Global Direct Investment Solutions

Corporate Development for a Networked World

OnTheShortList.com     SurgeUSA.org

ShortListNews.com

Bruce Donnelly - biographic note about PLATO

Bruce Donnelly   bruce@gdi-solutions.com    (Biography)

The world is still catching up with that vision of what individuals can do together as the technology to share knowledge and ideas quickly becomes more transparent.

Children today have a hard time imagining any other outcome, just as we have a hard time imagining life without electricity, cars, airplanes, or telephones even though these technologies have only been around for a few generations.  Today, instant messaging, e-mail, web conferencing, and the use of websites, databases, and educational games or graphical interfaces such as plasma panels for digital images are taken for granted.

In 1974, on PLATO, it was breathtaking and revolutionary.  It was like being on the Carousel of Progress ride (1964 NY World's Fair - I was there - later at EPCOT) and being able to jump decades into the future by simply walking into the next room.

I abandoned my study of computer engineering in 1975 after working briefly on PLATO (an acronym - Programmed Logic for Automated Teaching Operations, I think).  PLATO ran on a supercomputer network based at the University of Illinois' Computer-Based Education Research Lab (CERL).

The name presumably reflected some academic ideals among the founders about the Socratic method of learning through exploratory questions, dialogue, and creative thinking from different perspectives, rather than just the drill and practice "instructional" programs which pushed a packaged set of facts at students to memorize in that era.  I don't think it reflected the elitist vision of Plato's Republic (philosopher kings, etc.), as though academics were the highest form of life.  Instead, PLATO empowered individuals to explore openly shared knowledge in very flexible ways along many paths of discovery.

I was also influenced by reading my original copy (eat your hearts out, computer nerds) of Computer Lib, Dream Machines by Theodor Nelson.  Look at the reviews on Amazon.com, which you can find through the Bookshelf section under Knowledge Management.

Nelson's visionary work, which included some content about PLATO (on which he spent some time in that era), foresaw the Internet and hypertext in much the same way as science fiction writers were able to visualize a future which they couldn't actually create or reliably predict.  He spotted the trend and the potential power of flexible access to knowledge at a time when most people still thought computers were just for programmers and other experts, as if you quite literally needed to be one of those rocket scientists at NASA (remember : Apollo 11 was recent history - 1969) or a mathematician, engineer, etc.

At the time, people could visualize computers automating tedious or complex repetitive tasks like factory or office automation, but their role to quickly share universal access to vast knowledge and as a platform to communicate and collaborate flexibly as individuals over global computer networks was not widely recognized.

For example, note that even in the book and movie "2001: A Space Odyssey", the astronauts did not actually communicate or collaborate with people elsewhere through their Hal 9000 computer.  They communicated through was was basically a videophone (already promoted by AT&T at the time - as at the 1964 World's Fair in New York), which wasn't that different than seeing TV images from Apollo missions.  In some respects, even science fiction hadn't yet caught up to this vision of the future of computers in the early 1970's.  Note that the Hal computer to just help run such a complex spacecraft was envisioned as something large enough that Dave could float around inside it to remove memory cards, which reflected the large mainframe environments of that era.

Perhaps more important than his own hypertext idea (Xanadu), Nelson recognized the paradigm shift of vast computing power shifting to the individual and to networks of small computers, rather than to the existing mainframe hardware and software suppliers or programmers.  This was quite radical in 1974 after an era of social unrest as well as the nagging fears of powerful new technology for military and other applications.  The "power to the people" message wasn't the general direction of the computer industry.

Unlike the threatening vision of an all-seeing and incredibly intelligent computer run amok beyond human control as in "2001 A Space Odyssey" (movie : 1968) or more obscure variations on this theme (FailSafe, The Forbin Project, etc.), Nelson foresaw a scenario which was more like Dorothy waking up in Oz, or Alice falling into Wonderland, with both good and bad potential according to how this power was used.

In any case, incredible adventures lay ahead though the open sharing of knowledge.  Computer Lib was a call to action for individuals to seek and defend that new personal freedom which these sorts of Dream Machines already made it possible to envision.

http://www.platopeople.com

This is for the PLATO History Project by Brian Dear, who is writing a book for which he has interviewed many people and has already shared some interesting photos.  The lack of a reliable history of PLATO has led to confusion and ignorance of it, including prominent tech and computer museums where they should know better.  Someday, that should prove to be an interesting book, but it remains a work in progress.

http://www.cyber1.org

Cyber1 is aimed particularly at bringing many former members of the PLATO community together again, and actually provides the opportunity to experience again what PLATO was like through access to a working copy of an early version, so that one can emulate a PLATO terminal and use some of the most popular old programs.  It's interesting to think that the old CDC supercomputer platform can now be simulated on a Mac or a server.

If you were a PLATO user, or are in a position to support their efforts to preserve this unique piece of computer history, please contact them about how to help.

http://www.thinkofit.com/plato/dwplato.htm

The above website, by David Woolley, includes some useful insights into web conferencing, forums, and other communications tools.  He was one of several PLATO pioneers in that field.

Thanks to the above for sharing their knowledge and time, and informing people about PLATO's history, which has long been neglected by those who never experienced it, or think the networked computer world started with IBM mainframes, PC's, and the Internet.  The efforts of Control Data Corporation to market the PLATO system failed, especially after they tried to migrate some of the capabilities to PC's in the 1980's which were never really designed to bring so much power to the end-user.

We knew already, in the early 1970's, that companies like CDC with some incredible technology were still no match for the IBM marketing juggernaut of that era.  The 64 bit architecture behind PLATO in the early 1970's is finally emerging (not the same, obviously) in the next generation of Intel microprocessors, 30 years later.

Another summary can be found at :

http://www.livinginternet.com/r/ri_talk.htm

There were also interesting games on PLATO, both for individual users and for very rapid competition between users on terminals anywhere on the network.  You can find some history about this aspect of PLATO, including Talkomatic and Notes as well as Flight Simulator and other games, on

http://www.computergaming.com

I don't pretend to be an expert on the subject, but the basic networking model behind PLATO involved state-of-the-art supercomputers from Control Data, each with their own networks of up to 1000 terminals which were grouped together into 32 clusters (network interface units) of 32 or fewer terminals (like a classroom).  This meant communications could be multiplexed across plain old phone lines for many users at a site because, at any given instant, the process of communication keypress by keypress (rather than page by page) meant that there would be very few conflicts between 32 users for this limited bandwidth - and even many such sites wouldn't burden such a computer.

In effect, by receiving each key as fast as it was pressed, and already knowing most of the options available to the user from any page, the system could be ready in the background to respond very rapidly to almost any request, rather than delayed by communications bandwidth as large packets of information are sent or received between the computers.  That's why, even with high bandwidth communications channels today for the Internet users, the performance can be so slow.  What would the Internet experience be like today if anything could be accessed in a small fraction of a second, without the high costs of building and using a very high bandwidth communications infrastructure?  PLATO was designed to perform at such speeds over plain phone lines and modems that would be unthinkably slow to users today.

This differs quite a bit from the Internet or distributed client-server models, although the computers behind PLATO could potentially be networked through high-speed connections to work across more than the terminal's host computer.  The bandwidth was needed between the supercomputer servers of content - not between the user terminals and computers as would be analogous to ISP connections for individual users or office clusters today.

The addition of microprocessors to create "smart terminals" already started to distribute some of the routine or intensive computing activity (graphics, etc.) away from the central supercomputer servers and network interface to the end-user's terminal in the mid-1970's, such as to provided enhanced graphics capabilities and speed long before PC's.

Remember : this was in 1974.  Very limited PC's rolled out in the early 1980's, at which time few could imagine why an individual would ever need a 10MB hard disk.

With typical response times on the network of less than 0.2 seconds, one would literally hear groans of despair within seconds if there was no visible response to a key, as an entire room of users would realize the network was down in far less time than even a very fast webpage downloads today.

Compare that to the usual task at that time of submitting batch programs as a stack of Hollerith (keypunch) cards on IBM mainframes, where one could wait hours to discover that there was a keypunching error on one card.

We could literally cross the street from 1970's technology as computer engineering students to the radically different future that PLATO was demonstrating at CERL by creating a dynamic online community beyond anything the developers had imagined.

The estimated download time for this simple text page over a 56.6 phone line is 8 seconds.  Imagine going to any content on this website in roughly 0.2 seconds, and think about all the websites which are even slower because of complex graphics.  We only had low-resolution monochrome graphics (orange pixels), but there was experimentation with slow-scan TV technology and other ways to bring very realistic imaging and color to the system within the bandwidth constraints of something designed for wide-area networks of schools which, realistically, had very limited resources and therefore needed the cost per terminal and phone line to be low.

After 1975 I had no further exposure to PLATO until I eventually saw some of the PC software demonstrated much later, which left me completely unimpressed because it was now little better than the primitive drill and practice and test applications of long ago, when IBM and others were pushing "Computer-Assisted Instruction" as a very different approach than "Computer-Based Education".

CAI pushed instructional content at users fairly relentlessly, like an automation process in which the programmer decides what is appropriate for students to see at each step.  CBE empowered users to organize and use the content they wanted in very flexible ways to seek knowledge, in much the same way as hyperlinks easily provide a way to follow an almost infinite variety of paths through information.  This reflected fundamental principles about the future of education as an individualized process through which the computer was a tool for sharing knowledge flexibly and stimulating collaboration among users in ways which couldn't even be predicted, rather than just automating basic or repetitive tasks in a standardized classroom.

We networked such systems from Germany to Washington DC, which was not easy in those days, and developed a local database system to manage contacts and other work for the largest US consulate general in the world.  This enabled us to stop inviting dead or retired contacts to large parties, or sending out duplicate invitations, or getting their titles wrong, or just dealing with the simple fact that formal German address labels differ in format from US ones.  Revolutionary?

The CRM database and other tools I use for my work today are much better!  They reflect skills developed over the last 20 years at PricewaterhouseCoopers, Deloitte & Touche, the US Foreign Service, and by managing thousands of US business contacts for a British regional development agency.

As the three communities I serve converge (corporate executives, professional service providers, and economic development professionals) as a market around their common interests in the faster and better development of corporate investment projects, there will be other innovations to support the participants because the key to this business is not what I can accomplish alone, but rather what we can all achieve together. 

GDI Solutions is not about publishing another website full of many useful links.  It is about changing the way these three contact networks interact as a community to deliver much higher value.  It isn't about connecting to websites.  The real value is still in connecting talented people to work better together through very flexible and focused teams, as the lesson of PLATO made obvious thirty years ago.  The role of GDI Solutions is as an intermediary and catalyst to bring talent together with needs.

Thank you once again to the people at PLATO who completely shattered my plans 30 years ago to become yet another young computer engineer or programmer by demonstrating the parallel universe of a future empowered beyond the realm of IBM marketing and technological constraints.  Needless to say, I was not overwhelmed by the arrival of the PC, or the progress of the PC and Internet over the last 20 years.

Are we having fun sharing knowledge and ideas rapidly for real benefit yet?

What if PLATO hadn't failed?  What if it had successfully adapted to the emerging technologies, and kept bringing even more power to a growing user community?

In any case, it did fail.  We move forward with the technology we have available.  The paths we choose to share knowledge and increase collaboration among people have important global consequences, however, which are not always obvious at the time.  PLATO was, in the end, a missed opportunity to change the world 20 - 30 years ago.

We also had time-share access to a computer at the Illinois Institute of Technology over teletype terminals, so that we could prepare batch programs on paper tape and then run them.  This provide exposure to languages such as Fortran, and later study included PL/1, Algol, COBOL, and others about as useful to modern life as Latin.

Thus, I went from 1950's and 1960's technology to 2000-era technology within one year, which is another reason why PLATO was so mind-boggling in 1973-74.

The "green Coke machine", however, was a fun adventure.  Forget about "higher level" languages.  If you wanted the computer to add 2 plus 2, you had to program it in machine language of octal codes on paper tape, and manipulate the memory yourself.  Not even assembler language.  Just getting the machine to boot up and run long enough to process a program was an adventure, and it cranked out a lot more heat than useful output, but it provided unique appreciation for software such as operating systems and compilers for higher-level programming languages, and the term "memory dump" took on a whole new meaning.  It was a real dump, and figuring out why the machine did unexpected things was a constant challenge.

I don't think I have any photos to share from the days at CERL and other PLATO sites around the Champaign-Urbana campus.  Too bad.  Hopefully nobody else has pictures of me from that era either.

I will look forward to Brian Dear's book, and will list it in the Bookshelf section under Knowledge Management as soon as I know it is available through Amazon.com.  If anybody is aware of other relevant books on the subject, I welcome suggestions.  That includes not only PLATO, but also the contemporary issues of knowledge management, development of online communities, collaborative technology, etc.  Note that David Woolley's website ( www.thinkofit.com ) has some good tips.