Staging the OGAS, 1962 to 1969 109
network efficiencies, which were optimized to serve both national and local
needs, would achieve full effect by 1990, nearly thirty years after the Komso-
mol young scientists delivered their letter into Khrushchev’s hands.
As originally envisioned, the OGAS had several distinct features. Perhaps
the most meaningful contrast with that of modern networks is that the
OGAS was modeled after the economy of a factory writ large for a nation.
The basic unit of the OGASU was, as the initials imply, the ASU, the auto-
mated management system, or a local information and control system that
looped onsite mainframe computers into the industrial processes of a fac-
tory or enterprise to provide real-time information feedback, control, and
efficiencies. This kernel vision of a network as an expression of the nervous
system of a factory, writ large across a nation, magnified the image of the
workplace until it incorporated the whole command economy—a sort of
simultaneously metaphorical and mechanical collectivization of the indus-
trial household (or what Hannah Arendt calls the oikos).
The OGAS Project might be seen as preceding, although not precipitat-
ing, the current trends in so-called cloud computing. The national network
was to provide “collective access,” “remote access,” and “distance access”
on a massive scale to civilian users who could “access,” “input,” “receive,”
and “process” data related to the command economy (such older terms
appear to bear more descriptive heft than the modern computing meta-
phors such as upload, download, share, and stream). The decentralized net-
work was designed so that information for economic planning could be
transmitted, modified, and managed in relative real time up, down, and
laterally across the networked administrative pyramid. At the base of that
pyramid, in the network’s initial vision, were as many as twenty thousand
computer access points and ASUs distributed throughout the nation’s enter-
prises and factories. This base of computer centers would be connected to
one hundred to two hundred midlevel regional planning decision centers
in major cities, which would be connected to the central planning process-
ing center in Moscow by high-capacity data channels. The original vision of
a three-tiered pyramid network—with twenty thousand computer centers
on the bottom, one hundred to two hundred in the middle, and one on
the top—was scaled back in the original design of the technical base of that
network (the Unified State Network of Computing Centers, or EGSVT). The
first proposal for that technical network offered a modest blueprint where
one central computing center in Moscow would regulate only twenty-five
to thirty computing centers in city sites of “information flow concentra-
tions” and an unspecified number of “regional calculating center and
points of information gathering”^3 (figures 4.1 and 4.2).