R.I.P. Paul Baran

March 28, 2011 10:24 AM

The father of packet switching - Mr. Baran’s invention was so far ahead of its time that in the mid-1960s, when he approached AT&T with the idea to build his proposed network, the company insisted it would not work and refused.
posted by BigHeartedGuy (24 comments total)

I haven't read it yet, but everyone is raving about this profile of him from Wired in 2001.
posted by mathowie at 10:32 AM on March 28, 2011 [5 favorites]

There's an old joke that AT&T was a billing company -- that's what they produced. Bills. The entire architecture was centered around being able to directly account for every minute of every phone call.

There's an interesting argument that, by that standard, AT&T was right about packet switching. Say what you will about circuit switching, but it does create an end-to-end context you can throw onto a bill.
posted by effugas at 10:33 AM on March 28, 2011 [5 favorites]

There's an old joke that AT&T was a billing company -- that's what they produced. Bills.

Nice to know things haven't changed.
posted by Mister Fabulous at 10:37 AM on March 28, 2011 [1 favorite]

Just as well ATT turned him down; otherwise we wouldn't have anything like the Internet with all the patents on the basic underpinnings.
posted by Old'n'Busted at 10:40 AM on March 28, 2011

When Wizards Stay Up Late
posted by jcruelty at 10:46 AM on March 28, 2011 [1 favorite]

*Where* Wizards Stay Up Late is a great book. Highly recommendo!
posted by stenseng at 10:57 AM on March 28, 2011

. . . .
: :
::
posted by orthogonality at 11:04 AM on March 28, 2011

From the Wired interview linked above:

Somebody was doing a study on termination of wars. How the hell do wars stop? Interesting problem, but Congress got all pissed at the idea. They even passed a law forbidding government-funded defense researchers from studying surrender. They were afraid that somebody would think our study of surrender would indicate that we were exhibiting weakness. So the study of surrender continued, but you didn't call it that. We didn't emphasize that communications was important in cooling things off; we did emphasize getting the word around to go fire your missiles.

Sometimes certain terms take on a meaning of their own and become real. One was "minimum essential communications." The military said all they wanted was "minimum essential communications," and I believed them. So I thought data rate would take care of everything - get the word out, calm things down if necessary. You don't need a hell of a lot of communication for that.

posted by migurski at 11:07 AM on March 28, 2011

Other bits from the Wired interview:

The AM radio packet switched steganographic network of 1960:

So then I picked up on an idea from Frank Collbohm that the problem is the military depends heavily on high-frequency communications. A high-altitude nuclear burst takes out the ionosphere11 for many hours. So the only thing that was left was the ground wave12 - that's what you get from broadcast stations during the day in the short range. Collbohm's idea was for the radio stations to relay the message from one to the other. But there are a lot of them in the US. So I said, "Let's automate it." That would make it practical.

The first crack that I took at it in 1960, I got an old Johniac computer and a plotting board, and I plotted the locations of all the AM13 radio stations in the US. Yeah, there's plenty of paths; I said look at the range.

That went off in two directions.

One, I went out with a briefing chart, saying, "OK, here's the solution to your problem." I got push-back from the military: "That takes care of the president getting word to the missile, but what about me? I've got to speak to the troops. I've got to do this and that. I need more communication."

Meanwhile, the Air Force took the idea and gave it to Rome Air Development Center. They built it as a teletypewriter system and tested it. It worked just fine. And they did something cute: They used the AM radio stations, but slightly wobulated their frequency - around 20 Hz. You couldn't hear it on your radio, but it let us send a frequency-modulated14 teletype signal.

Was that implemented?

Yeah. It was implemented, tested with a dozen stations, and it all worked fine. That may have been the first packet-switching system.

The virtues of openness:

Let's put this in the Cold War context: The idea was to have basically retaliatory capabilities, so you wouldn't have to use them. But that only works if the other guy knows about it. What was going on in terms of visibility? Were you sending your papers to the Soviet Union?

We kept everything open that we could.

And trailed it in front of known spies?

We published it! I gave a course on it at the University of Michigan in '65. We were a hell of a lot better off if the Soviets had a better command and control system. Their command and control system was even worse than ours.

That sounds pretty enlightened. This was the Air Force saying, "We want the enemy to have the same second-strike capability as we do."

There's certain things you don't say. But yes.

posted by zamboni at 11:11 AM on March 28, 2011 [6 favorites]

Speaking of old school AT&T, I love this psychedelic video on microchips featuring William Shatner.
posted by exogenous at 11:27 AM on March 28, 2011

when he approached AT&T with the idea to build his proposed network, the company insisted it would not work and refused.

Situations like this always make me wonder how often big companies like AT&T are right. Sure we call say, "Haha, stupid AT&T you shouldn't have passed on this. Dumb Verizon, you shouldn't have passed on the iPhone." Etc. But how often does someone come to them with an idea that seems far ahead of its time, they tell the inventor that it won't work, it gets implemented by someone and fails?
posted by VTX at 11:37 AM on March 28, 2011

.
posted by jquinby at 11:38 AM on March 28, 2011

VTX, what you're talking about is essentially the heart of the book The Innovator's Dilemma. The main argument the author makes is that companies consistently make a series of short-term rational decisions that undercuts their ability to compete in the long-term.

It's also worth pointing out that a lot of ideas fail. I'd go far as to say that the majority of ideas fail, often times for reasons outside of the quality of the idea itself. Too early, too late, requires too much Kool-aid, incompatible, too expensive, can't produce enough, can't scale up, no business model, and on and on and on.
posted by jasonhong at 11:49 AM on March 28, 2011

.
posted by Xoebe at 12:08 PM on March 28, 2011

The early bird may get the worm, but it is the second mouse that gets the cheese.
posted by yesster at 12:57 PM on March 28, 2011 [1 favorite]

There is a component of that. Mostly I'm bothered stories about the big dumb corporation (they think they're so smart!) did something dumb without anyone asking if they usually get it right. I don't know if they do but it bugs me that everyone takes it for granted that they should have given some new tech a chance.
posted by VTX at 1:02 PM on March 28, 2011

Then there''s the issue that ideas are easy, implementation is hard. You should ignore the bright ideas folks. But you should probably not ignore the bright idea guy with an implementation strategy. Hence the funding and eventual buyout of startups.
posted by yesster at 1:05 PM on March 28, 2011 [1 favorite]

.
posted by lothar at 1:15 PM on March 28, 2011

"Better dead than Red" was a very common sentiment in those times, always spoken without any irony at all.

What does "dead" mean? What Baran says about the DoD people vis-a-vis TNW may or may not be reliable. There were many implications that were not just not understood, but which weren't even on the radar yet.

Eventually some people realized that it would mean "Everything". I appreciate their existence a hell of a lot more than that of functionaries like Baran. Screw him. Screw them all.
posted by Twang at 2:20 PM on March 28, 2011

.
posted by ZeusHumms at 2:31 PM on March 28, 2011

When Samuel Morse and Guglielmo Marconi died, there were universal moments of silence held across the mediums that they helped create -- the telegraph and the radio.

It's interesting to note that it would be unthinkable to do such a thing on the Internet today. First, because there is no single person hailed as its 'inventor' in the same way that Morse or Marconi were by the time of their deaths (in both cases somewhat unfairly, since both built heavily on the work of others and were in large part successful marketers and showmen), but mostly because it's simply not feasible or even possible to shut down the modern network for any meaningful amount of time.

That, I think, is a greater testament to Baran and the many others who helped create the modern communications network; not a moment of silence but the absence of it.
posted by Kadin2048 at 4:04 PM on March 28, 2011 [2 favorites]

.
posted by PHINC at 6:38 PM on March 28, 2011 [1 favorite]

About 8 years ago I was enraptured with BBN. First 6 nodes of the Internet, Internet Protocol, campus in Cambridge located betwixt MIT and Harvard.

I had worked for about 10 years in infosec...and BBN was the penultimate genius hot spot in my mind.

So, while I worked at another infosec bleeding-edge company, I applied to BBN.

Took about 2 months to get them to fly me in for an in-person interview.

On the plane I read "When Wizards Stay Up Late at Night."

I got the great campus tour, ending with the loading dock where the first 6 nodes were shipped out. Damn, nearly fainted, such a nerd.

They hired me, and I was lucky enough to work on the re-architecture of the infosec for NYSE.

On the days off, I'd sit and listen to the elders tell me about the early days of BBN.

Finest company I ever worked for, and learned more at BBN than anywhere in my 20 years in infosec.

Rest well, Paul. You did good. VERY brilliantly good.
posted by Dunvegan at 9:03 PM on March 28, 2011 [2 favorites]

BTW, BBN hired me back then even though I was a "girl."

One of two women working at BBN when women "didn't do" infosec.

Truly, BBN was a meritocracy...when almost no other company of their caliber was.

And Paul was a formatively decent person. It was all about your ability with him, and also, what he saw you could learn in a wicked fast manner.
posted by Dunvegan at 9:34 PM on March 28, 2011 [1 favorite]