Inkwell: Authors and Artists

    
Brian Dear steps into the spotlight at Inkwell to talk about his
adventures with people who built a community using one of the
prototypes for all social computing, PLATO, plus his fascinating
collaborations that have followed that experience!  Leading the
conversation is Ari Davidow, one of the community-building pioneers at
The WELL as well as on sites he has created.  

Brian and Ari, welcome!  Please say a little more about yourselves as
we get going, since my introduction lacks some of the juicy details. 
  

    <scribbled by ari Sun 8 Nov 09 14:43>
  

    
My name is Ari Davidow. I am currently the Director of Online 
Strategy at the Jewish Women's Archive (http://jwa.org). I became 
involved with the WELL a bit over 20 years ago, having gotten my 
first taste of the potential of online community with 
microcomputer-based BBS's. I tend to think that this makes me a 
long-time veteran of the online world. Hah.
 
My guest will be Brian Dear, whose experience with online 
community dwarfs my own alleged longevity. That matters because 
this interview is a bit different from the usual inkwell.vue  
feature. Instead of discussing a book just released, we will be 
discussing a book in the process of being written.
  

    
Brian Dear is founder and chairman of Eventful, Inc., which he 
founded in 2004.  Eventful is the world's largest search engine 
for events, and also offers the Eventful Demand service where fans 
can "demand" that an event happen in their town.  Prior to 
Eventful, he founded eBay Design Labs, the team responsible for 
user experience design for eBay's website. Other previous gigs 
include various management roles at Eazel, MP3.com, FlatWorks, 
RealNetworks, and Coconut Computing.

After dabbling in BASIC on a Wang 2200A personal computer in high 
school, Brian got his real start on computers in 1979 at the 
Univerity of Delaware where he was first exposed to the PLATO 
system.  He got a programming job on PLATO and wound up working on 
PLATO for 5 years, after which he joined Hazeltine Corp in 1984 to 
work on designing multimedia authoring systems.

Since 1985 as a hobby he began collecting the oral history of the 
original creators of the PLATO system.  Over the past 25 years 
that has grown into a major research project to document PLATO's 
history.  In 2010, which is the 50th Anniversary of PLATO, he 
expects to have the book finally published (http://www.platopeople.com/).  
He is also organizing a conference to celebrate all things PLATO at 
the Computer History Museum on June 2-3,2010.
  

    
Brian, all I know of the PLATO system is what I have read on your 
website, or on David Wooley's website. Why don't we start with 
some history? What was the PLATO system? Why did it matter? Tell 
us a bit of the story to get us started.
  

    

PLATO was the answer to a question posed by a physicist at
the University of Illinois in late 1959: is there something 
we, meaning the U of I, can do with computers to help education?
Education, especially math and science education, was a big
issue after Sputnik in late 1957, where the mood of the
country was, we just got beat by the Russians.  There was a
major mandate to improve education so we'd have the math
and science skills in the new generation to enter the workforce
and make sure the country was the global leader in science 
and technology, including space.

So PLATO, standing for "Programmed Logic for Automatic Teaching
Operations", was started in 1960.  The vision was, build a 
general-purpose, extensible system that could be programmed
to teach anything to anyone.  The system went through three
major architectural stages during the 60s, but the real breakthrough
was version IV, rolled out in 1972.

PLATO IV had gas-plasma, flat-panel display terminals with built-in
touch screens.  The display was 512x512, orange pixels.  The
terminals were connected to a heavily customized Control Data
CYBER mainframe.  The programming language was called TUTOR.

What made PLATO IV remarkable was how quickly it was embraced by a
large number of people from a very wide range of backgrounds, and
then what they did with it.  In addition to lots of educational
lessons, indeed, entire curricula, being written on the system, 
it turned out that TUTOR was ideal for other things: it had
features like "common" memory and there were commands that could
tell you how many people were running your program and who they
were.  These turned out to be ideal ingredients for multiplayer
games, multiuser chat rooms, and instant messaging.  Plus, you 
had by the mid 1970s a huge collection of message forums called
notesfiles -- very akin to The WELL's conferences.  In fact The
WELL's conference architecture (conference > topic > linear set
of responses), was the same as PLATO's (notefile > note > linear
set of responses). I've always felt that that architecture was
what made PLATO's notesfiles and The WELL's conferences so 
interesting and readable (as compared to the threaded model where
every response can have its own set of responses.) 

PLATO was the first computer system, network really, that scaled
with lots of people.  It was bigger than ARPANET at least for a
while, with many capabilities at a level of usage that we wouldn't
see until the 1990s.  A PLATO user didn't use a computer, they
"belonged" to the system.  It was a community.  For me, I was drawn
immediately to this sense that a computer wasn't for number-crunching
or lonely things like word processing, spreadsheets, or video games,
the way Apples, Commodores, etc were being used, but it was a "place"
where you could meet, interact, stay in touch, get answers to questions,
and share and make discoveries.

Why did it matter?  It's kind of like Ardi, the recently-discovered
4.5 million year ancestor of humans.  PLATO matters because it 
challenges our assumptions of how the online world evolved.  It 
rewrites the history.  It's as if we discovered Wilbur and Orville
Wright were not the first to fly a powered plane -- that it'd been
done faster and longer with a jet aircraft 30 years earlier.  
  

    
Wow.  When you say bigger than ARPANET (the precursor to the Internet)
approximately how man people are you talking about?
 
  

    
I've looked far and wide for an authoritative source showing 
*user* counts on ARPANET machines during the 70s, but all I ever find
in the histories is the number of "host computers" connected.  No 
indication of how many different *people* had accounts on those host 
computers. I'd love to hear of a reference if someone has one.

PLATO had only about 15-20 mainframes installed between 1972-82, with
more being installed in the subsequent five years, at sites all over 
the world.  Most of these systems served many thousands of users, 
mostly students and trainees.  Terminals were installed in sites ranging 
from K-12 schools and universities to government, military, and corporate 
training facilities.  I'd estimate that a high six-figure number of people 
encountered PLATO between 1972 and 1982.  I don't believe ARPANET hosts 
achieved that during that timeframe, but I'd be delighted to be corrected.
  

    
This is indeed like a buried history.  Were there contemporary
articles or books about PLATO that hinted at the educational and
community uses during the early days?  If it's like a lot of things
from the 70s and 80s, I'm wondering if there may be published accounts
that have never made it to the online world!
  

    
There are no trade books about PLATO; no historical or narrative 
nonfiction accounts, no documentaries, no films, basically nothing
except some rare local news articles and the very occasional magazine
article from the 60s or 70s.  There's very little footage (if there had 
been, I'd have done a documentary film in a second), although I have a 
tape of Donald Bitzer demoing PLATO on the Phil Donahue show circa 1977-78.

Now, there are TONS of academic articles, from the 60s through the 80s, 
dealing with the educational dimension of PLATO, as one would expect 
(that was PLATO's funded mission: education).  Though, interestingly, if 
you dig deep enough you can find some more social-sciences and 
psychological research on PLATO that is very much akin to the "Internet 
studies" sort of academic research that goes on today.  

While there are tons of academic papers, there would have been many more 
but it was widely known, and numerous educational researchers have told 
me this, that over the years, PLATO papers were more and more rejected 
from publication or from conference presentations simply because 
non-PLATO researchers were sick and tired of PLATO, and tired of what 
they felt was the "arrogance" of "those PLATO people".  The PLATO project 
was the best-funded, biggest computer-assisted instruction project in 
the world.  And that annoyed lots of people who weren't a part of it, 
because once PLATO was funded, funding essentially dried up, and it's 
never been repeated at that scale ever since.  

There are also a few academic books about PLATO here and there, aimed at
instructors, computer-based training professionals, instructional 
systems designers, etc.  But nothing like a "Hackers: Heroes of the 
Computer Revolution", "The Soul of the New Machine", or even something 
like a Sherry Turkle-style cyberculture book. 
  

    

I have a theory why there was little to no mainstream media
coverage of the "social" side of PLATO: the fact that here was
a large group of people "living digital" day in and day out,
with email, instant messaging, chat rooms, message forums, and
games as insanely addictive as Evercrack, er, Everquest, or
World of Warcraft, or Microsoft Flight Simulator.

My theory is that the user community knew full well how good they
had it, but they also knew how hard it was to successfully explain
your excitement to a non-technical person.  Today, if during a 
dinner conversation you mention you emailed someone that day, it's
no big deal.  In the 70s, if you mentioned you emailed someone,
it took the entire dinner to explain the concept of email, and
by the end, both you and the listeners were exhausted.

PLATO didn't luck out with a Stewart Brand / Rolling Stone type
article.  Had Stewart lived in Champaign-Urbana, the world would
prolly be a very different place, but he didn't, and he wrote about
what he saw at nearby Stanford.  Likewise, there weren't a lot of
curious media people in the midwest.

Worse, the public front of PLATO, both from the University of Illinois
as well as from Control Data, was that this was an educational 
system.  Education money paid for it, education money kept the 
power going.

Because education has never been "sexy" as a reporting subject,
and because PLATO was always marketed in the education space, the
cybercultural dimension of PLATO simply stayed under the radar, and
therefore went undocumented.
  

    

Ooh, I should add something to #9 above.  There is one interesting
account of PLATO from the 70s that got published in a book.  Well,
a few special pages of a rather unusual book: Ted Nelson's "Computer Lib /
Dream Machines" book.  Unfortunately, it's long out of print, and the
edition published by Microsoft Press in the 80s deleted some pages
(including stuff on, what else, PLATO).

Ted Nelson's PLATO adventure is the closest thing to a Stewart
Brand / Rolling Stone type of article that was ever published in that
era about PLATO.  I don't know if the original book is available
online these days, but it's worth looking for.
  

    

Update to #6, about how many people were on PLATO vs ARPANET.

Vint Cerf told me today that he thinks there were some 50,000
people on the ARPANET, but he doesn't say which year or over
what time period.

Noel Chiappa at MIT estimates there were 6,600 people identified as
of 1982, and about 14,500 as of 1984.  That sounds about right.

If those ARPANET numbers are accurate, then PLATO dwarfed ARPANET
perhaps as late as 1985.
  

    
That is simply amazing.  

(That accidental secret society aspect is mighty familiar, too.  Even
years later, when I got involved as a customer on The WELL in 1990, it
was awfully hard to tell outsiders why it was so captivating.  I told
my grandmother, who was an outdoorsy old time Californian who didn't
use computers, that it was like being around a campfire with many
interesting people, some of whom you knew and others who dropped in,
and everybody took turns talking, but she was rather dubious about that
metaphor!  Describing what a modem was turned out to be pretty
useless, too.) 

I can imagine futility of describing PLATO as a juicy, evolving
computer-mediated culture ten years earlier!

I'm curious about the cultural evolution, but first, were there any
general learnings from all the experience with using PLATO for
education that have made their way into contemporary "distance
learning" and all our modern adaptations?
  

    

Oh, a whole book could be written about that.

I have my theories and they're probably biased and presumptuous
and inaccurate and unjustified.  But here goes.

Education is a social science.  In my experience, the learnings
and discoveries in the social sciences do not accumulate and
propagate the way they do in the "hard" sciences.  In the hard 
sciences, somebody makes a discovery.  The paper is presented in
a peer-reviewed journal, people replicate the findings, those
get presented in peer-reviewed journals, and pretty soon the
whole world is thinking a new way about something.

WHat I found in education is that you can publish all you want,
but getting the world to pay atention was nigh impossible.

You figured out a terrifically effective way to teach?  Awesome.  
Even have the statistical data to show how proven your approach
is?  Fantastic.  Can you get a critical mass of people to follow
your approach such that it becomes a mainstream method of teaching?
Good luck.

And so it was with computers in education.  It's a tragedy that so
much brilliant work done on PLATO, not only in elementary education
but even in hardcore college-level courses, stuff that tens of
thousands of students earned full credit for as a formal part of their
educational curriculum, is no longer used simply because the underlying
technology platform has changed.*

Oftentimes instead there's web-based material that isn't even as
interactive or as well-thought-out as the PLATO versions of the 
same subject matter done in the 1970s.  Heck, I even saw this
first-hand on a tour of some learning labs at the U of I in 
1997 and 2003.  Labs that in the 1970s were full of students working
with their PLATO terminals  that are now full of students working
with browser-based PCs.  On PLATO, you typed your answers to
questions and got INSTANT results.  On the web, you type your answer,
click "Submit", and get a frickin' beach-ball busy cursor as some
stupid web server takes its sweet old time to come back with a "correct"
or "incorrect" message.  We've lost a lot of real insight into how
to make computers engaging with students as we have "progressed"
technologically over the past 20 years.




* There is an exception.  Some PLATO courseware written in the 1970s
and early 80s is still in use on the NovaNET system, which serves
tens of thousands of students all over the country every day, right
now.  NovaNET still runs the TUTOR language, and even has archives of
many PLATO notesfiles going back to the 80s, and in some cases, into
the early 70s.
  

    
Lag time is so dispiriting.  All of the potential sense of being in a
place reverts back to a feeling of hurling an object over a large chasm
and waiting to hear if it clunks at the bottom or reaches the other
side. 

Do you have any idea who has the rights to the lost courseware?  I
wonder if that is a lost treasure trove somebody might renew and adapt
to other platforms. 
  

    
The U of I has lots of rights, as do the original authors.
Some lessons here and there have been adapted to PCs and Macs,
soemtimes to great success.  The best example I can think of is
"How the West Was One + Three x Four...", which was a very popular
elementary math lesson/game created in the mid-70s that thousands
of kids around the country enjoyed for many years.  Then, it was
ported to PCs & Macs by the author, and it went on to be a big 
seller in the educational software market.  I believe it's still for
sale... yep, here it is:

   <http://store.sunburst.com/ProductInfo.aspx?itemid=176613>


Some college-level chemistry and physics lessons were adapted to
newer platforms, but the vast majority of PLATO's 70s-era courseware
has not ever migrated to newer platforms.  My guess is, a) the people
who might even know it ever existed are now retired or close to it,
and b) if someone is aware it exists, the likely perception is it's 
tainted by the very fact that it is so old (as in, how could
anything so old be any good).  Ironically, a lot of the lessons
are probably better than any competitive newer lessons even though
the new stuff has whiz-bang multimedia glommed on to make it look
shiny.

Finally, a big chunk of the original U of I lesson catalogue
is available for viewing on the cyber1.org "PLATO" system.  The
Cyber1 people got a special legal reprieve from the U of I to 
make the lessons available for historical purposes (can't be used
by actual students or for commercial use).  For instance, the
"How the West Was One" lesson is available as lesson "west".
  

    
Hmm, what an interesting licensing choice.  Better than nothing.

Most of the folks here at The WELL are probably most intrigued hearing
about the forum-like notesfiles, and how people built a community
there.  In your research for the book, have you discovered any7
consensus on a tipping point into an unmistakable community, that
interaction or personality or event that made it clear something
remarkable was happening?

(Welcome to all who are now reading along without logging in.  You're
invited to join us if you like, or you may simply email a question or
comment for Brian Dear, for posting here.  Send it to inkwell@well.com
-- please include "PLATO" in the subject line.)
  

    

As early as 1972, the early PLATO IV developers and lesson authors
realized they needed a way to communicate with one another.  So
they did the easiest thing they could think of: they opened a
source code file and instead of putting source code in it, they
left it open and anyone could write a question in there and then
save the file and let someone else come along and edit it and
append an answer.

As you can imagine, the honor system worked only so far.  It became
a routine annoyance that some joker would come along, and blow away
the contents of the file, or change the answers or the name of who
posted the reply.  So reliability and security were key drivers that
finally pushed Paul Tenczar, one of heads of the system programming
team, to assign Dave Woolley to write a real program that had
security and reliability and would end the use of these primitive
text files.  That was summer 1973.  Wooley was 17.

He had little to no experience, and the entire concept of "notesfiles"
or "conferencing systems" was not even invented yet.  He just went
and designed something that seemed reasonable, coded it, and in
August 1973, released it to the community.  At first there were
only three notesfiles -- one for system announcements, one for
public notes, and one for "help" notes.  They were all a huge 
success.
  

    

"Ask the Experts" on The WELL has always reminded me of helpnotes,
although helpnotes was for help on pretty much anything, even
mundane things like "anyone know a good mechanic for a '72 VW?"

I remember vividly when I first came across helpnotes.  It was this
sudden awareness that here was a computer where there were tons of
PEOPLE connected, all offering to help one another, for free.
In fact, it was abundantly clear that the more you helped people,
the more useful your answers were, the more likely you'd get help
whenever you wound up needing it.  Early online karma, for sure.
Something that I've also seen on The WELL pretty much since the
beginning.
  

    
Exactly.  So who was there?  Kids, professors, grad students?
  

    

All kinds of folks.  The CERL (Computer-based Education Research
Lab) staff was a hodgepodge of engineers, physicists, educational
psychologists, programmers, and hardware designers.  Add to that
the fact that PLATO quickly attracted people around campus and
around town who were just drawn to this one-of-a-kind, futuristic
system.  Just nearby nearly across the street was Uni High, the
elite high school made up primarily by kids from U of I faculty.
They tended to be exceptionally bright and a lot of 'em began to
hang out at CERL, learning how to program TUTOR and starting to
participate in the online culture that took off in 1973 as chat
rooms (Talk-o-Matic), instant messaging (TERM-talk), and conferencing
(PLATO NOtes) all opened up that year.

Plus, you had competing message systems hacked together by users
themselves -- "discuss" and "pad" were two.  "Discuss" died off,
but "pad" eventually became a notesfile, and it still exists today
on NovaNET, with a parallel "pad" on Cyber1.  They're kind of
like the "genx" conference here.  Random sayings, observations,
conversations.
  

    
I lived half a block from Antioch Law School in D.C. in the late
1970s, and I wandered in one day and saw this orange plasma screen with
a sign saying PLATO.  Somehow I managed to read some of the help files
and ended up playing a massively multiplayer space game.  That was my
only exposure to PLATO, but it clearly showed where computers were
headed.
  

    

One amusing trait of PLATO was that, given its educational
focus, the games typically had throrough "help lessons".
It became a tradition, in a way, that you'd write a game,
but it wasn't complete until the help lesson was written.
Some help lessons were enormous, encyclopedic in scope, like
the help lesson for Avatar, one of the precursors to later
MMORPGs.  "emphelp", the Empire help lesson, had its own
simulation of the actual Empire game, showing you step by
step how the game was played and what all the commands were.

I suspect many gamers were like me -- skip the help lesson
and just go in and play, figuring stuff out (and dying a lot)
until you got the hang of it.  Only later one might actually
spend some time in the help lesson to learn about all the
more obscure features.
  

    
Could you say a little more about the funding of PLATO? It sounds as
if the system would have been hugely expensive per student/user. Any
idea of the costs in 1960s-1970s $? Who funded it and championed it and
under what circumstances did the money eventually disappear?

Also, would it be at all possible to run PLATO and some of the apps in
emulation on computers today and thereby at least give people a
glimpse of what it looked like then? Is anyone working on that? Is the
source code available?
  

    

On the emulation front - as I mentioned, go see http://cyber1.org.
It is the PLATO system, for all intents and purposes.  It's a copy
of the system software running on an emulated CYBER platform on
a plain old PC running a variant of Linux.

The old CYBER mainframes ran the NOS operating system and the
COMPASS assembly language.  Cyber1 has NOS and COMPASS being
emulated on the Linux box, and then the "PLATO job", in its
original source code, is running on top of NOS.

You can read all about it at http://cyber1.org.

The funding story is more complicated -- will have to wait until
I get back from a meeting later today.  Suffice to say, the major
funding was from a Tri-Services contract (Army, Navy, Air Force),
ARPA, and biggest of all, NSF.