Interoperable Networks: Why Should Your President Care? Copyright 1990 CAUSE From _CAUSE/EFFECT_ Volume 13, Number 4, Winter 1990. Permission to copy or disseminate all or part of this material is granted provided that the copies are not made or distributed for commercial advantage, the CAUSE copyright and its dateappear, and notice is given that copying is by permission of CAUSE, the association for managing and using information resources in higher education. To disseminate otherwise, or to republish, requires written permission. For further information, contact CAUSE, 4840 Pearl East Circle, Suite 302E, Boulder, CO 80301, 303-449-4430, e-mail info@CAUSE.colorado.edu INTEROPERABLE NETWORKS: WHY SHOULD YOUR PRESIDENT CARE? by J. Patrick Casey and James G. Williams ************************************************************************ J. Patrick Casey and James G. Williams are associate directors at University Computing Services, Indiana University (see pp. 30-33). Casey's support unit is responsible for consulting, computer education, and user assistance with the planning of departmental computing environments. Williams' network systems unit is responsible for IU's wide-area network, as well as the campuswide and local area networks on the Bloomington campus. ************************************************************************ ABSTRACT: Higher education is facing a unique challenge -- achieving excellence in the face of a declining student population, heightened competition for faculty, and inadequate funding. With more to do and less to do it with, new tools and new thinking are called for. This article suggests that an interoperable network can bring new efficiencies to the use of resources, provide far-reaching access to services, and connect scholars to information. Over the next ten years, institutions of higher education across the continent will be faced with a rapidly changing environment. Declining populations in typical student age groups will heighten competition for not only the very best students, but for all students. More faculty are retiring than are entering the profession, increasing competition not simply for the best faculty, but for all faculty. Growing awareness in federal, state, and local governments of the seriousness of the current financial situation will lead to decreased or, in the most optimistic scenarios, flat public funding for higher education. Then there are some basic facts of economic life in higher education: "The prices of the goods and services purchased by universities as measured by the Higher Education Price Index (HEPI) have been rising even in relation to the Consumer Price Index ... universities have not only been paying more for the goods and services they buy but have also been buying more of them."[1] Taxpayers, parents, and research foundations alike are increasingly reluctant to allow tuition prices to escalate. All these factors will force higher education into new models of thinking. However, there are institutions that will not only weather this financial storm but thrive in it. Common to these "thriving institutions" will be their view of information technology: not as an expensive but (regrettably) unavoidable cost of running a modern institution, but as an enabler of institutional excellence, as a strategic investment. As we head into the 21st century, it will not only be industry that worries about "time-based competition," but also higher education. It will not be enough to have the best product; you will also have to get it to market first. That product can be viewed as teaching, research, public service, or all three; it doesn't matter. The institutions that thrive in the 21st century will be those that not only balance cost and benefit by achieving excellence in the quality of their basic commodities (teaching, research, public service), but that do so efficiently. A critical enabler to achieving this excellence and efficiency is the development of an interoperable network. Developing access for disparate technologies Increasing refinement and specialization within the computer industry has produced a variety of systems that have little in common with each other, and this has consequences for your institution's reach for excellence. A UNIX-based engineering workstation is completely different from an IBM PC, which in turn is worlds apart from the Macintosh environment. However, users of UNIX workstations, Macintosh workstations, and IBM PCs will all want access to resources that institutions will have available on their computing networks. Can your institution afford to disenfranchise its computer science, engineering, or mathematics faculty by not supporting UNIX networking? Can it afford to abandon its fine arts or theatre and drama people by turning a blind eye to the Mac world? Or can it ignore its business or public and environmental affairs schools, often heavy DOS users? What about your large shared systems, your mainframes? Many institutions are finding that they can no longer get the services they need from any one vendor. Can you afford to let either your IBM or DEC users go without connectivity? And above all: can your institution develop and maintain excellence simultaneously in all of these disparate technologies? Can your best technical or support staff really be in two or more places at once? Developing access to disparate services Basic cost considerations will make it impossible for even the largest universities to develop or maintain state-of-the-art expertise in all disciplines. Access to services in all of these environments can be assured for all devices by developing an interoperable network. A wide-ranging interoperable network allows the largest and smallest of institutions to tap into specific areas of expertise at other institutions without having to develop that expertise themselves. The network will serve to unite scholars in both intra- and inter- disciplinary efforts. The small engineering school can attract the scholar who needs access to a Cray Y-MP. The music historian need not be isolated in a remote college environment. An inter-disciplinary research group can exchange information regularly, independent of location. What is an interoperable network? Strange as it may sound, an interoperable network is to networking what relational databases are to information systems, and what UNIX is --or would like to be -- to operating systems: a way to provide information services across disparate hardware and software platforms. Specifically, by an interoperable network we mean a network based on an arbitrary lower-level protocol, such as Ethernet or token ring, and a specific interoperable higher-level protocol, such as TCP/IP or OSI/TP4. This higher-level protocol which "rides" on the lower-level physical protocol is designed to be independent of any particular computer or networking manufacturer. The two internetworking standard protocols in use or development across the United States today are TCP/IP and OSI. Although OSI has the official blessing of sanctioning bodies, TCP/IP has a much longer history and at present a much richer networking and application environment. Although everyone talks about a transition to OSI, or a shift away from TCP/IP, that shift or transition could be a long time coming. OSI protocols promise functionality beyond what can be achieved today through TCP/IP. However, the exact performance penalty that these protocols would place on end-user systems is at present unclear. Why build an interoperable network? Networking will play a key role in the attempt to recruit and maintain the best possible faculty and students for the institution. Just as it is impossible for each institution to develop excellence in all areas of teaching or research, it is impossible for each institution to match -- let alone exceed -- the excellence of every type of information resource present at every other institution. However, through the use of an interoperable network, it is possible to tie institutions together to provide access to these resources. Consider as an example Indiana University, an eight-campus system with two large campuses and six smaller ones. We take it as a given that scholars at the smaller campuses of IU should have access to all information available to scholars at the larger campuses. In fact, we have as a goal that no scholar should come to any IU campus and not have access to at least the same quality of computational and informational resources available at his or her previous institution. Another institutional factor that will come to be examined more and more carefully in the harsh light of tighter budgets is separate academic and administrative computing networks. This is not to say that "heads-down" records management will not have special networking requirements, or that they are the same as those needed to support 3-D, full-motion visualization (not to mention voice and video, coming soon to a computing network near you). Nor is this to say that academic and administrative computing organizations need necessarily merge. But the networking technology that enables end users to access these services will be forced to become more tightly interwoven. Very few institutions can afford -- financially, or from the point of view of "opportunity cost" --to put two or three or four workstations (each with different network technology) on the desk of every faculty and staff member. Likewise, no institution can afford to deny their faculty and staff access to key administrative and academic information. The solution to this problem is an "internetwork" that seamlessly connects both administrative and academic resources. How do we get there from here? To be effective, an internetwork does not have to cover the four corners of the earth and have hundreds or thousands of nodes attached to it. In fact, even work groups on campuses that have developed local area networks have seen the productivity gains that can be realized by this close sharing of information. It seems only logical that these local work groups would now want to extend this productivity gain across larger and larger segments of the university community. A carefully planned and designed internetwork allows these groups to do this in a transparent, simple manner. Because an internetwork is based on networking standards, rather than the proprietary protocols of particular vendors, local area networks that support national (indeed, international) standards are able to participate in the larger campus area networks. It is theoretically possible in the space of a week to go from having no nodes to having literally hundreds of nodes on a campuswide network by simply connecting established local area networks that support standardized internetworking protocols. We do not suggest that if your institution doesn't scrap SNA, DECnet, and ASYNC (or RS232) overnight, your president will come looking for you tomorrow with an early retirement opportunity. But we do suggest that the institutions that thrive in the 21st century will be those that develop strategies today for implementing ubiquitous access to computing resources tomorrow, making all computing resources of their institution --and indeed of the national and international research networks -- a seamless part of the scholar's workstation. ======================================================================== Footnote 1 Enhancing Quality in an Era of Resource Constraints: Report of the Task Force on Costs in Higher Education (Ann Arbor, Mich.: University of Michigan, 1990). ========================================================================