SAIL Sagas
Les Earnest, Executive
Officer of the Stanford Artificial Intelligence Lab, 1965-80, Associate Chair
of the Stanford Computer Science Department, 1985-88.
Below
are some stories that were included with emailed invitations to a conference on
“Sustainable archiving of SAIL” held at
Contents
Page Subject
2
2
Spacewar started videogames
2 Automatic
investments
3
Marriage refuted
3
David Shaw and Glenn Close are Close
3
Turing Awards
4 Hot times in the hills
5 Horse
rescue by Bill Schottstaedt
7 Grass fire narrowly misses computer music center, SLAC by Bob Beyers
and Karen
Bartholomew in Campus Report,
11
More gatherings?
11 Results of the 20th Annual Spring
Orgy, 1986 April 19
12 Off-road
13 Prancing Pony vending machine
Stanley won the race
The Stanford Racing Team, under the leadership of Sebastian Thrun, deservedly received a lot of media attention after
winning the robot car race across the desert in 2005 with their entry named
Spacewar started videogames
A recent article on "Videogames in Computer Space: the Complex History of
Pong" in the Annals of the History of computers can be seen at http://www.computer.org/cms/Computer.org/ComputingNow/computingthen/2009/03/CT-Lowood.pdf.
It points out that Spacewar, the first videogame, was
originally developed at MIT by Steve Russell and his colleagues in the Tech
Model Railroad Club. Steve then brought it to Stanford when he moved here to
join John McCarthy and he and others then improved it.
Meanwhile a company called Atari was formed independently to convert Spacewar into a commercial videogame but Bill Pitts, with
help from SAIL colleagues Ted Panofsky and Phil
Petit, beat them to it, putting it into the Stanford coffee shop and a local
bowling alley. Bill called it "Galaxy Game" because the term
"war" was a very unpopular on campus, which was deep in an anti-war
movement regarding
Automatic investments
Beginning in 1966 everyone in SAIL selected a two- or
three-letter identification for login. I see that 43 years later many,
including me, still use their original IDs even though they have moved to
different domains.
In the late 1970s a bright young SAIL computer scientist and entrepreneur
identified as RWW figured that money could be made by creating a computer
program that spotted trends in the commodities market and generated trades. He
put a program together, tested it on historical data and found that it made
money like a printing press. He then organized an investment club and sold
shares for $5,000 each. Whereas most investment organizations warn new clients
about the possible downside of investing, he built the warning into the name –
the Black Hole Fund.
Many people in SAIL each bought a share, including me, and some went for more.
Happily, a month after launch these shares had appreciated considerably and as
time went by they went up even more, reaching somewhere around $8,000 within a
few months. They then began to slip at an increasing rate and, when the fund
reached a preset cutoff at around half the initial value it was liquidated. We
later learned that the automatic trading program had indeed spotted a trend but
it turned out to be the attempt by the Hunt Brothers of Texas to corner the
silver market, which failed spectacularly, bankrupted them and cost us a few
bucks.
He later got some experience on Wall Street and, ignoring the Black Hole Fund
experience, went on to form a very successful hedge fund called D.E. Shaw &
Co. that pioneered the use of automatic trading, as discussed in an interesting
recent CACM interview at http://cacm.acm.org/magazines/2009/10/42365-a-conversation-with-david-e-shaw/fulltext
. He later founded D.E. Shaw Research, which focused on biochemical technology,
and there created Anton, a special-purpose supercomputer designed to speed up
molecular dynamics simulations by several orders of magnitude, a remarkable
accomplishment. David won’t be able to attend the reunion but sent an amusing
Regrets posting on
I lost track of the above interview URL for a time and, while doing a web
search for it, discovered that David married actress Glenn Close on
Marriage refuted
In my last blurb I noted that while doing a web search I
accidentally discovered that former SAIL denizen David Shaw had recently
married actress Glenn Close. However that is not true; a
classic case of name ambiguity. Who knew that there were two very
successful David E. Shaws who were both doing biotech
research? As it turned out, four former colleagues (Brian McCune, Paul Martin,
Bob Filman and Harlyn
Baker) straightened me out in a hurry.
David
Shaw and Glenn Close are Close
In response to my apology for mistakenly stating that
our David Shaw had married Glenn Close, he kindly responded:
"This happens all the time and is a source of endless
amusement to us, so please don't feel at all bad about it. Among other things,
Glenn Close's Wikipedia entry pointed to mine, and vice-versa, for quite a
while before we discovered it, so the news of our fictitious marriage was long
ago disseminated beyond any hope we might have had for epidemiological
containment."
He went on to say that he and the other David E. Shaw and their wives have
become friends and are now neighbors. "We can see each other's living
rooms from our respective windows." They apparently get together
occasionally to exchange misdirected mail.
Turing
Awards
Turing Awards are generally recognized as the Nobel Prizes
for computing, as discussed at http://en.wikipedia.org/wiki/Turing_Award.
Remarkably, of the 55 such awards made over the years, 16 have gone to people
affiliated with SAIL.
1969 Marvin Minsky for “artificial
intelligence.” He and John McCarthy had co-founded the Artificial
Intelligence Project at MIT in the late 1950s and Minsky
spent a sabbatical year at Stanford as Visiting Professor during 1964-65.
1971 John McCarthy for “artificial intelligence,” a term that
he introduced earlier. Prof. McCarthy was the founder of SAIL.
1974 Donald E. Knuth for his major contributions to the
analysis of algorithms and the design of programming languages. Prof.
Knuth was partially supported by SAIL and he often used the SAIL computer.
1978 Robert W. Floyd for “having a clear influence on
methodologies for the creation of efficient and reliable software.”
Prof. Floyd’s research was partially supported by SAIL and he often used the
SAIL computer.
1980
A. Antony R. Hoare for “his fundamental contributions to the
definition and design of programming languages.” Tony Hoare was a
visiting scholar at SAIL in 1969.
1984 Niklaus Wirth for “developing a sequence of
innovative computer languages: EULER, ALGOL-W, MODULA and PASCAL.”/ /Wirth was
a member of the Stanford Computer Science Department in the mid-1960s and, with
John McCarthy, contributed to the design of the first display-based timesharing
here, called Zeus.
1986 Robert Tarjan (with John Hopcroft) for “fundamental achievements in the design and
analysis of algorithms and data structures.” Tarjan did his PhD
at SAIL during 1977-80.
1991 Robin Milner for three major achievements in the mathematical theory of
computation, a field he worked in at SAIL during 1971-72.
1994 Edward Feigenbaum and Raj Reddy for “pioneering the design and construction of
large scale artificial intelligence systems, demonstrating the practical
importance and potential commercial impact of artificial intelligence
technology.” Feigenbaum joined SAIL when he
came to Stanford in 1965 and Reddy earned his PhD there in 1966.
1996 Amir Pnueli
for “seminal work introducing temporal logic into computing science and for
outstanding contributions to program and systems verification.” He
worked on mathematical theory of computation at SAIL during 1967-68.
2000 Andrew Chi-Chih Yao
“in recognition of his fundamental contributions to the theory of computation,
including the complexity-based theory of pseudorandom number generation,
cryptography, and communication complexity.” Prof. Yao
did some of his early work on those topics at SAIL during 1976-80.
2002 Ronald L. Rivest, (with Adi Shamir and Leonard M. Adleman) for “their ingenious contribution for making
public-key cryptography useful in practice.” Rivest
worked on his dissertation at SAIL beginning in 1969. The initiator of the
concept of public-key encryption was Whitfield Diffie,
also of SAIL.
2003 Alan Kay for “pioneering many of the ideas at the root
of contemporary object-oriented programming languages, leading the team that
developed Smalltalk, and for fundamental contributions to personal computing.”
Kay spent two years as a post-doc at SAIL before moving on to Xerox PARC and
elsewhere.
2004 Vinton G. Cerf (with Robert E.
Kahn) for “pioneering work on internetworking, including the design and
implementation of the Internet's basic communications protocols,
2008 Barbara Hubermann Liskov for “contributions to practical and theoretical
foundations of programming language and system design, especially related to
data abstraction, fault tolerance, and distributed computing.” She did
her PhD dissertation on chess end games at SAIL during the mid-1960s.
Hot times in the hills
Former SAIL people may recall that rows of Blue Gum Eucalyptus trees used to
line both sides of Arastradero Road, making it a beautiful country lane beneath
the overhanging foliage. People who attend the Walkabout will notice that the
trees have now largely disappeared. That was a result of the runaway fire that
was started just downhill from the DC Power Lab on the afternoon of
It was a very dry summer and the eucalyptus trees by the road turned into
torches. With help from a strong west wind the crown fire swept down
Arastradero, traveling a half mile to
One person at the lab saw that a barn was starting to catch fire at the “Poor
Farm” across the road, in an area that would later become part of
As the trees burned and homes along
After the fire,
During this fire, homeowners who evacuated as requested by fire officials
mostly lost their homes whereas those who stayed and fought it with garden
hoses saved theirs. However general conclusions probably cannot be drawn, since
risk assessments must be done on a case by case basis but quickly.
Horse
Rescue by Bill Schottstaedt
The part of the preceding story about the horse rescue during the big fire of
1985 brought forth an eyewitness account from Bill Schottstaedt,
below, that differs in significant details. This once again illustrates that
stories often drift in the retelling. Here
are Bill Schottstaedt's observations.
--------------------------------------
The fire continued to spread very rapidly. Most of the ccrmalites left as fast as possible, but about a dozen of
us stayed. For some reason none of us broke out the fire hoses. It
seemed improper for a non-fireman to mess with them, and we expected the fire
fighters to show up right away. But they didn't actually get to the lab for
more than an hour. As the fire increased in intensity, the giant resin-soaked
eucalyptus trees began to burn. As each tree reached ignition, first a
cloud of rock doves burst out of the foliage, then seconds later the treetop
exploded in flames.
We could look down the hill and see that the horse farm was burning. I think I
suggested checking on the horses, but Rob Poor thinks it was his idea. In
any case, we jumped in his car and drove down the hill. On the way down we
passed a half-size fire engine that had been abandoned to the fire. It
was weirdly misshapen, melted, I guess, from the heat. So, the fire fighters
had shown up, but had not made it to the lab.
At the horse farm, the fire was all around us, eucalyptus trees were going off
like bombs, the farm house itself was already in flames, and the propane tank
in the back yard was a roaring flame-thrower. I remember wondering if it
would explode while we were goofing around in the barn.
The barn door was open, so we had no trouble getting in, and could see through
the smoke that it was full of horses, locked in their stalls. I remember they
were nervously banging against the walls and coughing. The barn itself was not
on fire, but smoke was rising from the roof,
and coming in from the surrounding flames. I could not figure out how the
stall latches worked, but luckily Rob did, and he went through the barn
unlatching the doors. My job was to get the horses to leave the barn, but
they didn't need any help from me. My main contribution to the rescue was
to get out of the way. The barn eventually burned to the ground.
As the horses left, the ceiling started burning. I remember thinking that
Rob and I were in some danger because branches were falling and we were
surrounded by flames. As we ran back to Rob's car, I looked back at the
farm house. One of the horses had gone into the house,
and had started climbing up the stairs to the second floor! This bizarre
sight is probably my clearest memory of the entire fire. The horse must
have gotten away, because Rob says all the horses survived.
We drove back to the lab, and people finally started breaking out the fire
hoses. A number of them had rotted to the point that they were useless,
but 3 or 4 were intact enough to direct water onto the flames. We spent an hour
or so dousing the burning grass and trees on the Arastradero side of the road
that surrounded the lab. A fireman finally showed up and got us organized
as a proper crew, clearing deadwood away from the road. When I asked why
it had taken so long for anyone to get to us, he said we were at the juncture
of several (I think he said 5) fire districts, that there was a lot of
confusion as to whose fire it was: "too many chiefs, not enough indians" was his summary. The
By nightfall, the fire had burned itself out. In the dark, the lab was
surrounded by black fields of glowing logs. Each oak tree was encircled
by embers, and every now and then a branch would crash to the ground in a
shower of sparks, starting a small short-lived fire.
Bob Shannon and I stayed at the lab day and night for the next several
days. A Concerned Citizen apparently saw us and notified the
police. They drove up and I met them. I think I recognized the
policeman from a previous adventure with a run-away sheep (I could write a
story about how not to catch a sheep, but not today). By chance, I had
Stanford declared us Heroes, and offered us a banquet, but I was on night
schedule and never got my free meal.
[Bill Schottstaedt later wrote: I found the Campus Report for July 3, and as you wrote, they reported that the police considered it arson. They also reported someone driving a car through a wall of flames at the barn -- I don't remember this, but I'm worried that my memory is at fault. It's possible the "Rob" I was with was Rob Currie, not Rob Poor. Sigh -- it's no fun getting old.]
Below are photos and the text of the Campus Report article
that Bill sent along.
Burning Barn
Grass fire narrowly misses computer
music center, SLAC
By Bob Beyers and Karen
Bartholomew in Campus Report,
To Patte Wood, Monday’s grass fire near the D.C. Power Lab was a bad dream come true.
To Robert Currie, who sounded the first alarm, it proved an instance of remarkable derring-do: to help save trapped horses, he drove a car parked inside a barn through a blazing wall. The car was a Pinto, but it didn’t explode. [Pintos of that era were noted for having their gas tanks leak and catch fire after rear-end collisions.]
To Celia Fulton, one of 15 to 20 persons at the lab, home of
the
Old fire hoses, leaking water “just like you see in cartoons”—as Fulton put it—were used by Chris Chase, Julius Smith, Mike McNabb, Bill Schottstaedt, Robert Poor, Dave Jaffe, and others to preserve the wooden structure built in the 1960s. Often compared to an abandoned starship, it is the home of five people among those working there.
Horses let loose from the nearby burning
barn circled the D.C Power Lab.
Burning tree
One of the computer music “fire crew”
members sprays nearby pine trees. A circular driveway around the lab and
favorable winds helped protect it from the flames.
Winds blew the fast-spreading inferno away from the lab,
damaging or destroying a dozen homes in Los Altos Hills and scorching hundreds
of acres of rolling, tinder-dry grasslands and forest.
A separate blaze, started by sparks from a disabled car on Interstate 280, spread rapidly northeast from the Alpine Road underpass through Webb Ranch to the edge of the IR6 building at the Stanford Linear Accelerator Center. It caused no apparent damage to the structure.
Helicopters from the California Division of Forestry (CDF)
drew huge buckets of water from nearby Felt Lake to battle both fires, which
were brought under control about 7 p.m. CDF planes dropped fire retardant in
the area of the main fire. Approximately 200 firemen from
Eight Stanford police officers assisted
Some of the computer music students and staff who
helped protect D.C. Power Lab from fire pose for a photo after their ordeal:
(front row) /rob Poor with field glasses and Chris Chafe; (back row) Julius
Smith, Mike McNabb and Bill Schottstaedt
When Curie first saw the fire, it covered an area about 30
feet square near the lab. After calling the Fire Department, he and a volunteer
contingent went across
Around
Wood had seen it all before—in a dream two weeks ago.
Following the dream, she discussed her concerns with her colleagues, but
nothing had been done to have backup copies of their computer programs stored
at another site. Had the winds blown from the opposite direction, a lot of
their life’s work would have been lost. As soon as electric power is restored,
perhaps starting late Tuesday, they’ll put finishing touches on compositions
for a concert at
When Prof. John Chowning, who
heads CCRMA, called from
From one side of the lab, the view of the rolling foothills remained unchanged Tuesday. From the other, there was total devastation. “It’s surreal,” said Wood.
Sometime this fall, CCRMA will move to new quarters on the
main campus, as planned, its karma intact after a very close call.
More Gatherings?
[After the event] Most people seemed to have a good time at
the reunion this weekend. There now is talk of holding an event like this more
often than every 40 years or so. Given that the old folks have been given their
due, to a first approximation, any awards at future SAIL events would
presumably focus on recent accomplishments.
Of course we could include a nostalgic event such as another Spring Orgy,
including a Felt Lake Steeplechase (provided that there are
medical staff handy), a treasure hunt, and a programming contest run by
Don Knuth. Instead of holding both fast and slow bicycle races we probably
should focus on slow. And forget about the car time trial. The results of the
most recent such event, the 20th Spring Orgy held in 1986, are attached. Given
that there was a 14 legged race, we presumably would continue that incremental
event by having a 15 legged race next time.
Results of the 20th Annual Spring Orgy,
1986 April 19
SLOW BICYCLE
SLOW BICYCLE
SLOW BICYCLE
SLOW BICYCLE
SLOW BICYCLE
SLOW BICYCLE
600 METER DASH,FIRST PLACE,MICHAEL LOWRY,1:52.96 (track record is 1:44.0 by
Mike Hewett)
600 METER DASH,SECOND PLACE,BOB TUCKER,2:09.72
600 METER DASH,THIRD PLACE,JOHN MYERS,2:35.61
600 METER DASH,FOURTH PLACE,CHRISTOPHER LANE,2:58.06
600 METER DASH,FIFTH PLACE,VIC SCHEINMAN,3:05.39
600 METER DASH,SIXTH PLACE,JANICE
600 METER DASH,SEVENTH PLACE,MARTY FROST,7:15
600 METER DASH,SEVENTH PLACE,MARGARET LONGSTRETH,7:15
6 KILOMETER CRITERIUM,FIRST PLACE,MICHAEL LOWRY,11:39.52 (track record is 10:03
by Armin Staprins)
6 KILOMETER CRITERIUM,SECOND PLACE,LES EARNEST,12:24.21
6 KILOMETER CRITERIUM,THIRD PLACE,VIC SCHEINMAN
6 KILOMETER CRITERIUM,FOURTH PLACE,GUY MALACHI
FOURTEEN-LEGGED
FOURTEEN-LEGGED
FOURTEEN-LEGGED
FOURTEEN-LEGGED
FOURTEEN-LEGGED
FOURTEEN-LEGGED
FOURTEEN-LEGGED
FOURTEEN-LEGGED
FOURTEEN-LEGGED
FOURTEEN-LEGGED
FOURTEEN-LEGGED
FOURTEEN-LEGGED
FOURTEEN-LEGGED
FOURTEEN-LEGGED
FOURTEEN-LEGGED
FOURTEEN-LEGGED
FOURTEEN-LEGGED
FOURTEEN-LEGGED
FOURTEEN-LEGGED
FOURTEEN-LEGGED
FOURTEEN-LEGGED
FOURTEEN-LEGGED
FOURTEEN-LEGGED
FOURTEEN-LEGGED
FOURTEEN-LEGGED
FOURTEEN-LEGGED
SHERIFFS STEEPLECHASE,FIRST PLACE
SHERIFFS STEEPLECHASE,FIRST PLACE
SHERIFFS STEEPLECHASE,FIRST PLACE
SHERIFFS STEEPLECHASE,FIRST PLACE
SHERIFFS STEEPLECHASE,SECOND PLACE,
SHERIFFS STEEPLECHASE,THIRD PLACE,
SHERIFFS STEEPLECHASE,FOURTH PLACE,
SHERIFFS STEEPLECHASE,FIFTH PLACE,MARTIN FROST,16:07 by kayak
SHERIFFS STEEPLECHASE,SIXTH PLACE,JIM DAVIDSON,16:20
SHERIFFS STEEPLECHASE,SEVENTH PLACE,LES EARNEST,19:54
SHERIFFS STEEPLECHASE,EIGHTH PLACE,ALLEN VAN GELDER,21:31
SHERIFFS STEEPLECHASE,NINTH PLACE,LANCE BERC,24:17
SHERIFFS STEEPLECHASE,TENTH PLACE,JOHN MYERS,27:53
TREASURE HUNT,PLANNER,DICK GABRIEL
TREASURE HUNT,PLANNER,MARTY FROST
TREASURE HUNT,FIRST PLACE
TREASURE HUNT,FIRST PLACE
TREASURE HUNT,SECOND PLACE
TREASURE HUNT,SECOND PLACE
TREASURE HUNT,SECOND PLACE
TREASURE HUNT,THIRD PLACE
TREASURE HUNT,THIRD PLACE
TREASURE HUNT,THIRD PLACE
TREASURE HUNT,FOURTH PLACE
TREASURE HUNT,FOURTH PLACE
TREASURE HUNT,FIFTH PLACE
TREASURE HUNT,FIFTH PLACE
TREASURE HUNT,SIXTH PLACE
TREASURE HUNT,SIXTH PLACE
TREASURE HUNT,SIXTH PLACE
SPRING ORGY,MOST PERSISTENT COMPETITOR,VIC SCHEINMAN
SPRING ORGY,CHIEF REFEREE,LES EARNEST
Off-road
Given that the DC Power Lab was in a scenic location near
Whenever I saw someone tearing up the grass I usually
confronted them, sometimes with amusing results. For example, as I bicycled up
to the lab one time I saw two guys in an SUV just heading off-road and pointed
out the sign to them. Their response was to gun the engine and head off anyway,
so I gave chase on my bike. As they headed down-slope I stayed after them and,
in their eagerness to get away they failed to notice that the track they were
on ended in a swamp next to
I then sent letters to local towing companies telling them that they were not allowed to tow vehicles from our property without prior authorization, so that we could bring in a sheriff each time. The next time it happened I was duly notified, got the driver cited and authorized the towing. However when the tow truck went down the hill to pull the car out, it too got stuck. The towing company then sent another truck, which stopped further up the hill and attempted to pull the others out, but it too got stuck. The towing company then sent a monstrous tow truck that stayed on the pavement and pulled the other two tow trucks out, leaving the car stranded for a week or so.
On another occasion I confronted a motorcyclist who was apparently heading off-road and he agreed to stay on the pavement, but promptly took off on a trail that I knew led to the south side of the lab. I then bicycled over there and hid behind a bush as he came uphill toward the road. I managed to grab him around the neck and pull him off the bike before he knew what was happening. He promised to come back with his gang and take care of me the next day so I told him to bring lots of help. However I never saw him again.
Prancing Pony vending machine
The Prancing Pony Vending Machine was evidently the first computer controlled vending machine anywhere in the world. It was created to fill an unmet need.
Given that SAIL was about five miles off-campus and the
nearest food source was a beer garden (Zotts) about a
mile away, I initially set up a coffee and food room near the center of our
facility and it subsequently got named after a pub in Tolkien's
"Lord of the Rings". In fact all rooms in our facility were named
after places in Middle Earth and had signs posted on the doors showing their
names in both Latin and Elvish alphabets. At some
point the Stanford Buildings folks asked me to number our rooms and give them a
map. Instead we gave them a map showing room names in both alphabets. Their
response was to send out a carpenter with numbered tags, which he nailed on
each door.
Meanwhile we took turns buying coffee and food, which was offered for sale on
an honor system basis. That worked well for awhile but it suddenly started
losing money big time. We then negotiated with Canteen, which had an exclusive
contract with Stanford, to put in a couple of vending machines. However we
found that they were not restocked often enough and broke rather frequently.
I finally negotiated to rent a machine from Canteen that we could restock. They
seemed to like this idea since it would relieve them of making frequent trips
out to our distant facility. In fact they never billed us for the rental even
though I repeatedly called it to their attention. Meanwhile I got Ted Panofsky to make a connection to our computer so that it
could release the doors on the vending machine, thus making it possible to buy
either for cash or, though a computer terminal, on credit. I then wrote a
program that let people buy under password control and that billed them on a
monthly basis via email. It was set up to randomly give away whatever was
purchased on 1/128th of the purchases and offered a "double or
nothing" option, which had an honest 50:50 outcome. I noticed a cultural
difference in that almost none of the computer science students gambled,
knowing that they would win 1/128th of the time if they didn't, whereas many of
the music students did gamble. In both cases the Prancing Pony vending machine,
having taken on the name of the room, seemed quite popular and we organized a
team of volunteers to acquire the needed supplies and restock the machine at
least twice a day.
The Prancing Pony also sold beer but only on credit and only
to people over 21, since it knew everyone’s age. If a youngster attempted to
buy beer it responded “Sorry, kid.”
Some years later I found out why the honor system had failed in the Pony. I was
the founding President of Imagen Corp., which made the first desktop publishing
systems using laser printers, and after awhile there a young woman employee
felt obligated to confess that when she was a teenybopper, she and her
girlfriends used to ride their horses up to SAIL, then went in and stole candy
from the Prancing Pony. Thus, her misconduct contributed to a technological
advancement.