At 06:14 PM 11/18/03 -0800, Corey Bailey wrote:
Let me work on that one Dave.
I'm relatively sure that the forerunner of TCPIP was conceived by a UCLA professor. I read somewhere that the first loop tests of the protocols were between UCLA and San Diego state. It was called Cerfnet.

Corey, I can help you with that one, too -- at least a little.

I never heard of Cerfnet (at least not in that era), but it must refer to Vint Cerf. He had been at UCLA, but he moved to Stanford in 1972, just before he started collaborating with Bob Kahn (who had moved that year from BBN to ARPA -- incestuous organizations? or just a rather small community of interest; probably the latter, see below). In late '73 or '74, they published a paper, "A Protocol for Internetworking" or something like that. The protocol was indeed the forerunner to the IP protocol. The TCP protocol came later, and I don't remember if it was Cerf/Kahn or others who had the biggest input into that one.

While I still support Larry Roberts as THE father of the Internet, Cerf & Kahn have much more legitimate claim to the title than Licklider, IMHO.

The Cerf/Kahn IP paper was published in the IEEE Transactions on Data Communications, a "refereed journal". That means that any paper published is first reviewed by experts in the field. Adam Lender (the editor of the journal) flattered me by inviting me to be one of the three reviewers.

BTW, my review was mostly favorable; my main objection was the lack of accountability (what we'd think of today as security or identity assurance) in the protocol. That was never rectified. I guess I should be glad of that; from 1997 to 2000, I made a very good living implementing an Internet firewall to try to give users the security that the protocol never did.

Back to Internet contributions from UCLA:
* Wesley Chu is generally credited with the first analysis of packet multiplexing, published in the early '70s when he was at UCLA. But he was at Bell Labs in the '60s, and knew Randy Pilc and Pat Marino there -- who had done earlier, similar analyses but were prevented from publishing proprietary info. (I have some personal pain -- and a rueful grin -- about that.) BTW, Wesley shared a house with my brother and another bachelor engineer when he was at Bell Labs -- yes computer networking WAS a real small world in those days.
* Leonard Kleinrock of UCLA did a lot of the packet traffic theory in the '70s, when it was a very hot topic. I didn't meet him personally until 1977, when he came to Bell Labs to give a guest lecture to a course I was teaching.
* Jonathan Postel, who went on to become perhaps the most important protocol innovators for the Internet, started his networking career as a graduate student at UCLA in 1969. But I don't think he did his most important work there. That probably happened in the mid-'70s at MITRE (Massachussetts) and SRI (Palo Alto), then back in the LA area at USC in 1977, where he made many major contributions until his passing in 1998.


Hope this sheds some light on your comment and, if you choose to research it further, gives you specific enough handles to use Google fruitfully. And I won't apologize for the name-dropping; it was fun for this retired Labbie to wander down memory lane.

Cheers!
DaveT




Reply via email to