'arpa' Episodes

Networking for a Nuclear War, the Americans

     8/11/2019

In this episode we are going to explore the ARPANET. This is a companion to the last episode, which covered contemporary Soviet attempts to create an early internet.

Like with last time, today we are still in the Cold War era. Now, this won't be a point by point comparison of Soviet to US networks. They are totally different beasts. Instead, what I want to do is look at how ARPANET was developed, what influenced it, and how it would kick start the creation of the internet.


Bob Tayler: ARPA to PARC to DEC

     1/15/2021

Robert Taylor was one of the true pioneers in computer science. In many ways, he is the string (or glue) that connected the US governments era of supporting computer science through ARPA to innovations that came out of Xerox PARC and then to the work done at Digital Equipment Corporation’s Systems Research Center. Those are three critical aspects of the history of computing and while Taylor didn’t write any of the innovative code or develop any of the tools that came out of those three research environments, he saw people and projects worth funding and made sure the brilliant scientists got what they needed to get things done.

The 31 years in computing that his stops represented were some of the most formative years for the young computing industry and his ability to inspire the advances that began with Vannevar Bush’s 1945 article called “As We May Think” then ended with the explosion of the Internet across personal computers. 

Bob Taylor inherited a world where computing was waking up to large crusty but finally fully digitized mainframes stuck to its eyes in the morning and went to bed the year Corel bought WordPerfect because PCs needed applications, the year the Pentium 200 MHz was released, the year Palm Pilot and eBay were founded, the year AOL started to show articles from the New York Times, the year IBM opened a we web shopping mall and the year the Internet reached 36 million people. Excite and Yahoo went public. Sometimes big, sometimes small, all of these can be traced back to Bob Taylor - kinda’ how we can trace all actors to Kevin Bacon. But more like if Kevin Bacon found talent and helped them get started, by paying them during the early years of their careers… 

How did Taylor end up as the glue for the young and budding computing research industry? Going from tween to teenager during World War II, he went to Southern Methodist University in 1948, when he was 16. He jumped into the US Naval Reserves during the Korean War and then got his masters in psychology at the University of Texas at Austin using the GI Bill. Many of those pioneers in computing in the 60s went to school on the GI Bill. It was a big deal across every aspect of American life at the time - paving the way to home ownership, college educations, and new careers in the trades. From there, he bounced around, taking classes in whatever interested him, before taking a job at Martin Marietta, helping design the MGM-31 Pershing and ended up at NASA where he discovered the emerging computer industry. 

Taylor was working on projects for the Apollo program when he met JCR Licklider, known as the Johnny Appleseed of computing. Lick, as his friends called him, had written an article called Man-Computer Symbiosis in 1960 and had laid out a plan for computing that influenced many. One such person, was Taylor. And so it was in 1962 he began and in 1965 that he succeeded in recruiting Taylor away from NASA to take his place running ARPAs Information Processing Techniques Office, or IPTO. 

Taylor had funded Douglas Engelbart’s research on computer interactivity at Stanford Research Institute while at NASA. He continued to do so when he got to ARPA and that project resulted in the invention of the computer mouse and the Mother of All Demos, one of the most inspirational moments and a turning point in the history of computing. 

They also funded a project to develop an operating system called Multics. This would be a two million dollar project run by General Electric, MIT, and Bell Labs. Run through Project MAC at MIT there were just too many cooks in the kitchen. Later, some of those Bell Labs cats would just do their own thing. Ken Thompson had worked on Multics and took the best and worst into account when he wrote the first lines of Unix and the B programming language, then one of the most important languages of all time, C. 

Interactive graphical computing and operating systems were great but IPTO, and so Bob Taylor and team, would fund straight out of the pentagon, the ability for one computer to process information on another computer. Which is to say they wanted to network computers. It took a few years, but eventually they brought in Larry Roberts, and by late 1968 they’d awarded an RFQ to build a network to a company called Bolt Beranek and Newman (BBN) who would build Interface Message Processors, or IMPs. The IMPS would connect a number of sites and route traffic and the first one went online at UCLA in 1969 with additional sites coming on frequently over the next few years. That system would become ARPANET, the commonly accepted precursor to the Internet. 

There was another networking project going on at the time that was also getting funding from ARPA as well as the Air Force, PLATO out of the University of Illinois. PLATO was meant for teaching and had begun in 1960, but by then they were on version IV, running on a CDC Cyber and the time sharing system hosted a number of courses, as they referred to programs. These included actual courseware, games, convent with audio and video, message boards, instant messaging, custom touch screen plasma displays, and the ability to dial into the system over lines, making the system another early network. 

Then things get weird. Taylor is sent to Vietnam as a civilian, although his rank equivalent would be a brigadier general. He helped develop the Military Assistance Command in Vietnam. Battlefield operations and reporting were entering the computing era. Only problem is, while Taylor was a war veteran and had been deep in the defense research industry for his entire career, Vietnam was an incredibly unpopular war and seeing it first hand and getting pulled into the theater of war, had him ready to leave. This combined with interpersonal problems with Larry Roberts who was running the ARPA project by then over Taylor being his boss even without a PhD or direct research experience. And so Taylor joined a project ARPA had funded at the University of Utah and left ARPA. 

There, he worked with Ivan Sutherland, who wrote Sketchpad and is known as the Father of Computer Graphics, until he got another offer. This time, from Xerox to go to their new Palo Alto Research Center, or PARC. One rising star in the computer research world was pretty against the idea of a centralized mainframe driven time sharing system. This was Alan Kay. In many ways, Kay was like Lick. And unlike the time sharing projects of the day, the Licklider and Kay inspiration was for dedicated cycles on processors. This meant personal computers. 

The Mansfield Amendment in 1973 banned general research by defense agencies. This meant that ARPA funding started to dry up and the scientists working on those projects needed a new place to fund their playtime. Taylor was able to pick the best of the scientists he’d helped fund at ARPA. He helped bring in people from Stanford Research Institute, where they had been working on the oNLineSystem, or NLS. 

This new Computer Science Laboratory landed people like Charles Thacker, David Boggs, Butler Lampson, and Bob Sproul and would develop the Xerox Alto, the inspiration for the Macintosh. The Alto though contributed the very ideas of overlapping windows, icons, menus, cut and paste, word processing. In fact, Charles Simonyi from PARC would work on Bravo before moving to Microsoft to spearhead Microsoft Word.

Bob Metcalfe on that team was instrumental in developing Ethernet so workstations could communicate with ARPANET all over the growing campus-connected environments. Metcalfe would leave to form 3COM. 

SuperPaint would be developed there and Alvy Ray Smith would go on to co-found Pixar, continuing the work begun by Richard Shoup. 

They developed the Laser Printer, some of the ideas that ended up in TCP/IP, and the their research into page layout languages would end up with Chuck Geschke, John Warnock and others founding Adobe. 

Kay would bring us the philosophy behind the DynaBook which decades later would effectively become the iPad. He would also develop Smalltalk with Dan Ingalls and Adele Goldberg, ushering in the era of object oriented programming. 

They would do pioneering work on VLSI semiconductors, ubiquitous computing, and anything else to prepare the world to mass produce the technologies that ARPA had been spearheading for all those years. Xerox famously did not mass produce those technologies. And nor could they have cornered the market on all of them. The coming waves were far too big for one company alone. 

And so it was that PARC, unable to bring the future to the masses fast enough to impact earnings per share, got a new director in 1983 and William Spencer was yet another of three bosses that Taylor clashed with. Some resented that he didn’t have a PhD in a world where everyone else did. Others resented the close relationship he maintained with the teams. Either way, Taylor left PARC in 1983 and many of the scientists left with him. 

It’s both a curse and a blessing to learn more and more about our heroes. Taylor was one of the finest minds in the history of computing. His tenure at PARC certainly saw the a lot of innovation and one of the most innovative teams to have ever been assembled. But as many of us that have been put into a position of leadership, it’s easy to get caught up in the politics. I am ashamed every time I look back and see examples of building political capital at the expense of a project or letting an interpersonal problem get in the way of the greater good for a team. But also, we’re all human and the people that I’ve interviewed seem to match the accounts I’ve read in other books. 

And so Taylor’s final stop was Digital Equipment Corporation where he was hired to form their Systems Research Center in Palo Alto. They brought us the AltaVista search engine, the Firefly computer, Modula-3 and a few other advances. Taylor retired in 1996 and DEC was acquired by Compaq in 1998 and when they were acquired by HP the SRC would get merged with other labs at HP. 

From ARPA to Xerox to Digital, Bob Taylor certainly left his mark on computing. He had a knack of seeing the forest through the trees and inspired engineering feats the world is still wrestling with how to bring to fruition. Raw, pure science. He died in 2017. He worked with some of the most brilliant people in the world at ARPA. He inspired passion, and sometimes drama in what Stanford’s Donald Knuth called “the greatest by far team of computer scientists assembled in one organization.” 

In his final email to his friends and former coworkers, he said “You did what they said could not be done, you created things that they could not see or imagine.” The Internet, the Personal Computer, the tech that would go on to become Microsoft Office, object oriented programming, laser printers, tablets, ubiquitous computing devices. So, he isn’t exactly understating what they accomplished in a false sense of humility. I guess you can’t do that often if you’re going to inspire the way he did. 

So feel free to abandon the pretense as well, and go inspire some innovation. Heck, who knows where the next wave will come from. But if we aren’t working on it, it certainly won’t come.

Thank you so much and have a lovely, lovely day. We are so lucky to have you join us on yet another episode. 


(OldComputerPods) ©Sean Haas, 2020