Slashdot Mirror
ACM to Honor TCP/IP Creators with Turing Award
bth writes "The New York Times reports that Vint Cerf and Robert Kahn will receive the ACM Turing Award. According to the ACM website: The Association for Computing Machinery, has named Vinton G. Cerf and Robert E. Kahn the winners of the 2004 A.M. Turing Award, considered the "Nobel Prize of Computing," for pioneering work on the design and implementation of the Internet's basic communications protocols." Commentary from Groklaw also available.
TCP/IP has played a pivotal role in the revolutionised age of information and communication.
Seriously folks, I think this news really fits the "news for nerds. stuff that matters " slogan.
"There is no flag large enough to cover the shame of killing innocent people."--Howard Zinn
What about Donald Davies and Paul Baran, the guys who invented packet switching in the 60s? Their work directly led to the development of the first internet protocol, NCP. TCP/IP didn't replace NCP fully until 1981, although we should be glad it did.
Congratulations to some truly innovative pioneers.
You'd have thought they would've received this during the dot-com boom or before that.
OCO is Loco
It's suprising the people who architect some of the finest PC ideas are not recognized more by the media. Everyone knows who Bill Gates is, but when you ask someone who were some of the people behind TCP/IP or C++ or anything besides windows, they have no idea.
Seriously though ... award them for creating the TCP layer, which breaks down massively under (non-congestion related) packet loss? Award them for creating IP, which trivially allows source address forgery in yet another DDoS against my IRC server? ... in fact it was essential for the Internet today, and they deserve praise, but what they did was far from an optimal implementation. Considering all the mad fanboying going on here, I just felt I had to post this.
I'm not saying what they did was all bad
http://en.wikipedia.org/wiki/TCP/IP#Layers_in_the_ TCP.2FIP_stack
Kahn and Cerf deserve credit they are getting but not based on the mere fact that the whole world uses TCP/IP. I mean to say that if you'd reason merely by size then good ole Bill would be a candidate for the Turing award. The reasons why IP has become the default network protocol should be stated more clearly.
IMHO the genius of Kahn and Cerf lies in the fact that they "thought deeply of simple things" almost exactly like Thompson and Ritchie did with Unix. For me, the transmission error handling and the routing are simply beautiful.
If a packet is lost, IP and UDP simply don't care and neither should the underlying layers do (forget about x.25 for a moment.) Try explaining this apparently frivolous approach to an IBM SNA guy -or even to most non networking CS people. Hell, IBM even built quality of service stuff in their Tokenring stuff. Nice to have, if you can switch it OFF. If a packet or frame is lost: too bad, TCP will take care of it, anything else should stop whining about it.
The fact that part of the routing is done by IP on any node is also marvelous. It made the protocol usable in small networks without having to buy or explicitly set-up a router. You know, equipment used to be horribly expensive. Ever studied SNA or OSI?
There would be loads of jobs for us techies in supporting the Internet if it were made up SNA, OSI or NetBIOS. But who'd want them?
Would Metcalf deserve the same honor as Kahn and Cerf but then for inventing Ethernet? I'd say yes.
I hadn't the slightest objection to his spending his time planning massacres for the bourgeoisie... (P.G. Wodehouse)
It's a serious insult to compare this to Einsteins innovations. This just strengthens the view that computer science is a pseudo science like social sciences in the scientific community.
Well, *most* sci/tech awards don't measure up to Einstein's work. Those are big shoes to fill. Perhaps they should rank the awards, or offer a Century Award for the biggies. Einstein didn't even get a prize for relativity, it was something slightly more obscure IIRC.
But you are right in that much of software and computers revolves around "soft" issues. Nobody can "prove" that relational or OOP is better then what came before/aside them. That is why we have so many language and paradigm holy wars. But this just means that software is transcending the physical world, and the variations and possibilities of virtual worlds is wonderous, personal, and confusing; making judging difficult.
Table-ized A.I.
Thank you for posting these links. I never knew Jon Postel, and I was a toddler whenever RFC #1 came out in the very early 1970's (and I'm just a plain old midwestern hacker-for-pay now.) But reading Cerf's remembrance of Jon Postel always make me cry, like right now.
What a strange beast, the Internet, which can be a vessel of human connection, understanding and sharing of feelings, aside from all the latching shift registers and so forth.
Mr. Morse transmitted over an early electronic network, "What hath God wrought?" Don't know the answer to that, but I do know what Morse, Cerf, Postel and others hath wrought.
Thanks for reminding us.
award them for creating the TCP layer, which breaks down massively under (non-congestion related) packet loss?
The greatest majority of traffic on the Internet is TCP acknowledgments (35%), meaning that TCP is the most used transport layer protocol of the few other alternatives. If it is as bad as you say it is, why is everybody using it ?
If you're such an expert, spend time fixing the problems you think exist, by contributing to the IETF, rather than running an IRC server, and complaining anonymously about DDoS attacks on slashdot.
The Internet's nature is peer to peer - 20050301_cs_profs.pdf
It seems to me that TCP/IP is an fine engineering result that has benefited from being in the right place at the right time. If circumstances were different we would be lauding the inventors of Banyan Vines or DECnet or some schlock M$ protocol. Thankfully we are not. But the idea of associating workmanlike engineering results with a theoretical genious like Turing and other deserving winners of the Turing Award is irksome.
an ill wind that blows no good
It's no doubt that we would speak about Internet protocols a little differently had these guys not done what they did, but to me it seems like we'd just be saying some other acronym (does anyone really buy that they invented the idea of packets and it didn't come about until 1973?) They invented the basic scheme, but the real cleverness seems to have come as a result of the various exponential-backoff mechanisms and other complexities in today's implementation of TCP/IP, not the basic protocol they designed in the 70's.
Looking at the previous winners it's kind of hard to tell what the point of the Turing award is. In some cases it's given to researchers that have made very influential theoretical break-throughs and others that seem to have invented something that became popular. Maybe I'm just being sidetracked by what is essentially the old debate about whether "systems" research is true research since it's often difficult to comparatively evaluate alternatives.
I just like to see the award go to people that did something that no one else (or at least very few people) working at the time would have been likely to think of and I'm not sure this meets that criterion.
Remember, it's not how popular it was, it's how important and valuable it was. I thought the Turing award was, unofficially, only for deep theory shit.