Merit Network , the pioneering Michigan research network that played a key role in the development of the Internet, this week will celebrate its 40th anniversary. President Emeritus Eric Aupperle, a 32-year Merit veteran who served as director and president during the organization's heyday in the mid-1980s and 1990s as the operator of the National Science Foundation 's NSFnet, spent some time reminiscing and looking ahead with Jim Duffy.
Can you highlight Merit's key accomplishments over the past 40 years?
There are an awful lot... One of the very first would be to allow the student and faculty and staff of [Michigan's] three major universities to share and access and use the computing resources of all three, and build a network to do that. Merit's task was to build that network. In those days, it was not possible to go out and buy off-the-shelf router technology, as one can do today. So one of the first things we had to do was develop technology -- essentially routers, although at that time we called them communications computers. We built three of them -- one to be attached to each of the three mainframe systems and interconnected by phone lines. The design and development of that was certainly historic in terms of an accomplishment. The only other organization that did something similar in a comparable time frame was ARPA with ARPAnet and IMP technology.
In the mid-1970s, we added the ability to dial in directly to our network. In the early 1980s, we modified our software so we could also carry the then-DARPA protocols, the TCP/IP protocols. I think we were the only network that not only ran our own protocols but also the DARPA protocols.
Then probably the next major thing was when Merit teamed up with IBM and MCI in 1987 to bid on the NSFnet activity. We successfully won that bid and had a T-1 network up and running by mid-1988. We upgraded that to T-3 and then closed it down in 1995. Gopher had been developed and the Web had been developed but the spawning of the dramatic growth of the commercial Internet followed right on the heels of the NSFnet activity.
I've often commented to people that the National Science Foundation invested something on the order of US$50 million in the networking activities associated with the NSFnet. Fifty million dollars -- when you consider the expenditures of the federal government -- the relative payoff is just incredible. A great return on investment.
What do you think have been the biggest accomplishments in the industry over the last 40 years?
When we started in the late 1960s and early 1970s, we used teletype to VR terminals and we used punch paper tape to load the software used on our communications computers. The next major thing that happened was the development of personal computer technology, so one of the major developments early on was replacing teletypes and punch paper tape with PCs.
The development of the World Wide Web has just really transformed the way the Internet was used relative to the much more awkward and arcane procedures that were involved before.
I think probably the two most significant developments over that period of time were the cost and improvement of PCs and their abilities to have, over time, much more memory and improved use of technology -- going from floppy disks to DVDs. And then the related software developments, the Web being probably the most dramatic of those.
It would have been impossible to envision how much has been accomplished in such a relatively short period of time.