Network Startup Resource Center Bush and Klensin Copyright 1992, Randy Bush & John Klensin. All rights reserved. Overview -------- US research and engineering scientists need access to the Internet to/from developing nations for when they travel, for collection of data from distant parts of the planet, and for collaboration with R&E scientists in the ever more sophisticated but still developing world. It is proposed to formalize and set up a minimal-cost human and computer resource for catalyzing, coordinating, and disseminating the information and tools necessary for nations and areas in which communications or computer facilities have traditionally been primitive to acquire and use low-cost and appropriate technology to bootstrap themselves on the path to joining the Internet. This network startup resource center would: o make tool kits for low-cost IP, UUCP, FidoNet, and other appropriate networking technologies available to networkers via FTP, UUCP, mail- based archive service, and the post, o make documents on starting and early phases of networking available by the above means, o assist networkers in new areas in contacting appropriate assisting agencies and people and get them in communication with each other, o catalyze and coordinate provision of networking textbooks etc. which are often exceedingly hard to acquire in information-deprived areas, o assist US scientists traveling abroad to contact existing or nascent networking services, o gather, especially by collaborating with those who already gather it, information about existing facilities in or near network-starved areas so that new networkers may lever off of them when possible, o disperse that information as widely and freely as possible via FTP, diskette, WAIS, paper, and any other reasonable means, and o speak for, or catalyze folk to speak for, the needs of the emerging networks in the fora of international networking engineering and administrative decision-making institutions. The network startup resource center exists informally now, but relies on people's evening and vacation hours and on networking facilities which are not funded past the second quarter of calendar 1992. It is proposed to fund minimal personnel time and support of networking facilities using existing computer resources at a cost of $48,800 per year for three years, plus roughly $10,000 of non-recurring expenses. All costs to using countries, institutions, and individuals are and would continue to be on a cost-recovery basis. The next three years are the critical period of international networking proliferation where the interests of US R&E workers should be ensured at the global level. By the mid-'90s, the initial proliferation game will have mostly played out, and the US researchers will either be strong participants, collaborators, and leaders, or have been left behind. Background ---------- Unlike the situation of thirty years ago, when almost all important scientific work occurred in Western countries, and the US was clearly predominant in most fields, the scientific community is becoming increasingly international. Important work and areas of study occur all over the world. Collaborations and ability to access sources of data and other resources are increasingly important to scientific progress. In many fields, we see more and more inter-institutional collaborations on research and papers that draw on the strengths of each of these institutions. Exchanges of ideas and collaboration and review of proposals should not be limited to one country, or even to developed areas. Especially in such areas as the health and social sciences and in all of the various fields that study "global and environmental future" issues, participation of scientists in developing areas has become crucial. This is true whether the scientists themselves are indigenous to, or visiting in, those areas; indeed, as the community becomes more international, the distinction between the two is gradually becoming less clear. Communication facilities based on computer networks, especially the low-level ones such as electronic mail, have become critical to these types of collaboration. The post is simply too slow to permit real interaction, and fax, while faster, does not lend itself well to group interaction, much less true collaboration on shared materials. The US has taken, and should continue to take, a leading role in spreading international networking, both to preserve an international technological lead and to support the work of our own scientists, who might otherwise find themselves at a disadvantage. It is worth noting, for example, that US scientists have been excluded from several European efforts by lack of effective connections to ongoing discussions. The more the network connection infrastructure can be opened up, the more scientists can participate in international efforts on the basis of interests, skills, and knowledge, rather than based on where they happen to live or work. One can best understand the future in this area by looking at examples of the recent past. Mathematicians in the US have begun to establish relationships (professional society to professional society) with mathematicians in Africa. Others have established active collaborations with colleagues in Russia. We are establishing links between nutritionists in the US and elsewhere and their colleagues in the island South Pacific and Southeast Asia, with Africa and Latin America to follow as soon as the infrastructure can be put in place and personnel trained. The UN is leveraging academic and research networking all over Central, South, and Latin America. And, of course, a very large number of collaborations have sprung up between scientists in the US (and Europe) and their counterparts in Eastern Europe and the Soviet area, as the changing environments in those countries have made these links feasible. Similarly, data gathered or prepared in remote locations often must be transferred back to the US - or, if the US does not have the facilities, elsewhere - for evaluation or analysis. The examples usually cited are climate or rainfall data, but similar issues arise with health and nutrition status, mortality and disease figures, and even certain types of economic statistics. Without computer networks, the options are to mail machine-readable media (often unacceptably slow and unreliable) to send the data via fax (typically requiring rekeying or the use of OCR techniques, which are not completely reliable), or long-distance international data calls at great expense. The demand from the US scientific community has been present for some years as evidenced by the "how do I access a network to stay in touch with colleagues at home while I'm in the field or to work with colleagues at remote institutions from my home institutions," inquiries which appear at very frequent intervals on popular network mailing lists and news groups. What is relatively new is the potential to satisfy those requests and unleash some of the demand. This proposal is intended to provide the tools needed to get such network connections operational quickly and smoothly when the demand exists but the networks do not. Many Latin American, Caribbean, and African universities tell of invited US scientists who would not come for sabbaticals or extended research visits because they would not have email access to their colleagues. The problem of being without network facilities is not merely one of keeping up correspondence, but of being isolated to the point of ineffectiveness if reliable email is not available. On the other hand, when networks are available, previously-unanticipated collaboration seems to come into being almost spontaneously. Again, the underlying causes seem to involve a latent demand that remains latent as long as joint work requires either the disruption of waiting for the post, the continual retyping of texts transmitted by post or fax, or the need to secure large budgets and approvals for extensive international travel. When computer networks are available, people quickly become comfortable with using them, the collaboration seems to happen often and quickly in many disciplines and work groups, and needs only a little bit of outside stimulus to get it started in others. Although the availability of computer networking continues to spread throughout the world, we are reaching a point where the growth curve may be slowing. This slowdown is not due to overall saturation or a lack of need, but to the fact that the "easy places" have become saturated and the remaining places are, for one of several reasons, less easy to penetrate. These reasons include: o A lack of on-the-ground technical knowledge. Perhaps obviously, it is easier to put wide-area networking in place in an area where there is already well-developed knowledge of networking, pre-existing LANs, etc., than when no such expertise or experience exists. This lack of knowledge includes not only the rudiments of networking, but, more important, an understanding of the options, tradeoffs, and the most appropriate technologies to be used in particular situations. That understanding has traditionally not been available even in fairly sophisticated communities, since they have not had to concern themselves with situations normal to developing areas, e.g., where the telecommunications infrastructure is marginal at best. Many of the people who are normally considered experts can make poor guesses and give poor advice if cost and technology tradeoffs are radically different than they are in countries with established telecommunications and network infrastructures. This situation prevails, almost by definition, in many less developed areas. Although the reasons may be different, it is also prevalent in many areas of Eastern Europe and other portions of the former Soviet area of influence. A very similar situation occurs in the US K-12 arena, where the cost of a single phone line can be a major administrative obstacle. o Inability to create adequate demand to marshal the resources that are perceived to be necessary until the network is in place. In other words, building and maintaining networks costs money and other resources (and these costs are often overestimated as part of the process of resisting change). Unless there is significant demand, the investment may never be made. But the demand often does not arise until networks are actually seen to be in place and working. This inertia, caused by perceived low demand levels, is not limited to developing areas. One could argue that EARN represents one of the greatest successes in "selling" networking to a somewhat resistant audience. EARN almost certainly would not have come into existence had not IBM funded the initial effort on a basis that made the initial network infrastructure very low cost for the institutions. IBM then withdrew funding after people and organizations were "hooked" and the network had evolved means of standing on its own. The present evolution of EARN toward the use of TCP/IP and OSI technologies may be viewed as the inevitable final step in a process for which NJE was an appropriate first-step technology given hardware and personnel infrastructure already in place in the relevant countries. A very similar situation exists in the US K12 (kindergarten through 12th grade educational) community. Few of the innovative teachers, the potential initial users who perceive the benefits in advance, have the political and economic power to affect their networking destiny. Network inertia and data hoarding are rife in the administrative infrastructure, and trickle-down theories are regularly disproven in the data networking arena. Network development and installation may, as a result, not be possible unless the resources for network bootstrapping are available readily and at minimum cost. But it has been shown time and time again that a minimal networking channel will quickly build a user base which soon fills that channel. Even in less-developed areas, the perception of value rapidly builds to the point that the user base then manages to find its own funding to continually increase the available bandwidth. o The wrong user community. The history of starting networks and network connections has predominantly rested in computer science and computer technology-oriented departments, businesses, and other data communities. There is some history of these communities constructing networks for their own use and then using various mechanisms - costs, perceived complexity, lack of support to then hoard the resource. When computer *user* communities - scientists, educators, or the general public - for whom the computer is a tool for communication or computation (but not an interesting device in its own right) need access to networking technology, they often need to start over. At the same time, those user communities are much larger and, in many places, represent the largest potential user community. They may ultimately have access to greater political, cultural, and financial resources than computer technologists. At the same time, the community which is more computer-oriented can, and typically has, managed to establish communications when they see it as sufficiently important. While they may not understand optimal approaches for a given area, they usually have access to sufficient information to make something work. Other scientists and educators have had less success, since they do not have the technology readily available and may believe (or "know") many things that are untrue, such as the need to focus computer networks around mainframe systems or centralized remotely-accessed hosts or the requirement for very high-bandwidth circuits to do anything useful. These misperceptions are encouraged by publicity releases from vendors of high-end systems and by a media focus on the "next generation". o A special version of the "last kilometer" problem. Most of our traditional models for connecting networks have involved establishing links to hubs, and then spreading the networks outward from those hubs. Progress in the last several years has actually produced actual or potential hubs in many areas of interest, with their own connections. Sites of interest to new users - including sites at which US scientists are spending time or with whom they are collaborating - are often not connected to those hubs and, if they are not, rarely know of their existence. Efficiency of connections and local availability of technical advice argue strongly that these new connections should be made, when possible, to the existing sites, thus utilizing resources that would otherwise have to go to long-distance connections for a single site into improved bandwidth for both new and existing nodes in the area. However, taking this approach requires that there are sources of information about existing sites and potential hubs and the technologies needed to connect to them, as well as, in many cases, some outside encouragement to focus resources on improving very wide area connection infrastructure rather than on parallel and likely low-performance connections. o The hardware equivalent of the "last kilometer" problem also causes difficulties. In many cases, the critical factor preventing a network from being put in place is a single, marginal, piece of hardware that may be quite inexpensive in the grand scheme of things or even available for donation if we could only bring potential recipients and potential donors together. This might be a simple modem or an IBM PC clone that could be dedicated as a mail hub, or, at the next level up, ethernet adaptors or related technology or high-bandwidth modems. In a few cases, there are no problems here worth fixing. The people who do not have networks and computer-based communications have no need for them, and would not know what to do with them if they had them. In a larger number of situations, however, we are dealing with a latent, and sometimes unrecognized, demand or an articulated demand that is waiting to be realized. Specific interest by a US researcher - such as an agreement to visit, lecture, or collaborate only if facilities are in place - in a particular area may focus demand, turning "perhaps this is something we should do someday" into "this is something that should be done now". If we are not prepared to support that transition at the opportune moment, we may miss the opportunities and see connections develop at other times and in circumstances that are less optimal for access from and by US locations and scientists. This proposal addresses the cases where the demand exists or is on the threshold of breaking out. It is designed to catalyze and help support existing or potential networking providers with advice, training, information about initial moderate-cost connection capability, and support structures needed to turn vague interest in network connections to the international academic and research community into actual connections. As these connections are increasingly used, "data demand" and consequent bandwidth requirements will move these connections along the path from "low cost and marginal but useful" into the beginnings of the type of national and regional interest and infrastructure that will ultimately lead to full Internet connectivity. Since "low-cost" connections can actually become very expensive if the traffic volume becomes high enough, the US should then have to assume, at most, only the costs of the domestic end of actual internet connections by the time circumstances make them necessary. US researchers can not and should not bring the net with them wherever they go or wherever they need data. Typically when they do so, they spend excessive time organizing networks rather than doing research in their substantive areas. Costs will tend to be higher than necessary because such single-user links cannot benefit from the economies of scale of shared infrastructure. Equally important, networks that are perceived of as being set up by, and for the convenience of, "outsiders" tend to collapse as soon as the outsiders depart. Even when a visiting US scientist sets up a network facility, that facility should be structured so that it can expand into something permanent, serving local users and linked with other facilities in the region, rather than providing only a "call home" capability. If nothing else, the alternative is that the next visiting scientist is likely to have to repeat the earlier effort, a considerable waste of time and resources. The critical focus, in all cases, should be to get the networks in place, and local engineers and other personnel trained and given access to tools and information so that experience and demand accumulate and connections become permanent rather than transitory. For maximum benefit to the US scientific community as well as those in remote areas, development and deployment should be in local hands, using local expertise, and primarily at local expense. While IP links are desirable and the clear eventual goal, one of the major goals of this project is to provide people with information and advice about the alternatives for when medium- and high-speed IP does not appear to be immediately practical, including UUCP, Fidonet, commercial providers such as CompuServe or MCIMail and even, where the appropriate structure already exists and can provide helpful bootstrapping, mainframe-based technologies such as NJE (BITNET/EARN). One aspect of the proposed work will be providing easily available and well-documented libraries of tools for the lower-end network technologies to run on the commonly available platforms. Such kits can be constructed and refined for IP on IBM PC clones (MS-DOS), UUCP (Unix and MS-DOS), and FidoNet (MS-DOS and Unix) for use over dialup phone lines and/or X.25 connections. We will investigate whether such kits need to be prepared in "regional" form - e.g., reflecting local languages, connection arrangements, or telecommunications constraints - and encourage or provide such tailoring as necessary. Some of this translation work has already been done with the assistance of colleagues in the Latin American and Caribbean regions. In order to permit low-cost networks to be developed and installed, several tasks should be pursued, more or less in parallel. Some of these contribute directly to network installation when the demand is already clear, others to building the demand and momentum, and still others to improving the technology or understanding of available technologies in ways that would make such networks either more efficient or better connected to the rest of the world. (1) Facilitating network installation. The critical path to solving the majority of the problems described above does not lie in buying things for people or even in installing networks for them. It lies in a series of information resources: putting potential donors together with potential refurbishers, recipients, and shipping mechanisms; arranging for databases of, and easy access to, information about hubs, potential collaborators, and technical information and alternatives. This, in turn, implies that the investment with the highest payoff is not in hardware or travel dollars (although some of each are clearly needed), but significant improvements in consistency and availability of data, tools, discussions of alternatives, and other documentation. An important part of this is drawing information together about efforts going on in parallel that might be coordinated in ways that leverage both into higher-quality services. Documentation collections should include small case studies of "how others have done it and how to learn from that"; such examples may be as important as specific instructions of how to put tools together. (a) Installation kits. This project will improve existing kits that permit someone to arrive in a particular area and quickly and efficiently establish network connections to the outside world and, ideally, to other sites in the immediate vicinity. Improvements to documentation and automated procedures are as important as improvement to the technical networking parts of the kits themselves. The better these arrangements can be made, the more likely it is that a scientist who is not a computer expert can take advantage of them. We will investigate different ways of organizing and presenting databases of sites that might accept connections from new sites coming online so as to make that information more available to the actual users of the installation kits. Neither the "where are the networks and what are they called" efforts of Quarterman, nor the "how to reach an address of the following form" efforts of Frey and Adams adequately address this issue. The raw UUCP, Fidonet, Internet, and BITNET/EARN/etc maps are not integrated nor are they organized properly for this purpose. (b) Initial IP installations. While IP connections to the rest of the world are desirable, we recognize that they will typically be a second or third step, not the first one. If nothing else, the processes of registering domain names, obtaining IP addresses, and, in some cases, export controls on necessary technology, mitigate against the concept of ad hoc "instant" network installations that are key to the minimal-cost networking model. At the same time, some planning and guidance can facilitate the installation of IP connections as soon as they are practical. Such planning and guidance should focus on preventing the types of user-visible incompatible changes that would argue against transition to IP technologies. The other major barrier to widely available IP installations has been the cost of router technologies. We are at a point of opportunity in this area, since a new generation of routers are now being installed within the US and parts of Western Europe. We propose to establish a coherent mechanism for soliciting donations or low-cost consignment sales of "obsolete" routers, reconditioning them for use in developing areas, and then making them available to network developers and aid agencies at costs consistent with installation in developing areas. When that option is not practical, we will provide support for PC-based routers for lower speed applications. This mechanism will be designed to make the critical information and contact points available to aid agencies and assistance organizations (e.g., Vita or the Peace Corps) who could reasonably be involved in actually brokering specific arrangements, shipping hardware, and so on. As an extra benefit, these same information resources could be used by people concerned about pre-college school networking in the US to obtain appropriate hardware and tools for those applications. We would use existing relationships with members of the IETF working group in that area and other groups and organizations to facilitate that use. (c) Other hardware assistance As discussed above, in areas where the major obstacle to installing a low-cost network may be a single piece of hardware, we need ways to bring potential donors together with potential recipients and, when necessary, arrangements for checking, refurbishing, and shipping. A situation parallel to the router one occurs here, although on a much larger scale: as companies and other organizations shift to 386- and 486-based PCs from 8088- and even 80286-based ones and from 1200 and 2400 baud modems to ones with capabilities in the 9600 baud and above range, the effective resale value of the older equipment drops to "not worth the trouble". This is especially true for those organizations who must deal with the tax consequences of the sale of this older equipment before its allowable life has run out. But this very same equipment, while "obsolete" in the US, could be of immense benefit if placed strategically in developing areas. Consequently, this effort will strive to establish the right framework, including ways to match those who "have and would like to get rid of" with those who need and those who are able to test and refurbish. (d) Training and technical assistance. It is vital to enable people as well as technologies through documentation and in-country training. Even if network connections are established by visitors, local personnel must maintain them and be able to spread the word and facilitate additional connections. This is particularly important for network clusters that are isolated from the major network hubs (which are predominantly in developed areas). Unless significant subsidies will be available, more users, more activity, and more nodes are all critical to cost-justifying the higher speed and more regularly serviced links that make network communication efficient. We will offer references to supplemental documentation for the maintainers of relatively-local network hubs, and to individuals and groups who can provide specific training for selected groups. Where necessary, we will collaborate in the development of tools to make training activities more effective. Part of the goal should be to bring engineers in developing areas up to the level that they can participate actively in the training of engineers in other developing areas, rather than requiring ever-escalating US or European participation in training and installation activities. More important than these specific efforts, we will expand on existing efforts, including the INFOODS/UNDP mailing list of people interested in technical support for expanded networking in developing areas, to provide a network-based support infrastructure for low-cost hub and spoke sites. Once a new group has network contact with the outside world, it is vital to provide access to people who have been down similar paths and who can and will provide ongoing advice, assistance, perspective, and collective problem-solving. Some of this documentation should be in non-traditional forms. For example, we should begin to develop (and solicit from the community) a library of lecture and training materials including scripts, slides, and videotapes that would facilitate both local training and demonstrations and discussions by less-trained but interested travelers. (e) Perspective and experience. In addition to these specific activities, we bring considerable perspective and experience in low-cost networking, developing areas, and appropriate technologies to this work. We also have established and extensive contacts with other active participants in this work (both in developed and developing areas) who will be anxious to participate with us in supporting and advising new sites. Such perspective and experience is important in this area. Much damage has been done - usually with the symptom of creation of isolated "networks" that cannot interoperate with the existing academic and research network community - by those who are well-intentioned but ignorant of current technological alternatives and opportunities. A relatively steady flow of discipline-oriented "informatics" and "networking" conferences, most of which seem to ask the wrong questions or posit isolating technologies, testifies to the importance of making more real experience and perspective widely available. (2) Building demand and momentum. As suggested above, demand and momentum rarely happen by themselves. Most of the communities where they will arise spontaneously already have computer connections (although of varying quality and sometimes only within those communities). The arrival in a developing country of a US-based researcher equipped with an installation kit and a desire to communicate with "folks back home" should be supported if it gets a connection installed. However, our goal should be self-sustaining connections and spreading infrastructure. That typically requires involving local personnel, ideally on a peer basis and with the assumption that, given a little training and assistance, they can sustain and manage their own networks with locally-arranged financing. We need to develop a sufficient structure of information about who is interested and what is already operating in a particular area so that, if there is will to cooperate, every effort in a particular area reinforces every other effort and strengthens the links to the outside. This, in turn, leads efforts down the path toward regional backbones as the most effective mechanism for providing adequate bandwidth through, and out of, the countries and regions. Interested parties can best leverage their own needs into effective networks if they have information about other interests and activities. If existing sites are not inclined to cooperate with new ones, the best solution is to simply develop parallel infrastructure, gradually leaving them out. That, of course, requires the same databases, training, and information as would be the case if there were no existing connections to the area. The existing volunteer cooperative efforts in the networking area provide considerable potential for very high leverage on installation and dissemination of network technology and experience. We already have good contacts with those efforts and, by connecting the volunteers with other efforts, expect to be able to build, and loosely manage, a growing and dynamic worldwide support structure. (3) Technological investigations and improvements. Several emerging technologies have great potential for low-cost networking, either for the long-distance connections needed to connect emerging networks to the rest of the world or for dealing effectively with "last few kilometers" problems within destination towns, laboratories, and campuses. (a) Local technologies. We propose to explore, and report upon the potential and tradeoffs of, several of the potential local technologies, including low-cost radio and laser and low-power microwave Ethernet bridging technologies. (b) Wider-area technologies. Technologies such as VSAT continue to have great potential but little actual realization. Part of this has been due to the fact that the vendors of these technologies have not successfully interpreted and developed applications-level protocols, such as SMTP, that would permit efficient carrying of mail that interoperates successfully with other environments. We would work with these vendors and researchers in the field to assist in the design, development, and deployment of electronic mail that would efficiently use these transport mechanisms in the context of relatively low-cost networks. At the same time that we provide information and advice to make these products effective for transporting SMTP, we will contribute as necessary to extensions to, and variants on, SMTP that make it more compatible with, e.g., high bandwidth but long turnaround delay transport products. We are, of course, ideally positioned to advocate such extensions and variations. This component of the effort has another benefit, which is a potential improvement in the competitiveness of US-based vendors by helping them design and tune products and services to the needs of the expanding markets of the developing nations. Resources --------- John C Klensin John Klensin has been on the permanent research staff at MIT for most of the last 25 years, participating in or directing a wide variety of projects, many of them involving the applied use or development of computer networking or related technologies. He was involved in the original design of the Internet's File Transfer Protocol, and now chairs the Internet Engineering Task Force (IETF) working group on extensions to the SMTP protocol. Dr. Klensin helped organize the "info-nets" internet mailing list, which has evolved into one of the primary cooperative sources of information about existing network connections. He is also the primary maintainer of the "header-people" mailing list, which has been the principal long-term discussion forum about electronic mail formats and gateways. He also serves a member of the CREN Technical Advisory Committee and was one of the primary Internet technical resources for the design of the "INTERBIT" BITNET/Internet mail gateway systems. Dr. Klensin chairs the ASC X3 ("ANSI") Technical Committee X3J1 on PL/I and is ISO/IEC JTC1/SC22 Project Editor for that language and former convenor of the corresponding working group. He chairs the Standards Committee of ACM, and is a member of ANSI's Information Systems Standards Board (ISSB). He is Director of the Secretariat and Project Coordinator for the International Network of Food Data Systems (INFOODS), a United Nations University Project to improve the quality, availability, and comparability between countries of data about the nutrient composition of foods. In conjunction with UNDP, he organized and operates a mailing list which has become the focus of several efforts to assist those who are trying to bring networking to developing areas. As part of the INFOODS effort, he has organized or promoted several links between the nutritional and health communities and international scientific and research networks, the most recent of these being a connection to Thailand, developed in combination with Mr. Bush, Peace Corps personnel in the country, and personnel from New Zealand and Thailand. Dr. Klensin has acted as advisor and consultant to CIESIN, a research consortium in Ann Arbor Michigan responsible for integrating Earth Observation and other remote sensing data with social, political, and health information. Randy Bush and PSGnet R Bush, a compiler and realtime kernel writer and software engineering manager with over 25 years in the computer industry, is Director of Enginering at Servio Corp, the leading commercial developer of object databases and tools. He has been a user and occasional implementor of networking in the US from the ARPANET to the current day Internet, UUCP, and FidoNet. He chairs the US Modula-2 language standards committee, and is the author (but not the designer) of the basic FidoNet technical standards. PSGnet, Mr. Bush's pro bono and not-for-profit outside-of-work effort, has been involved for over three years with the deployment and integration of appropriate networking technology in Southern Africa. He designed, taught about, and helped deploy a multi-country (South Africa, Botswana, Namibia, Zimbabwe, and soon others) network using, in order o FidoNet technology on dialup lines, o UUCP on dialup lines, and finally o low-cost IP technology based on 9600 baud and below links using old PCs and publicly available PC-based SLIP routing software. PSGnet provides the US end of Southern Africa's university networks' link to the 'outside' world. This link has moved from UUCP to TCP/IP during the second half of '91. Over the next six months, it is planned to move to the end goal, direct connection from Southern Africa to the NSFNET in the US. The UNDP and Union Latina sent Mr. Bush to do the technical work and training for the first networking within Peru, and to establish the first links to the US. PSGnet is currently Peru's UUCP link to the rest of the world. PSGnet provides cost-recovery-only UUCP links for R&E sites and NGOs in Israel, Japan, Peru, South Africa, the UK, and around the US. In doing so, PSGnet is careful not to compete with commercial providers such as UUNET, or with the NSFNET connectivity arrangements; and tries to act as a steppingstone to such more 'normal' arrangements. Mr. Bush is the founder and technical coordinator of the Portland Oregon IP metronet, RAINet. This is a cost-recovery-only SLIP/PPP network based on leased 9600 baud lines which has over a dozen sites around the Portland metropolitan area. RAINet serves individuals (disallowing commercial traffic out of RAINet) as well as public schools and public access sites in Portland. PSGnet serves as the RAINet NIC and NOC. RAINet has volunteer engineers maintaining and enhancing PC-based IP routing software such as ka9q and PC-Route, developing networking tools for low-cost IP and UUCP links, and coordinating with FidoNet software developers and those developing gateware between FidoNet and RFC822-based networks. PSGnet provides FidoNet's links from North America to/from the world's other continents, with a few dozen direct dial-up links to Argentina, Australia, Germany, Hong Kong, the Netherlands, Poland, Singapore, South Africa, Taiwan, Thailand, and the UK. These links essentially bind FidoNet across the globe from the USSR, through Europe, the Middle-East, Africa, North and South America, Oceania, and Asia. PSGnet is the major gate between FidoNet and the Internet/UUCP world, moving almost 2,000 messages a day. As FidoNet in most other countries has no access to their national networks, all of their traffic goes through PSGnet and then over the international FidoNet links. PSGnet also provides the FidoNet/Internet gateway for specialized projects such as K12-Net and other public services. Budget ------ Financial administration for this project will be handled through the International Nutrition Foundation (INF), which serves as the primary US fiscal point of contact for the United Nations University's Food, Nutrition, and Hunger Programme and, as a result, as a past or present contractor or grantee under several US Government programs. INF has agreed to accept the project proposed here as a complementary adjoint to INFOODS. There will be no overhead on the budget for this project. PSGnet and John Klensin would deal with the INF, which would administer and report the use of funds. Klensin and Bush would be responsible for all non-financial status and project reporting. We propose support for the 56k IP link at PSGnet, and duplication and distribution costs incurred at PSGnet in direct support of the project. This facility already has computers, modems, an AlterNet IP connection, and the beginning of the resource information center we propose. We propose support for the following at direct cost: o $1,100/month AlterNet connection o 150/month Leased line to AlterNet POP o 200/month Estimated duplication and postal costs for document and floppy distribution The IP connection is currently funded until May '92. We propose the consultancy of John Klensin not to exceed $2,000/month and associated administrative support and telephone expenses not to exceed $200/month. We propose paying travel expenses of John Klensin to IETF and closely related meetings to represent the needs of US R&D in developing areas. With an average of one intra-US trip every two months, we estimate these costs at $600/month. PSGnet does not have a printer and will likely run short of disk space during the three years. Currently, printing facilities are borrowed elsewhere. We would budget $5,000 for purchase of a printer and disk sometime during the three years, but not plan to spend it until needed. It has been suggested that delivery of technical assistance by videotape (NTSC, PAL, and/or SECAM formats would be needed) or similar hi-tech media would be effective. $5,000 over three years should cover such experiments. Very rarely, paid consultation of experts in technical or international matters may be needed. We would budget $2,400/year in case it is found necessary. Summary ------- We propose minimal funding to continue and expand our ongoing efforts in support of entry-level networking to allow US researchers to have access to the global network when abroad and/or when they wish to gather data from abroad. We propose to continue our orientation toward aiding and empowering the local engineering and networking infrastructures, both abroad and in net-poor segments of the US such as K12, by making available and transferring by any reasonable means entry-level tools and technology, documentation, training, etc. We propose to catalyze and offer, but not impose or control, coordination of networkers who are working in developing areas abroad and within the US to minimize duplicate efforts and expenditure of resources. ============================== Network Startup Resource Center Bush and Klensin Project Summary A set of existing informal arrangements for providing assistance in setting up new computer networks in developing areas and linking them to the US will be enhanced and focused on the particular needs of US Scientists travelling to, or developing collaborations with, these areas. The resource center will provide for easily available installation tool kits for network software installation together with documentation, advice, and information about possible linkages to existing networks and facilities. It will provide a catalyst and focal point for making textbooks and surplus hardware available to critical individuals and hub facilities from which national and regional network spines might be expected to evolve. By building on a existing support networks of people and facilities, the center will provide very high leverage on the use of NSF funds. While emerging and novel means of delivery of materials will be investigated and evaluated, the primary focus will be on the availably and use of proven, time-tested, and reliable technology in all areas so as to make networking facilities available and useful to scientists who are not computer experts.