Who Invented Packet-Switching?

Saint Aardvark writes "It's how the Internet works, and now who invented packet-switching is under dispute. A posthumous paper by British scientist Dr. Donald Davies disputes the claim by Leonard Kleinrock to have invented the technique, saying Kleinrock never took it beyond the case of a single node. Kleinrock, whose lab was the first node on Arpanet, is willing to concede that Davies invented the term "packet-switching.""

Comments

Let me get something straight

[Mignon]

A posthumous paper...

I thought it was "publish or perish." Now you're telling me it's "publish and perish"?

I'm glad I got out of academia.

Naming Rights and Name Switching

[Kailden]

Then again, who invented packets? Who invented bytes? who invented bits? who invented the binary system? (I heard it was sheperds who needed to count lots of sheep so the used thier fingers as representations of powers of two.) Who invented numeric series? who advance math beyond numeric purity? who invented thought? Why does everyone want to take credit for everything? Too many battles are fought over naming rights....when the history revisionists will ultimately decide...not any single inventor. I mean there are still works of music, writing, and math from the Middle Ages on up that are attributed to the wrong people...(who invented inflated biographies?)

What about Paul Baran?

[Florian Weimer]

According to Hafner's and Lyon's "Where wizards stay up late" (an interesting read), Paul Baran was the guy who invented packet switching in the ARPANET context (he called packets "message blocks"). But AT&T (the company which eventually had to provide the communication lines) wasn't very happy about this idea, so he stopped working on this issue. About the same time, Davies started his experiments (and so did Kleinrock). Kleinrock might have considered packet switching in his very early theoretical articles on data communication, but it's not that clear that he was the first one to do practical, large-scale experiments (this was Davies, IIRC), or to consider packet switching the ARPANET context (clearly Baran).

All just a bit of history repeating

[Aztech]

When it comes to British inventors/inventions this is all too common occurrence, there is some great innovation in the UK but traditionally they concepts aren't followed through to commercialisation.

It happened to Sir Frank Whittle and the jet engine and consequently the first supersonic fighter, the Bell 1 which was based on the British design after the British Government withdrew funding for the project.

There was also the debacle over public key crypto research at GCHQ.

Donald Davies worked a the National Physical Laboratory in Middlesex, unfortunately the British Govt/Grants agency didn't see the potential of the invention at the time and no funded was given, so he went over to APRA who were throwing money at anything.

Donald died June last year at in Australia, where he went to retire, he didn't get a lot of recognition outside of a few small circles, but he did get quite a few awards from the various computing institutions in the UK, I think he's still relatively unknown in the US, probably because he was too modest, which is why some many scientists can claim to have invented Packet Switching.

This is a big mess.

[The Cunctator]

This is something I've done a lot of research about, and in fact have discussed this issue with Kleinrock and Stewart Brand (who is pro-Baran & Davies), and read Davies' paper. The first thing people should know is that Katie Hafner is the author of "Where Wizards Stay Up Late", a book she assiduously researched but which many of the participants within claim she did a bad job on. I think they might be confusing some of it with Janet Abbate's "Inventing the Internet", which I wasn't terribly impressed by.

This is Hafner's passage of interest:

By the end of July, 1968, Roberts had finished drafting the RFP...It was a rich piece of technical prose, filled with an eclectic mix of ideas. Kleinrock had influsenced Roberts's earliest thoughts about the theoretical possibilities. Baran had contributed to the intellectual foundation on which the technical concept was based, and Roberts's dynamic routing scheme gave an extra nod to Baran's work; Roberts had adopted Davies' term "Packet" and incorporated his and Scantlebury's higher line speeds; Clark's subnet idea was a stroke of technical genius.

Then she continues with a quote from Baran that "Everything is tied to everything else" with respect to who did the most important part.

It's weird, because from my perspective the participants seem to be arguing and use strong language like "spreading lies" to describe the alternative history, but when you look at the specifics, the dispute lies on some very fine nuances which are evidently impossible to untangle now, and may only be creations of recent times.

The number one question, to my understanding, is whether packet switching is such a central concept that the work by Baran and Davies which details it (since they both built experimental packet-switching networks and then wrote extensively about them, providing a base of information for Roberts) is important, or whether it really should just be understood as a relatively arbitrary (and self-obvious) implementation of multiple-node store-and-forward queuing theory, which Kleinrock is the father of, no question.

Did Baran and Davies' work matter to the ARPANET? It pretty much has to have. Baran wrote multiple volumes of detailed information from his experimental network; those volumes were available to and used by Bob Taylor, Roberts' boss and (according to Brand, at least) in the Baran camp, and Roberts credits them heavily in his early work.

All the early documentary evidence points only to Baran and Davies. However, the close association of Roberts and Kleinrock, the fact that Roberts helped Kleinrock do his thesis by doing programming for it (funny fact: the third guy in their little Lincoln Lab thesis group was Ivan Sutherland), and Kleinrock's lab's role as the first IMP site and ARPANET analysis center makes it absurd to believe that Kleinrock's influence wasn't major.

Of course, framing the dispute this way ignores how crucial the work of BBN was in all of this; they were amazing in designing and building the IMP. While Roberts' RFP had insane amounts of information, the IMP Guys did equivalent amounts of new work and recreation of ideas in their proposal.