History of computers, part 2 — TCP/IP owes a lot to Xerox PUP

An interview with Robert Taylor, former manager of Xerox PARC, shows TCP/IP designs were based a lot on the PARC Universal Packet (PUP) networking protocol suite.

TCP/IP owes a lot to Xerox PUP
Mike Knell (CC BY-SA 2.0)

To understand where we are going, we first must understand where we have been. This applies equally well to the history of nations across the globe as it does to computers and computer networking.

With that in mind, we’re taking a slow (somewhat meandering) stroll through the history of how computers talk to each other. Last time, we talked a bit about dial-up Bulletin Board Systems (BBSs) – popular through the 1980s and the bulk of the 1990s.

Today, I’d like to talk about one of the most influential, but rarely discussed, networking protocol suites: PARC Universal Packet (PUP).

Finding a good place to start this wonderful tale of nerdiness is somewhat challenging, as there are so many people and projects that led to PUP’s creation at Xerox’s Palo Alto Research Center (PARC). I think a nice starting point is… 1973.

Specifically, May 22of that year. On that day, Robert Metcalfe wrote a memo in which the name “Ethernet” was born. For those curious, it was named after “Luminiferous ether,” a disproven theory (in the 1800s) for “omnipresent, completely passive medium for the propagation of electromagnetic waves.”

Of course, Ethernet, itself, is great. But having a networking protocol and suite to use with it is even better. Thus, PUP was born in 1974.

Routing, packet delivery, a reliable byte stream, printer spooling, Telnet, FTP, name lookup — PUP was a robust networking suite, to be sure. All of which came a few years before TCP/IP came into full existence.

Xerox PUP inadvertently revealed

That leads me to a story I find rather amusing – a snippet from the oral history of Robert Taylor, manager of Xerox PARC throughout the 1970s, published by the Computer History Museum (pdf):

“The Xerox lawyers kept us from making PUP public, letting it outside. Vint Cerf at Stanford, who was a faculty member at Stanford in 1975, I think, or 1976, created a design committee amongst Stanford people mainly to come up with a protocol design to put on the ARPANET. He invited some people from PARC to come to these meetings, and no one outside of PARC knew that we had a working PUP.

“But the Xerox lawyers told the Xerox guys who went to this meeting at Stanford that they could go to the meeting, but they couldn’t tell the people at Stanford about PUP. So, Metcalfe and [John] Shoch went to these meetings, and [Dave] Boggs and [Ed] Taft might have as well.

“So, the Xerox guys went to this meeting, and some Stanford guy would lay a design idea out on the table and the Xerox guy would say, ‘Well, yes, but if you do that, you might have trouble with this idea over here or with this feature over here or this will get in the way of that.’

“And this conversation went around and around for a few times, and finally one of the Stanford guys sat back from the table and said, ‘You guys have already done this, haven’t you?’ And indeed, they had.

“So, Vint Cerf and Bob Kahn owe their fame to the Xerox attorneys. If it weren’t for the Xerox attorneys, nobody would have ever heard of TCP/IP. TCP/IP designs were based a lot on PUP.”

That’s right. Vint Cerf and Bob Kahn, often considered to be the fathers of the Internet, truly owe a great deal of that fame to the engineers at PARC who developed PUP.

That’s a little fact of computing history that doesn’t get talked about nearly enough.

A few years later, in 1977, PUP was succeeded by Xerox Network Systems (XNS), which, in turn, influenced several other early 1980s networking systems, including AppleNet. But that’s a story for another day.

Join the Network World communities on Facebook and LinkedIn to comment on topics that are top of mind.
Must read: 10 new UI features coming to Windows 10