come_to_think (come_to_think) wrote,

Prehistory of the Internet

Reading:  M. Mitchell Waldrop, The Dream Machine: J.C.R. Licklider and the Revolution That Made Computing Personal (Viking, 2001).  Jim gave it to me.

An intellectual and business history of computer development from the 1960s to 1990 --- the prehistory of the Internet, woven around the career of one "Lick", whom I had never heard of but who turns out to have been pretty important.  He had a vision, even during the mainframe era, of linked computers for the masses, and he was good at assembling & inspiring smart people and raising money for them from various government, academic, & business bureaucracies (they seem not to be all that different).  I couldn't actually follow the story (far too many names & acronyms to tell apart), but it is pleasing (tho chastening) to read about people who know their business and are doing something worth doing.

I suppose the vulgar short title was supplied by Viking.  The book itself is written in a vigorous colloquial style (sometimes a bit breezy) with remarkably few obeisances to journalese.

Tho some of the older people in the book (including Licklider) died before Waldrop started researching it, he did get to talk to very many of the younger ones, and of course there is a lot in print or in archives about their work, and he seems to have read it all.  I occasionally wondered how he knew what was going on in people's minds, but the copious quotations are all referenced, and I am inclined to trust him not to have made anything up for the story's sake, the way people do these days.

In a couple of places he mentions things I happen to know a little about, and (this always seems to happen) he makes me raise my eyebrows.  Some of his heroes were academic psychologists at some point in their lives, and they participated in the change of fashion from behaviorism to cognitivism, so he describes behaviorism at some length --- mostly, it seems, on the basis of cognitivist folklore, which he ought to have looked up.  In particular, the following sentence, about B.F. Skinner, is disgraceful:

He had raised his own infant daughter partially inside a baby-sized, Plexiglas "Skinner box" rigged for the appropriate rewards and punishments....

The box was not a Skinner box (tho it was a box, and invented by Skinner); it was an air crib.  A real Skinner box is a laboratory apparatus for training small animals, and it is indeed equipped for rewarding them (Skinner did not like punishing).  An air crib is a crib, that is, a place to put a baby to sleep (is that what "partially" "raising" one means?); like most cribs, it is not "baby-sized", but a good deal bigger than a baby (  It may contain a few toys & decorations, but it does not reward or punish.  It differs from ordinary cribs in being enclosed, so that the temperature & humidity can be controlled and the baby does not have to be bundled up.  Behaviorism played no role in its invention; it could have been invented by a Freudian or a Marxist or an imaginative nurse with no academic training; and the article cited suggests that its chances of adoption would have been better if it had not come from the notorious Professor Skinner.

I hasten to say that I am not & never have been a behaviorist.  I was brought up Freudian, and I have since become skeptical of all attempts to develop a technical language for discussing human behavior that does better than ordinary mentalistic language with critical additions & deletions.  I do not admire Skinner as a philosopher of science; I think he had a bad case of physics envy, which (as usual) he might have been disabused of if he had learned more physics.  I thought Chomsky's hatchet job on Verbal Behavior was well deserved.  But I do admire Skinner as a moralist.  We need a lot more of his experimental approach (Yankee ingenuity!) to making life better in this or that detail.  The air crib is one example; his utopia Walden Two contains plenty of others, well worth sorting out from the occasional foolishness.  (The commune I belonged to, Twin Oaks, was originally inspired by Walden Two, and has managed to confirm some of its suggestions.)

In his description of SAGE --- the network of state-of-the-art (vacuum-tube) computers that was built around 1960 to coordinate continental defense against Soviet bombers --- Waldrop is of course mainly concerned with its role in advancing computer technology.  He does make one comment about what other use it may have been:'s hard to say how effective the SAGE system really was in military terms, since it was (fortunately) never used in combat.  Arguably, in fact, the SAGE system was obsolete almost from the day it was commissioned, since by that point the United States and the Soviet Union were hard at work on intercontinental ballistic missiles that could be used to deliver a nuclear warhead across the North Pole in under an hour....

On the basis of chitchat I heard at Lincoln Lab about that time, saw in the press thereafter, and am able to confirm in part with Google, I think he might have gone farther:  SAGE was a kluge and a boondoggle --- the SDI of its era.  At best it was supposed to shoot down 75% of the incoming bombers, never mind missiles.  Also, it required a continuing programming effort to integrate new weapons into it.  A sizable bureaucracy was set up for that purpose, and it never caught up, even with our own improvements, much less the Russians'.  Years afterward, SAGE was being cited in some circles as a good example of what not to do with computers.  But who knows?  If a war had happened soon enough, it might have saved a few of us.

Oh, and PM wasn't a "socialist magazine".  It was a somewhat fellowtraveling daily paper.  And "sputnik" doesn't mean "little traveling companion".  It is the standard Russian astronomical term for satellite.  It also means traveling companion, but the "little" doesn't belong there at all.  "-ik" is indeed a diminutive suffix, but "-nik" is not.  (Bitch, bitch, bitch.)
Tags: behaviorism, internet, licklider, sage, skinner, waldrop
  • Post a new comment


    Anonymous comments are disabled in this journal

    default userpic

    Your reply will be screened

    Your IP address will be recorded