Cybernetics Behind the Curtain
April 9, 2015 11:57 AM   Subscribe

"Computers, once vilified and now championed, were constant in one thing: They amplified the virtues and deficits of the system that implemented them." The Tangled History of Soviet Computer Science.
The results of top-down computerization were devastating. New computer systems accumulated ever-increasing amounts of raw data and generated terrifying heaps of paperwork. In the early 1970s, roughly 4 billion documents per year circulated through the Soviet economy. By the mid-1980s, after Herculean efforts to computerize the bureaucratic apparatus, this figure rose by a factor of 200 to about 800 billion documents, or 3,000 documents for every Soviet citizen. All this information still had to pass through narrow channels of centralized, hierarchical distribution, squeezed by institutional barriers and secrecy restrictions. Management became totally unwieldy. To get an approval for the production of an ordinary flat iron, for example, a factory manager had to collect more than 60 signatures. Technological innovation became a bureaucratic nightmare.
The article's author, Slava Gerovitch, is the author of From Newspeak to Cyberspeak: A History of Soviet Cybernetics. Socialist Cybernetics have been discussed previously on MetaFilter.

Bonus, from the Nautilus article comments: the history of Hungary's "TPAs".
posted by Kadin2048 (15 comments total) 34 users marked this as a favorite
 
and go read Francis Spufford's Red Plenty
posted by Bwithh at 12:16 PM on April 9, 2015 [6 favorites]


I've had a fascination with Soviet Computing for a while and would love to see these old machines revived. The thing that piqued my interest most was the Setun system, which was a ternary computer.

Furthermore, and more about Cybernetics in general (and apart from "computing" in particular) it's a shame to not mention the grandfather of Russian/Socialist Cybernetics: A.A. Bogdanov (I didn't see any mention of him in the previous threads)...

He had a public argument with Lenin regarding his philosophy known as "Empiriomonism" (Lenin arguing against it in his "Materialism and Empiriocriticism" essay). But relevant to this, particularly, is his philosophy called Tektology which was a sort of precursor to systems theory.

Here's a couple links to monoskop copies of the now out of print translations to his work:

Version 1 (I believe this is the full book)

Version 2 (it says "Book 1" so it might actually have been published in 2 parts, haven't gotten too far into it - I think I ended up with the Gorelik (the first link) one over this one, but either way, here you go if you're curious).

He also did blood transfusion instruments on himself with others proclaiming that these did seem to have beneficial effects on his health. He ended up dying of a blood infection from these experiments. If you're looking for an old Bolshevik, one of the originals who did some interesting and fascinating work outside the political machine, it's a great place to look :)
posted by symbioid at 12:24 PM on April 9, 2015 [10 favorites]


A friend of mine used to work with an old Russian who got his PhD in mainframe technology in the USSR. Or as he used to put it, "I have a PhD in a field that doesn't exist anymore from a country that doesn't exist anymore."
posted by Itaxpica at 12:49 PM on April 9, 2015 [6 favorites]


This explains Tetris.
posted by srboisvert at 1:10 PM on April 9, 2015 [2 favorites]


I've always wondered if a centrally-planned economy could have worked if they had software around to do the central planning. Listing out the inputs and outputs of every industrial process and optimizing for maximum utility was impossible with paper and mechanical calculators, but it just might be doable with something akin to a modern ERP.
posted by miyabo at 1:54 PM on April 9, 2015 [2 favorites]


miyabo: they even tried that in Chile, they brought in Stafford Beer to help out. The CIA had other plans.
posted by idiopath at 2:05 PM on April 9, 2015 [7 favorites]


A large centrally-planned economy can work (sortof). Sprawling corporations with diverse integrated verticals and barely related business properties commanding capital bigger than some state GDPs are pretty much proof. I'm not sure how much computers are responsible and I'm not sure that means it all could've scaled to the size of the USSR, but there you are.
posted by weston at 2:19 PM on April 9, 2015 [4 favorites]


Soviet computing was constantly sabotaged from within, with multiple research efforts set off against each other in competition for scarce resources and political sponsorship. It was realised that IT was essential to implement an efficient planned economy, but only by enabling localised autonomy through efficient access to remote and local data, and sufficient processing and storage to make sense of it all. In other words, the tools to make everything work were too powerful to be allowed to be used.

Nevertheless, there was a strong campaign to build the rough equivalent of the Soviet Internet, based on indigenous technology (albeit developed in conjunction with Western companies such as Brtain's ICL, who was interested in bulking up enough to compete properly with IBM) and with a nationwide data network and a lot of back-end databases accessible to planners and management. Against that was a strong campaign to limit computing to non-networked devices based on East German-driven clones of Western technology (at first IBM, thence mini- and micro-computer tech). This was promoted internally as being a more efficient use of resources, as R&D efforts would be a lot less in hardware and negligable in software (which would just be copied wholesale).

The latter won, with a few dirty tricks employed such as leaking to the Guardian a 'Soviets plan Big Brother computer' scare which misrepresented the former ideas but did give their opponents enough material to kill them.
posted by Devonian at 4:38 PM on April 9, 2015 [2 favorites]


I vaguely recall that an old National Lampoon has some comics about "How we view Russia" through the cold war. I think the 1970's had someone who looked like Brezhnev confusedly staring at a bunch of telephones, cables, and printouts.
posted by ovvl at 5:52 PM on April 9, 2015


I don't know how much I buy into the author's premise of Russian rejection of computer technology. His article focuses on early computer technology in the 1950s. Hell, Americans were fearful of computers back then too. But I don't think the evidence is there over the long term. For example, the Russians were cloning the Intel 8080A chip in the 1970s, while it was still Intel's most modern product.
posted by charlie don't surf at 7:09 PM on April 9, 2015 [2 favorites]


Oh man, I get a warm feeling just looking at all those vacuum tubes.
posted by MtDewd at 8:38 PM on April 9, 2015


they even tried that in Chile

Project Cybersyn. Dig the control room! It looks like a conference room on the 2001: A Space Odyssey moonbase, but in 70's kitchen decor orange.
posted by thelonius at 3:33 AM on April 10, 2015 [2 favorites]


Russian cloning of Western computer IP was so rampant, you had engineers doing things like this: VAX - when you care enough to steal the very best
posted by kjs3 at 7:51 AM on April 10, 2015 [2 favorites]


Dig the control room! It looks like a conference room on the 2001: A Space Odyssey moonbase, but in 70's kitchen decor orange.

Pardon, please, my self-link, but I thought I'd share my hi-res retouched photo of the same room that I used as the cover image for my article on the subject.

I found and scanned a large black-and white photo of the control room, corrected it for lens distortion, filled in some of the blank screens using other photographs where they were not blank, then used color sampling from the smaller color images available to colorize the scan. It was a lot of work but I was happy with the outcome. Direct link to image.
posted by Hot Pastrami! at 8:22 AM on April 10, 2015 [6 favorites]


I don't know how much I buy into the author's premise of Russian rejection of computer technology.

I don't think that's really his premise at all; the author says that by the early 60s the political tides had turned very firmly in favor of computers (via the vehicle of "cybernetics") and Soviet thought was very warm -- at least in theory if not exactly in practice -- to the idea of central planning and economic management via computers.

What the Soviets were calling "cybernetics" back in the 60s and early 70s has today fragmented; some of the more far-reaching theoretical parts today seem to get lumped under "systems theory" (which today seems to be out of favor), but the more practical stuff is now "ERP" and you can buy it off the shelf from any number of commercial vendors.

Where I disagree a bit with the author -- and I haven't read his book or anything, which is presumably more nuanced than the article -- is that the West has really gotten the last laugh as far as industrial planning or automation is concerned. The comically high failure rate of ERP implementations, even in very limited cases like a mid-sized industrial concern, just makes the Soviet attempts at implementing what amounts to a hemisphere-sized ERP system all the more crazily ambitious. Even if all they did was succeed in generating a lot of paperwork (which is all that a whole lot of ERP systems do anyway), that's a build-the-Pyramids level of ambition.
posted by Kadin2048 at 9:24 AM on April 10, 2015


« Older droplet ballet   |   "I have never said these words before: good job... Newer »


This thread has been archived and is closed to new comments