Internet2: What does it mean for Libraries and Librarians?

Report on LAUC-B Research & Professional Development Committees May 17, 1999 Program

Speaker: Clifford Lynch, Executive Director, Coalition for Networked Information

While almost everyone is familiar with the Internet, "Internet2" seems to be surrounded by "murk and confusion" for people not directly involved. What is it? How does it relate to other things?

Lynch addressed two separate but related matters: 1) What is Internet 2, and, 2) What difference is it likely to make in higher education’s future?

Lynch sketched the history of the Internet, from its beginnings in the 1960’s as part of Federal government defense projects, through its backbone stage with funding from NSF (National Science Foundation) and access to supercomputer centers, and finally to academia and education centers. In the early 1990’s there was a "huge transformation" to enormous commercial enterprises, where the Internet was no longer just for the research and education communities. From 1990-1996, the Federal government extricated itself from most of the running and support of the Internet, and it took on a consumer component.

Today the Internet is predominantly a commercial enterprise. Academia, federal and state government agencies, and K-12 certainly have a presence on the Internet. But if one looks at the percentage of the population on the Net, or, if one looks at the revenue the Network generates, these are now infinitesimal. The Internet is now a resource for all. However, what the people might want as part of commerce might not be what academia necessarily wants. And although in the mid-1990’s, universities and libraries were using the Internet in their day-to-day business, there existed no "road map" on how to support the "next generation" of advanced large-scale applications universities have been dreaming about, primarily in the sciences. Out of these tensions has come Internet2.

Internet2 is a collaborative effort to develop advanced Internet technology and applications toward the research and educational missions of higher education. In 1996 a group of major universities got together, initially under the auspices of EDUCOM, asking how could they supplement the existing Internet to create a high-performance test-bed for what scholars and scientists wanted to explore. Internet2 is essentially a high performance, pre-commercial network.

By 1998 there were 70-80 universities, and now there are about 150 members. There are also partners in industry and government, and there have been many gifts and donations to Internet2 efforts. Some donations have been equipment or bandwidth, all of which translates to "incredible capacity", speeds close to backbone capacity directly out to applications. Internet2 is a project "among regions and between campuses", involving Wide Area Networks and "gigapops".

Getting started means each campus must upgrade infrastructure to even participate. Speaking of backbones and gigapops, Lynch likened the event to "someone showing up at the edge of each campus with a very large fire hose". Just to get started, we’re talking about OC-12, 655mbits/second. Each campus can expect to see a complicated political process, with upgrades in certain places and others to get upgrades later. However, there was a members meeting in April 1999 in Washington D.C., and Internet2 is rolling-out pretty fast. Connections are going in for Internet2 sites. [In the Q&A session at the end of the talk, a campus IS&T spokesperson said Berkeley was essentially ready and taking proposals. The first programs at UCB are probably going to start with astronomy and computer science.]

Internet2 is a US based project, but interest is emerging from other countries and entities, Canada, Singapore, U.K., Netherlands, Europe, Japan, Nordic countries, CERN. Memos-of-understanding are being written, and ways to hook networks together at really high speeds are being found.

Some groups think universities are seceding from Internet"1", but this is not the case, no matter how strongly this urban legend persists. Internet2 is a "supplement to the commercial internet, and the university community will remain fully connected to the public internet."

Bandwidth is what gets the most attention in I2 today. But there’s another part, implementing new network services which will enable new applications. One type of new service is Quality of Service (sometimes called differentiated services). The existing internet is fundamentally democratic –"Give me your packets, and I’ll do my best to deliver yours and everybody else’s." But some applications -- telemedicine, remote observational devices like telescopes, interactive visualization -- require tighter performance criteria than e-mail. As the network loads up, quality of service will imply political and economic arrangements for resource rationing for high-performance applications. The other new network service is "multicasting", which facilitates multipoint video in a much more efficient fashion and will be essential in a video and audio rich environment.

Today the majority of the advanced applications are still in the sciences. Beyond the science applications, Lynch talked about "broad commonality" applications. What does Internet2 mean for the future of libraries? In the narrow interpretation of the question, it doesn’t change things much. You may add tele-reference; it may facilitate more graceful deliveries of image collections. Internet2 opens some doors, for creating user interfaces in a different way, for designing "systems" that are a lot more graceful. Offsetting these gains, it is important to remember that a lot of people won’t be on Internet2 for a long time.

Interpreted in a broader sense, the question is "interesting". What can libraries do with this stuff? Some have thought that there were enormous video/audio/media collections just waiting to be shared. But there are not as many such collections around as was thought, and many are heavily encumbered with property restrictions and rights. In the near term, Internet2 will not enable an enormous number of library applications. But way down the line in time, it will create new "content genres". E.g., it will enable different ways of doing science, with researcher distributed geographically and creating a shared space or environment, e.g., the Upper Atmosphere Research Consortium at the University of Michigan. Such collaborations or "colaboratories" will produce "artifacts", and these can be stored and annotated. Internet2 will involve using high-speed networks to structure all kinds of activities to produce artifacts and events that are intrinsically part of the network world. Past meetings will become "object" and managed content.

How will advanced networking possibilities change scholarship, change how business is done? Lynch does not believe in the apocalyptic futures for higher education that so many of us have heard pronounced by one or another pundit (or gloom-sayer), with some melding of Disney or Hollywood and education. This is an "extreme view", and the future won’t play out quite that way. More importantly, however it plays out, it will not play out on Internet2. Internet2 is not a platform for distributed education or distance learning, as commonly proposed. The speaker does see distance-learning applications for sharing specialists’ seminars from premier universities. For such shared seminars, it will be "easy" to get data across the bandwidths of Internet2. The more difficult part will be getting the cameras and wires into classrooms, deciding who creates the analog of TV Guide and deciding who stores and organizes. Internet2 will set a stage for libraries to rise to new content and new roles.

In closing, Lynch stated Internet2 can change higher education in the near term; it will be a test-bed in the long term toward future commercial applications; and, finally, it will become irrelevant in 2 to 10 years time, if it does its job, as commercial applications will come in. Perhaps then there will be an Internet3. For now, Internet2 will give us a "window" on what kind of commercial applications we will see on these ubiquitous broad-band networks in the future.