Professor Alan Kay
From Professor Alan Kay distinguished American Computer Scientist known for his work on Object
oriented coding and winner of the Turing Award in 2003.
I knew Maurice Wilkes slightly many years ago, Roger Needham years ago, and most
recently have met Andrew Herbert. The latter two had spent some time at Xerox Parc, and it
was Andrew who told me about your society after I mentioned LEO as a good follow-on
exhibit/story to TNMOC’s recreation of EDSAC. He also explained about the society’s
museum presence in Cambridge.
Still, I think having at least “a wall” at TNMOC about LEO would greatly add important
context to the larger story of computing, and especially British computing.
To me, there is more of an interesting parallel to Whirlwind development than to the IAS
machine (especially with regard to scope and software). I think the LEO software schemes
for the Lyons’ businesses were both landmarks and early, and deserve to be more told in some
kind of museum exhibit.
The ARPA and then PARC computer work was very inspired by both the big ideas of the
past and especially the amazing amount of work under difficult and primitive conditions to
not just think about software uses, but to include both the design and making of software and
hardware part of an integrated working practical whole system.
Still, it’s hard to beat the LEO story in this regard (I love it!)
I do know Dag Spicer, and others at CHM, and have donated various artifacts to CHM,
including a couple of Altos, and a copy of the original Dynabook cardboard model. In
addition, CHM is planning a “50th Anniversary of Smalltalk” for this Fall, which will
probably partially feature the “Smalltalk Zoo” (working versions/recreations of all the
Smalltalks going back to 1972).
A huge challenge (for a computer museum especially) is to provide actual explanations and
context beyond just exhibiting artifacts as static objects. Visitors are supposed to “keep
moving”, and there is little time or place to learn more (a partial exception used to be the Met
in New York, which had a great store next to a large cafeteria, so a visitor could buy more
context and spend an hour in the cafeteria learning more).
Still, for computers, what’s really important to understand about them is not actually shown in
a learnable way, especially with regard to software. And the shops are often very lacking in
this regard also. For example, a terrific little project at TNMOC is a valve flip-flop kit, which
requires one to put together and solder, etc. But there is no explanation of flip flops, their
history, and especially no explanation about how the kit itself actually works, or how flip
flops are an essential part of computers, etc.
I started to talk to Andrew Herbert because I found that I had one Alto left in storage, and that
the TNMOC had some interest in doing an Alto related exhibit. This got me to revisit
“context” in general, and to think about “more” in general. And this led back to thoughts
about earlier heroic computer projects, like Whirlwind, and especially LEO.
One of the nice and important properties of LEO is the wealth of documentation and detail —
although I gather that this doesn’t extend to the actual software that otherwise could be
resurrected via emulation. However, the wealth of early system planning and ambition is
priceless. Parc was almost the opposite in that, though the planning was pretty crisp, there
were not a lot of planning documents, but there are readable tapes of large parts of the SW,
and working Altos (also very detailed emulations including the microcode etc), with the
result that quite a bit of what actually got done can be demoed today. Very useful partial
exceptions are CSL’s minutes of the weekly “Dealer” meetings in which progress and
discussion were shared, and also the 6 month and 2 year reports (I wonder if these got
preserved somewhere, I didn’t take mine from Parc, and now wish I had). (LEOPEDIA
editor’s correction: All the manuals (except volume 2 CLEO) are digitised
at http://settle.ddns.net/LeoMan/Manuals.htm and there is a working emulation of the Master
Routine and Intercode Translator. Intercode programs may be written, translated and run either
freestanding or under the Master Routine. In addition we have listings of several customer
programs, mainly from London Boroughs, as well as a few CLEO programs.}
In any case, though I am mostly thinking about how more context could be shown about the
Alto, I can’t help thinking about what could be done with LEO’s history to make it more
understandable by the public (and I’m sure the LEO Society has thought about this much
more than I have). I’m certainly willing to donate funds to help make the LEO story more
widely known. https://en.wikipedia.org/wiki/Alan_Kay
