Date: Thu, 18 Jun 1998 23:06:17 +0100
From: Willard McCarty <Willard.McCarty@kcl.ac.uk>
Subject: sub rosa
In the 1960s I worked as a computer operator at the Lawrence Radiation
Laboratory in Berkeley, California. I began with an IBM 7090 (core memory,
tape drives, card reader, etc.) and IBM 1170 (primarily cards-to-tape), then
moved to the newly designed and constructed Control Data 6600, which I saw
being installed. The 6600 arrived with an operating system that -- mirabile
dictu -- could run 8 programs simultaneously, using a central processor and
13 independent peripheral processors. The CDC 6600 was fed by a CDC 6400,
which handled I/O, storing up work for the bigger machine to do and taking
its output and scheduling the printing. With such support the 6600 CPU could
keep itself busy for quite a while, unlike the 7090, which had to be given
one job at a time, and which would come to an utter halt until its tape
requests and the like were handled. I can still remember the moment when,
sitting at the console of the new 6600, on the "grave shift", I realised
that the machine did not really need me. Not in the same way the 7090 had.
That moment of realisation marked for me a turning point in what computers
represented -- semi-independent entities, automata.
More than 30 years later, a couple of days ago in fact, I had another such
For a few months now I have been participating in an experiment to evaluate
the utility of ISDN for connection between home, where I run Windows NT 4.0
Workstation, and the College network. The major impact of ISDN on my work is
not so much the improved speed but the fact that the dedicated ISDN line
allows connection to be made to or from my machine at any time without my
having to do anything. The line is simply raised when it is needed, then
dropped after a few seconds of idleness. One consequence of this is that
anyone anywhere can contact my home computer when it is on; if the call is
accepted by the machine, my line is charged the standard connection rate.
So, you can imagine my concern when I began to notice that independently of
what I was doing the lights on the router near the machine would every once
and a while begin to flash and I could hear hard-disk activity. Eventually
the network chap at the College began to watch the traffic and discovered
that there was a great deal of data coming from a Microsoft site. (Wait,
don't jump to conclusions yet!) What puzzled me was that this activity would
occur when NO applications were running. It took me many days but finally I
was able to catch the machine in flagrante delicto while I had the NT Task
Manager running and showing the active processes. I saw something called
LOADWC.EXE suddenly jump to life and grab a lion's share of the CPU cycles.
What, I wondered, could LOADWC ("LOAD Water Closet"???) be? To find out, I
searched the NT registry and found that it was called "BrowserWebCheck".
Being suspicious, I guessed that it had something to do with Internet
Explorer, which has deep roots into NT. I ran Explorer and discovered that
(I then remembered) weeks ago I had subscribed experimentally to a number of
news services in Explorer.
In other words, NT/Explorer recorded my subscription, then set up automatic
acceptance of data from the Microsoft service, which data LOADWC was having
stored on my hard disk, all without my needing to be aware of it or to run
Explorer. In still other words, the network has taken on a life of its own,
just as the CDC 6600 did all those years ago.
It's worth pondering what this sort of thing means. I'd be most interested
in your thoughts.
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
Dr. Willard McCarty, Senior Lecturer, King's College London
voice: +44 (0)171 873 2784 fax: +44 (0)171 873 5801
Humanist Discussion Group
Information at <http://www.kcl.ac.uk/humanities/cch/humanist/>