN, VAX, and Pyramid hardware. There is also an experimental system operating of "Pink Book" (the UK Academic Community defined mapping of X.25 onto Ethernet LLC2). This will be of more interest when we obtain a second piece of pink book hardware! We have been looking at the performance of these systems at various levels, and hope to report in more detail later. There are three separate, but related, X.400 activities. The first project was to develop an X.400 to JNT Mail protocol converter based on the EAN X.400 system from University of British Columbia. This project was funded by Alvey. Work was done to increase the robustness and flexibility of the system. It was brought into a state where it could be operated by a group which did not need to understand its internal operation. It was operated as a pilot service at UCL, and then was transferred to University of London Computer Centre (ULCC), where it is now operating as a UK Academic Community Service. This is an important step towards making mail a full service: Previous UK mail gateway facilities have only been operated by their developers. The same system has also been used for a range of X.400 interworking trials. UCL has been involved in a number of interconnection exercises with various manufacturers, including user trials with British Telecom services. Some of these connections, for example with the French Academic Network, have developed into experimental services. The department uses these links for communication within several projects. Given the experience gained whilst developing the EAN Gateway, the group is now working on a more advanced protocol conversion facility. This system, "PP", will be a generic X.400 system which supports RFC 822 based (text) mail, X.400(1984) and X.400 (1988). By storing messages as directory trees, it will be able to handle Multimedia messages in a flexible manner, and perform a wide range of protocol conversions (e.g. JNT Mail <-> X.400) and format conversions (e.g. G3 Fax <-> SUN bitmap). Work is well underway on this system, and it is expected to be operational during 1988. It Westine [Page 6]  Internet Monthly Report December 1987 is intended to use this system as the basis for a range of experiments and pilot services. Code will be made available to other research and academic groups, and its availability will be announced in this forum. We are working on a number of aspects of directory services. This ranges from a system analogous to an online white pages telephone directory, to a system for high speed lookup of OSI information. It is generally agreed that such services are urgently required, but there is very little experience in this area. The THORN project (The Obviously Required Nameserver) is an ESPRIT project to develop a Directory Service aligned to the emerging standards. There are a number of industrial and academic partners: Olivetti; Bull; GEC; ICL; Siemens; SW; CERN; DFN; INRIA. These groups are working to develop a pre-competitive implementation of a directory service, in line with the emerging standards. The first phase of the project has produced an implementation according to ECMA TR 32, with simple replication facilities specified by the project. UCL is working on two aspects of this project. The first is to organise and participate in what is described as the Large Scale Pilot Exercise. This utilizes the system produced in the first phase of the project, and aims to gain real experience with the usage of directory services. An experimental service will be provided within UCL and in other selected parts of the UK Academic Community. This will interact with similar communities in France and Germany. It is hoped to examine a number of different uses of the directory service, with support of electronic mail as a major consideration. This system is just being released within UCL, and contains several thousand entries (people and hosts). In the context of THORN, and of another Esprit project (INCA - the Integrated Network Communication Architecture), UCL is working towards the development of X.500 directory services. Steve Kille & Jon Crowcroft (jon@CS.UCL.AC.UK) UNIVERSITY OF DELAWARE ---------------------- 1. For the last eighteen months I have been watching the NSFNET Backbone fuzzballs and gateways, learning from bizarre bogons, menacing martians, ominous overloads, rotten routing and all the other things that thrive in a healthy, growing Internet. Over the last few months massive fuzzware changes were made to cope with the recent explosion in traffic, number of nets and frequency of routing spasms. I have come to the conclusion that further improvements in the areas of throughput and robustness are not in the best interests of my own research projects and cannot be justified in view of the impending Westine [Page 7]  Internet Monthly Report December 1987 replacement of the fuzzballs scheduled for July. I will continue to evolve the fuzzball in other areas as a highly useful research tool and network surgical instrument. 2. One of the goals during this latest upgrade in fuzzball functionality was to implement and evaluate a priority- queueing service discipline in order to improve response times for interactive services like TELNET at possible expense of bulk-transfer services like FTP. The new discipline was installed at all NSFNET Backbone sites and most other sites early in the month. Subjective reports, confirmed by measurements under typical traffic conditions, indicate that TELNET traffic across the NSFNET Backbone has about half the delay and variance of non-TELNET traffic. However, since this effectively pushes non-TELNET traffic further back in the queues where quench and preemption are more likely, more study and possible modification of the fairness mechanisms may be necessary. 3. In preparation for the leap-second which ended 1987 and in despair of the fact I have spent the first moments of the last seven years advancing the year in all fuzzballs known to me, the fuzzball timekeeping code was extensively revised. The result is an automatic year rollover and automatic leap-second insertion via the Network Time Protocol (NTP) and fuzzball local-net protocols. Early on 31 December the leap-indicator bits were set at selected fuzzball hosts equipped with WWVB and WWV radio clocks. These bits were then automatically distributed to all NTP-speaking hosts and then further distributed to other hosts supporting the fuzzball local-net protocols. Upon the leap, all fuzzballs at U Delaware, U Maryland, ISI and Linkabit inserted the extra second and sailed accurately into 1988. The next step is to modify the firmware in the radio clocks themselves so they don't go cuckoo and have to resynchronize following the leap. 4. To see how the general Internet population coped with the leap second and in order to update data collected in 1983, a glorious three-day experiment was carried out with the approximately 5000 hosts and gateways listed in the NIC HOSTS.TXT file. Four packets were sent at five-second intervals to each host using UDP/NTP, UDP/TIME and ICMP/TIMESTAMP protocols and the clock offsets (in milliseconds) for each protocol averaged with respect to local time, which is synchronized via NTP to a radio-clock host. There were 46 hosts responding to NTP, 1158 responding to TIME and 1963 responding to ICMP requests. The distribution functions for each are shown below. More discussion in another place at another time. Westine [Page 8]  Internet Monthly Report December 1987 Decile NTP UDP ICMP ---------------------------------------------- 1 11 4632 5698 2 37 18238 27965 3 66 38842 68596 4 177 68213 127367 5 364 126232 201908 6 567 195950 285092 7 3466 267119 525509 8 20149 422129 2.91426E+06 9 434634 807135 5.02336E+07 10 1.17971E+09 1.59524E+09 2.11591E+09 5. Mike Minnich and Dave Mills attended the INARC Workshop, which was held at BBN on 17-18 December. See the INARC report elsewhere for further information. Dave Mills attended an NRC meeting on survivable networks in Washington on 8-9 December and presented a tutorial Advanced Topics in Internetworking at the Second TCP/IP Interoperability Conference on 1 December. Dave Mills (Mills@UDEL.EDU) Westine [Page 9]  Internet Monthly Report December 1987 NSF NETWORKING -------------- UCAR/BBN LABS NNSC Craig Partridge led a meeting on network management at the Second TCP/IP Interoperability Conference in Washington. Dick Edmiston represented the NNSC at the NSF meeting where participants discussed NSFNET transition issues. Karen Roubicek (roubicek@NNSC.NSF.NET) NSFNET BACKBONE SITES CORNELL UNIVERSITY THEORY CENTER November and December have been spent primarily stabilizing the NSFNET backbone network through software changes, in the fuzzballs, the surrounding gatedaemons, and the tables for each. The backbone currently carries traffic and routing information for about 105 networks - the actual number varies, of course. Here are Doug Elias's totals for November and December. Details are available upon request. ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ NSFNET TRAFFIC REPORT Period: Nov 1 - 30, '87 Total Traffic Figures Between Sites Ethernet Input 111381608 61279840 Output 105722051 55124918 In+Out 217103659 116404758 ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ NSFNET TRAFFIC REPORT Period: Dec 1 - 31, '87 Total Traffic Figures Between Sites Ethernet Input 110551496 62241199 Output 107690883 56274351 In+Out 218242379 118515550 ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ UNIVERSITY OF ILLINOIS AT URBANA-CHAMPAIGN Line to Minnesota Research Network came up as planned. The connection between P4200's is humming along. The gated configurations allowing their sqawking to NSFNET came along mid- December. Link to IMP94 is still no joy. Westine [Page 10]  Internet Monthly Report December 1987 T1 connection to UofI - Chicago is in but needs DSU/CSU's. This will replace various connections currently in place in the Chicago area (NW, U of Chicago, Argonne NL). May also be connection to Fermi, but that is not settled. by Ed Krol (krol@uxc.cso.uiuc.edu) JOHN VON NEUMANN NATIONAL SUPERCOMPUTER CENTER No report received. NATIONAL CENTER FOR ATMOSPHERIC RESEARCH AND UNIVERSITY SATELLITE NETWORK PROJECT (USAN) The Braden/DeSchon NNstat/STATSPY code has been installed on windom.ucar.edu and has already proved very useful not only in determining traffic patterns on USAN, but in detecting a network that was flooding USAN with rwho packets. Some statistics over a 72 hour period show the following: Total IP Packets (3.1 M). Total Packets (4.5M) IP Packets destined for UCARnet (47%). Pass-through Packets (53%) Top nine Source/Destination (S/D) IP nets account for (50%) of traffic. Top 3 S/D IP nets (34%) Top S/D IP net (15%) is 128.138 (University of Colorado) by Don Morris (morris@scdsw1.ucar.edu) PITTSBURGH SUPERCOMPUTING CENTER No report received. SAN DIEGO SUPERCOMPUTER CENTER Our PSN's first trunk line (to USC/ISI), installed on 29 October, has had the last 150 ft. at ISI installed. We are now awaiting the Telco folks to come test that last piece. The second line to UCLA, service date 4 Dec, is still not yet in sight. We are still looking forward to 7.4, along with an additional 0.5 meg, for our Proteon p4200. We have had no change in our line configuration this past month. In January, we expect to bring up a 56k line to UC Irvine to connect their campus network to NSFNET. Paul Love (loveep@sds.sdsc.edu) Westine [Page 11]  Internet Monthly Report December 1987 NSFNET REGIONAL AFFILIATED & CONSORTIUM NETWORKS BARRNET (No report received) JVNCNET (Refer to JVNNSC backbone report) MERIT Merit was recently awarded a cooperative agreement by the National Science Foundation for the upgrading, expansion and management of the NSFNET backbone network. This award extends over the next five years. Our proposal includes the extending of the current backbone to thirteen from the current six sites and using T1 lines to gain higher bandwidth. The proposal also includes joint agreements with IBM and MCI. IBM will be providing the hardware and the software to implement NSFNET's packet switching and network management facilities, while MCI will play a key role in providing the transmission backbone for the NSFNET. The T1 circuits supplied by MCI within their backbone will be implemented through DS3 carriers. Individuals wishing further information about this project may contact us at: "Merit_Computer_Network@um.cc.umich.edu." We started to beta test the traffic statistics program developed by Bob Braden at ISI. Among other functions, the program allows gathering a net-to-net traffic matrix about the data flow on the attached Ethernet. As this function will be a valuable tool to determine traffic flows within the NSFNET backbone, we started to use it to gather statistics on the Ethernet attached to the NCAR NSFNET backbone node. This testing was possible thanks to Don Morris at NCAR for providing the necessary access to his machine. We have installed and are starting to use the SGMP (Simple Gateway Monitoring Protocol) software in our Proteon gateways while runninig the user application program (available from NYSERNET) on a Sun workstations. We also experimented with the SGMP version available from Jeff Case (UTK) for IBM-PC type machines. Hardware and software designed at the University of British Columbia has been installed at the University of Michigan to implement TCP/IP protocols for mainframes running the MTS operating system. The applications for Telnet and FTP are currently in alpha test. Western Michigan University will gateway its DECnet installations to the Internet world using a MicroVAX 2000, which is supposed to be delivered soon. A T1 line has been ordered by Michigan State University to upgrade the link between the University of Michigan and Michigan State. Westine [Page 12]  Internet Monthly Report December 1987 This will be used for both Merit internal as well as for Internet traffic. MSU has received a 56Kbps line to the Purdue Arpanet IMP, the modems and hardware to connect the line on its end, but still waiting for the line to be connected at Purdue. by Jessica Yu (jessica_yu@um.cc.umich.edu) MIDNET (No report received) MINNESOTA REGIONAL NET (MRNet) (No report received) NCSANET T1 to UI-Chicago is up and running. University of Chicago and Northwestern University will be connecting through UI-C via T1 by March 15. Currently these schools share a 56Kbps circuit to NCSA. Argonne National Laboratory, currently connected through UI-C at 56Kbps, will upgrade to T1 as well. Illinois Institute of Technology is in the process of planning to connect through UI-C at 56Kbps. The University of Wisconsin at Milwaukee has ordered a 56Kbps circuit to NCSA, scheduled to be up by March 15. The Vitalink earthstation connecting Indiana University is scheduled to be moved (due to local construction) from UI Digital Computer Lab to UI Computing Applications Building at the end of January. by Charlie Catlett (catlett@ncsa.uiuc.edu) NORTHWESTNET All but two nodes of NorthWestNet are up and running; the last two, Washington State University and the University of Idaho, will be connected this week. A 56 kb line to the NSFNET backbone from Seattle to San Diego will be installed by the end of January or early February. by Hellmut Golde (Golde@cs.washington.edu) NYSERNET (No report received) SDSCNET (Refer to SDSC backbone report) SESQUINET The complete initially proposed SesquiNet configuration has been operational now for five months. The following campus networks are being served, and are advertised via EGP to the core: Westine [Page 13]  Internet Monthly Report December 1987 Baylor College of Medicine 128.249 Houston Area Research Center 192.31.87 Rice University 128.42 Texas A&M University 128.194 Texas Southern University 192.31.101 and the University of Houston 129.7 The phone line from NSFNET/NCAR to SesquiNet/Rice has been completed, and we have been using this line operationally for the last three weeks to a limited extent. The line connects a pair of cisco gateways: one integral to SesquiNet (at Rice) and the other integral to WestNet (at NCAR). Since the gateway at NCAR shares an ethernet with USAN and with gateways for the NSFNET backbone and the NASA Science Network, it has lots of potential usage. We have not yet, however, completed the arrangements for the exchange of routing information that will allow this line to be used to the fullest. Activities for the coming month include: > Testing the dual protocol (IP and DECnet) support for the cisco gateways within SesquiNet. > Adding Prairie-View A&M University to the network. by Guy Almes (almes@rice.edu) SURANET (No report received) WESTNET 1. In New Mexico, New Mexico Technet, New Mexico Inst. of Mining & Technology, and New Mexico State University are now interconnected. The 56 kbps serial line from New Mexico Technet to NCAR is projected to be installed by Feb. 15, 1988. The University of New Mexico is yet to be connected, as they are in the process of performing some adjustments to their campus network before installing the cisco gateway. 2. The University of Wyoming is connected to the Internet, but is experiencing minor problems with the 56 kbps serial interface card which are in the process of being resolved. 3. We are continuing to investigate the best manner in which to provide service to Brigham Young University and the University of Utah. A mini-proposal is being prepared in this regard, to be forwarded to NSF. 4. We have been negotiating with Eric Aupperle where the T1 line to serve Westnet should be installed. We have been told by NSF that the University of Utah is the favored location. Westine [Page 14]  Internet Monthly Report December 1987 5. Both Colorado State University (129.82) and New Mexico State University (129.138) have successfully completed the transition to Class B addresses. 6. David Wood has discussed with Scott Brim running the "HELLO" protocol on Westnet. The software is available, and we are waiting for it to be installed at NCAR before trying it. We anticipate testing it during the month of January. by Pat Burns (pburns%csugreen.bitnet@cunyvm.cuny.edu) Westine [Page 15]  Internet Monthly Report December 1987 TASK FORCE REPORTS ------------------ APPLICATIONS -- USER INTERFACE No report received. AUTONOMOUS NETWORKS No news to report for the ANTF. Deborah Estrin (Estrin@OBERON.USC.EDU) END-TO-END SERVICES Nothing to report this month. Bob Braden (Braden@ISI.EDU) INTERNET ARCHITECTURE About forty people attended the INARC Workshop held on 17-18 December at BBN. The theme of the workshop was technical issues of the next-generation internet. A report on the agenda items and discussions, which included a bomb scare, will be available shortly. Dave Mills (Mills@HUEY.UDEL.EDU) INTERNET ENGINEERING 1) There has been another round of growth in the IETF mailing list. After exposure at the December TCP/IP Conference and in the December `Connexions', requests swelled the list to almost 250 names and about 10 exploders. We can't really host meetings of 250+ folks yet and I suspect (based on the actual attendence of the last few meetings) that many are more interested in the email discussions, rather than becoming actively involved. Therefore, I feel it is time it re- institute the ietf-interest list, which I had eliminated a few months ago. Toward that end, I divided the full list by putting all names of people who attended either the last two meetings into the main list and the rest (and all exploders) into the interest list. Westine [Page 16]  Internet Monthly Report December 1987 >From now on, all general discussion should be addressed to ietf-interest@gateway.mitre.org or ineng-interest@venera.isi.edu. (the lists at ISI and Mitre are identical.) The main list is ietf@gateway.mitre.org or ineng-tf@venera.isi.edu. (again, the lists are identical.) The interest list, of course, includes the main list. The only usage for the main list at this time will be to announce specific meeting details (ie, time, date, place). The interest list will get announcements of upcoming meetings but details will be lacking so that anyone wishing to attend will need to request information. This will act as a primative registration procedure, which may have to improved in the future. So let me repeat, in the interest of giving widest exposure to the task force discussions, please address all mail in the future to the interest lists. 2) At the last meeting, we decided to start a draft document series to act as a clearinghouse for Task Force papers and proposed RFC's. There are now four files in the directory at sri-nic.arpa, which represent 3 TF documents and an index of the documents. These files are available by anonymous ftp in the usual way. The four files are: IDEA.INDEX - Index of the current or upcoming documents residing in the directory at sri-nic.arpa. All files are available by anonymous ftp in the usual way. IDEA002.txt - `A Comparison of "Link State" and "Distance Vector" Routing Algorithms', Ross Callon (BBN) Title is self-explanatory. IDEA003.txt - `Guidelines for the use of Internet-IP addresses in the ISO Connectionless-Mode Network Protocol', Ross Callon (BBN) and Hans-Werner Braun (U Mich) Update of RFC-986, which proposes a mapping for DoD Internet Addresses on the ISO OSI addressing format. IDEA004.txt - `Routing Information Protocol', Chuck Hedrick (Rutgers) Westine [Page 17]  Internet Monthly Report December 1987 Description of the Routing Information Protocol (RIP) (IDEA001.txt, a description of the goals, history, charter and organization of the Internet Engineering Task Force, is in preparation.) This is essentially an administrative mechanism meant to help streamline the RFC review process within the IETF. There are already about 4 other documents already in the IDEA pipeline, in degree of preparation. Anyone wishing to register a good IDEA, contact me or one of the working group chairs. 3) The date of the next IETF is March 1-3 in San Diego. Those wishing more information should send to ietf- request@gateway.mitre.org. Phil Gross (Gross@MITRE.GATEWAY.ORG) INTERNET MANAGEMENT No report received. PRIVACY Work proceeded during December on revision of RFC-989 per discussions at Privacy Task Force meetings during 1987; it is anticipated that an updated RFC will be distributed in January 1988. Scheduling of the next task force meeting for 2-3 March, 1988 at Lawrence Livermore National Laboratory was confirmed. John Linn (Linn@CCY.BBN.COM) ROBUSTNESS AND SURVIVABILITY No report received. SCIENTIFIC COMPUTING No report received. Westine [Page 18]  Internet Monthly Report December 1987 SECURITY No report received. TACTICAL INTERNET No report received. TESTING AND EVALUATION No report received. Westine [Page 19]