Interview with Dean of Computing Ira Fuchs
Nassau Weekly Article (11/7/85)
By Randy Schoenberg
The University has recently appointed Ira Fuchs the new Vice President for Computing and Information Services. Nassau talked to him about this plans for Princeton:
Nassau: Why do you think the University established your position?
Fuchs: I think there was a need for some focal point that could oversee computing for all aspects of the University. There are areas which require a more or less uniform approach to solving problems. The campus network is a good example. It is unlikely that we would want to have two totally independent networks for administrative use and for academic use.
Today there are small computer clusters in the residential colleges, and I hear that you may be planning to build some others in the upperclass dorms. Where do you see the network going in terms of undergraduate use of computers?
I see one goal as generally increasing the accessibility of computing, not just to the student body, but to the faculty and also to the administration. Specifically with respect to the residential colleges, it seems that the Forbes College pilot has been a success, and that it would make sense to replicate some or all of what Forbes has done at the other residential colleges.
As for the upperclass dorms and other buildings that undergraduates normally have access to, I think the goal in general is to extend the cable that was put in this summer to reach the remaining buildings, and to then wire each of the buildings to permit local area networks to extend to any place that one could rationally want to connect with a computer.
All connected to the Computer Center?
Once you have a local area network, and you can have any machine talk to any machine, the computing center can be viewed as a computational server on the network, the same as a machine in the Physics Department or a machine in any department, or the machine in Forbes College. In other words, they are all resources available on the network accessible to anyone who has access to the network and who has authority to use that particular source.
The same goes for high quality printing services. It’s one of my goals to improve or to increase the access to high quality printing. I haven’t witnessed it yet, but I’ve been told that at the time when junior papers and senior theses are due, there’s a great crunch of people trying to use the laser printer at the Computing Center. The cost effectiveness of distributing that particular resource I think is such that it doesn’t make much sense to force all the printing to be done at the Center. The thing that sort of ties it all together is the network. But along with doing that you have to figure out how to put the services on the network and make them easy to use and available anywhere.
Is the Network you are talking about TIGERNET?
There seems to be some confusion, at least in some people’s minds, over what it is that people mean when they say TIGERNET. My understanding is that it is a broad-band cable that was initially run between the Computing Center and the E-Quad, and has now been extended to somewhere around half of the campus. The goal would be to extend that further to reach all of the buildings on campus.
At Dartmouth there are terminal hookups in every room, and every freshman is required to buy a computer. If a student can’t afford it he has to take a loan. Using the network a student can access library listings, have conferences, and access the mainframe. Is Princeton going to go toward something like that?
I think in terms of the accessibility of resources like the IBM 3081 at the Computer Center or databases that might be available on various machines on campus that the answer is yes. We will seek to wire up dormitories and wire up in the rooms, so that a student who has a Macintosh or PC or some other compatible device will be able to plug it into the wall for access to the network.
I don’t foresee in the near future Princeton requiring that every student purchase one of these computers. On the other hand, the price of these computers is coming down almost on a weekly basis, and I think more and more we’re going to see students coming with their own computers. It may be that the University will encourage it by having access to all these resources. It has encouraged it already in terms of entering into pricing discount arrangements to make it less expensive to students. I suppose there is the possibility that the University could go further and provide financing or credit for acquisition of machines. I haven’t heard any discussions about that, but other universities have done that, just short of simply saying bring your own computer.
I guess as long as you have terminal hookups in the rooms…
…And clusters in as many places as is practical. That means that if a student doesn’t have his own computer he isn’t going to have a hard time getting one.
It leaves it up to the student.
How is the supercomputer at the Forestal Center going to affect the entire University?
The supercomputer will shortly be connected to TIGERNET. There will be a high speed connection from the new John von Neumann Center to the Computer Center and the ultimate goal is to make that supercomputer look like yet another server, another resource on the network throughout the University. That doesn’t mean that any student plugged into a computer terminal would automatically have his or her programs run on the supercomputer. If they had accounts there and if they were doing research which was funded by the National Science Foundation then, potentially at least, they would have access to the center. But the goal would be to make it as easy to send programs and jobs and to receive output from the supercomputer center as it is to the current central facility or any other machine.
Does the University have a goal towards centralization or do you think they are going to rely on you to formulate that? Does the University have an official stance on centralization?
I think that the University stance is that computers are important in education and by all accounts they appear to be becoming more important, that is to say they’re being used in more disciplines, and there’s more software being written that makes the use of computers important and truly useful in teaching and in research. Computers are obviously also used in the administration and management of the University. I don’t think there are too many people who see a reverse trend. On the other hand, there’s no intent to try and turn Princeton purely into an Institute of Technology, and that’s not what we’re trying to do.
I think that in talking about centralization, I don’t see what I’m doing, or what the University seems to be committed to, as implying greater centralization. I think that the University is looking for leadership, and trying to create a coherent environment for the use of technology at the University. In some ways, I suppose some people could view it as a further emphasis on some form of decentralization of computers. In other words, by making it possible for all of the departmental machines and distributed machines to be used in an equally easy way, you are in some ways encouraging distribution of resources.
Right now if a student is in a computer science course or a physics course, and he needs to use a computer for it, he has to use a computer, not at the Computer Center, but at the E-Quad or at the physics department. Do you see those departmental computers being linked up so that someone could work from the computer clusters or from his room?
Absolutely. With the cable reaching all the buildings, and with the local area networks gateway to the cable, and then with the appropriate set of standards or protocols used by all the machines on the network, it will be possible for any work station, any personal computer, to ask to connect to any of the servers on the network.
How much time and money is it going to take?
I really can’t estimate that yet, it’s something that I was working on when you came in. It’s going to be a significant project. I would hope that it could be completed in the next couple of years, and one of our problems will be to decide which pieces get done first. I certainly think it’s achievable and I think that there is a commitment on the part of the administration and of the faculty to see it happen.
How do you think Princeton stands now in comparison to other institutions as far as computer accessibility goes?
Better than some, worse than others. There are plenty of campuses that have not yet, for example put in the kind of broad-band cable systems that TIGERNET represents and that are planning to do so. And that’s a significant and expensive effort and it’s one that Princeton is already well on the way to completing. The development of local area networks on the campus is perhaps slower than some of the Universities that are at the forefront—Stanford, MIT, Carnegie-Mellon, and so on—but we’re not that far behind.
The technology is changing very fast in this particular area. So, it may be the case that it’s fortuitous that we haven’t made too many decisions yet, because we get to take advantage of all the new advances. Some people think that the best strategy in computing is not to do anything because it’s always cheaper and better if you wait a little longer. But then you don‘t have anything either. I think though that in the case of Princeton, it’s the right time to be doing the things we’re doing. Five years ago, if we had tried to make some of these decisions, it probably would have been quite difficult, if it was even possible. And yet I don’t see any point in waiting another five years to see how it all turns out at other universities. I don’t think we can afford to do that.