The Digital Antiquarian

My search for the original Oregon Trail code as first written at Carleton College having turned up nothing, I set out to find the earliest version I could. That turned out to be the one posted by one Deserthat in his blog. This version appeared in the July-August 1978 issue of Creative Computing. But before I get to that let me talk just a bit about the seven years that elapsed between the game’s creation and its appearance in Creative Computing.

Rawitsch, Heinemann, and Dillenberger were roommates enrolled in the student teaching program at Carleton College in 1971; Heinemann and Dillenberger taught math, Rawitsch taught history. It was Rawitsch who first conceived of and designed The Oregon Trail as a board game, and Heinemann and Dillenberger who suggested that it be computerized and then programmed it in Time-Shared BASIC. The first outside of this trio to play it were the students in Rawitsch’s history class, on a teletype machine which Rawitsch wheeled into his classroom on December 3, 1971. According to Rawitsch, they “loved it.” Rawitsch and many other teachers in the Minneapolis school district used it frequently for the remainder of the term. When Rawitsch left the district in 1972, however, he deleted The Oregon Trail from the system and took it with him as a long role of printed paper which he ended up tossing into a drawer somewhere and forgetting about.

As it happens, Minnesota was something of a center of computer innovation in these days. In 1973 the state’s legislature founded an organization called The Minnesota Educational Computing Consortium (MECC), with a mandate to seek out and implement new applications for computers in education. A UNIVAC 1100 mainframe was installed at MECC’s Minneapolis headquarters, and some 1500 terminals were connected to it from schools throughout the state. In 1974, MECC hired Rawitsch to work as “a liason between MECC and a group of community colleges.”

Encouraged to think about new applications for computers in education, Rawitsch recalled that yellowing printout of The Oregon Trail. He punched it into a teletype connected to MECC’s UNIVAC over a “long Thanksgiving weekend” in 1974. The UNIVAC was a mainframe rather than a minicomputer like the HP-2100 — a very different proposition altogether. Luckily, it did have a version of BASIC available. [Actually, this is not quite correct. I learned after writing this that Rawitsch initially keyed the program into another HP-2100 system. See Part 4 of this series for more about that.] In addition to presumably being modified by Rawitsch or someone else at MECC to suit the UNIVAC’s implementation of BASIC, The Oregon Trail was also enhanced by Rawitsch himself to be more historically accurate, consistent, and entertaining. That version was played by thousands of schoolchildren all over the state during the next several years.

In 1977, MECC replaced its aging UNIVAC with a top-of-the-line CDC Cyber-73 system, and The Oregon Trail was modified once again to run on that system. This was the version that appeared in Creative Computing in 1978, and the version I back-ported to HP Time-Shared BASIC in order to experience The Oregon Trail in something close to its original form.

Now, as for Creative Computing… well, there’s one hell of an interesting story there as well.

Founded in 1974 by David Ahl, Creative Computing was the first mass-market magazine devoted to computers, predating even Byte by a full year. In keeping with its name, Creative Computing approached its subject not as an exercise in business and engineering, but as an artistic and cultural phenomenon. Its pages have plenty of technical advice and program listings, but also plenty of speculation on what the impending computer revolution really means. These “soft” articles are perhaps not surprising when one considers that many of the people who read the magazine could only dream of and speculate about access to a real computer of their own. There is much enthusiasm, but also some trepidation, most prominently in repeatedly expressed concerns about civil liberties in the new world to come. Both its idealism and its anti-authoritarian bent seem anchored in the by then fading counterculture of the late 1960s. I’m always reminded when I read it of the atmosphere around Berkeley’s Community Memory project, as described so well by Stephen Levy in Hackers.

Before founding Creative Computing Ahl was Education Marketing Manager for DEC. In keeping with this background and with its humanistic focus, Creative Computing was always very interested in the use of computers for education, devoting many pages to the subject in almost every issue. It’s thus not a big surprise that The Oregon Trail ended up there; if anything, the surprise is that it didn’t appear sooner.

By the time it did, MECC, which shows every sign of having been quite the visionary organization, was already getting involved with the new generation of “microcomputers,” as they were called in those days. MECC began installing Apple IIs into many Minnesota classrooms that year. And, yes, The Oregon Trail was ported yet again to run on them.

And so, with that history dispensed with, I’ll finally have a close look at the game itself next time around, which is after all where I’ve been trying to go with this shaggy-dog story all along.

I mentioned last time around that the original Oregon Trail was written on an HP-2100 series minicomputer. That’s a pretty interesting topic in itself.

HP’s first computer line, the 2100 series could be equipped with a number of possible operating system. One of the most common, and the one under which The Oregon Trail was written, was called HP Time-Shared BASIC. This system was very unique in its time, and perhaps even visionary. Rather than placing the user in the command-line driven environment typical of virtually all other OSs of the period, Time-Shared BASIC, true to its name, dropped the user after login into an interactive BASIC environment. Not only could she write programs here using BASIC, but all of her other immediate interactions with the system — loading and saving files, etc. — were also done using BASIC statements.

This was the design concept later used by many of the 8-bit generation of personal computers, as anyone who ever typed “LOAD ‘*’,8,1” to start a game on a Commodore 64 can attest. Following the norms of the time, even the original IBM PC dumped the user into a little used and seldom remembered BASIC environment if it didn’t find a DOS disk to boot at power-on. I’d been curious for years how we got from the command-line driven environments typical of most institutional computing to the interactive BASICs of these machines; now I think I have an idea.

Time-Shared BASIC represented a more welcoming environment for working and programming than was typical of the time, and this fact combined with the relatively low cost and easy maintainability of the HP-2100 line made these machines favorites of universities and even high schools. HP seems to have put considerable effort into designing and marketing the HP-2100 as a more user-friendly, accessible sort of machine. This manual is particularly interesting, being an introduction to BASIC programming pitched to the complete novice. It’s actually really well done, managing to walk the fine line of being friendly and accessible without falling into condescension. In light of all this, then, it’s not at all surprising that an HP-2100 would have found its way to Carleton College.

There were quite a lot of games and educational programs written in Time-Shared BASIC, and some of these have ended up on the Internet in the form of an unorganized dump to a huge tape image. So, I decided to try to bring up an emulated version of Time-Shared BASIC on my computer and to look through this mass to see if there might be an copy of the original The Oregon Trail in there somewhere. Well, it sounded like a good idea at the time, anyway…

You see, while there is an HP-2100 emulator available thanks to the amazing efforts of The Computer History Simulation Project, Time-Shared BASIC was a pretty complicated configuration. It was in fact TWO HP-2100s, one serving as a sort of gateway to users who connected via remote terminals and the other hosting the core of the OS itself. So, emulating the thing means running two separate HP-2100 emulators, loading the appropriate software onto each, and linking them together via sockets. Finally, one opens a THIRD window on one’s PC to telnet into system via the loopback address. I never would have gotten anywhere close to a working setup if hadn’t been for a Yahoo group dedicated to the platform, who host in their files section an emulator setup that almost worked right out of the box. I won’t bore you with the details of my struggles to get from almost to completely working; suffice to say that I finally got my own little Time-Shared BASIC system up and running.

And so I started going through the tape dump. This was almost 6MB of data, a large quantity indeed for a collection of BASIC programs often only a few kilobytes in length. Alas, though, no joy on The Oregon Trail.

But what I did find was pretty darn interesting, and more than justified the time it took to get to this point. Here were literally hundreds of BASIC programs: games, educational programs dealing with every subject, practical scientific and mathematical tools, etc. I even found what appears to be the original version of the old Star Trek game. There was obviously quite a thriving culture of program development and trading on this system from the late 1960s to the late 1970s. That’s perhaps not so remarkable in itself. What is, though, is that these programs were being written and used not by a priesthood of professionals as in the world of the IBM mainframe or a collection of focused hackers and researchers as in the world of the DEC PDP line, but rather by everyday students and educators. This gives their work a very different character. And if this sampling of their work is anything to go by, these people were very, very interested in games.

This mother lode deserves more attention, and I’m going to try to give it some and perhaps post a bit more about it in the future. (In particular, I want to see if I can find a version of the Hamurabi game Jason Dyer mentioned in a comment to the first post in this series.) But before I do that I’ll get back to The Oregon Trail proper next time around.

I recently got a copy of 1001 Video Games You Must Play Before You Die. It’s not really a very good book, for reasons that are interesting on their own and that I hope to talk about in another post very soon. Right now, though, I want to talk about the very first entry in the book, on The Oregon Trail, because that entry sent me down a rabbit hole from which I have only just emerged, blinking and reconsidering the history of interactive narrative.

If you’re of a certain age and nationality (i.e., mine), you almost certainly know The Oregon Trail. From the early 1980s until well into the 1990s virtually every public school in America seemed to have at least a few Apple IIs off in a corner somewhere, and one of the titles available on them was guaranteed to be this little educational game which placed the player in the role of a would-be settler setting off from Missouri on the long journey to the Oregon Territory. Those versions communicated mostly in text, but they spiced up their presentation with lots of colorful graphics, and were appealing enough to become favorites among students then and to still be nostalgically remembered by millions today. In fact, I just learned that there is now a Facebook app of the game.

What’s not often realized is that even when it first arrived on the Apple II The Oregon Trail was already a very old game. That’s why it’s the first entry in the chronologically arranged 1001 Video Games. (Actually, the fact that they got this date right is kind of surprising, because there are a heck of a lot of others that they got wrong. But I promised not to kvetch about the book right now…) It was in fact first written in 1971 by three educators at Carleton College, a small liberal arts college in Northfield, Minnesota. Don Rawitsch, Bill Heinemann — no relation to the “Burger Bill” Heineman who worked on The Bard’s Tale series among other games — and Paul Dillenberger wrote the game in BASIC on an HP-2100 series minicomputer.

When I was writing my history of IF, I named two programs as the most important predecessors to the landmark Adventure (1976-77): Joseph Weizenbaum’s Eliza (1966), which first deployed the basic system of IF interaction (albeit in the context of an elaborate parlor trick rather than a game) and Gregory Yob’s Hunt the Wumpus, a simple game in which the player moves from room to room in a maze while attempting to avoid and eventually kill the eponymous Wumpus. The Oregon Trail makes me think that there should be a third entry on that list.

Let’s consider the state, such as it was, of interactive narrative in 1971. While there had been experiments with interactive storytelling before in the mystery genre, examples of the form were pretty thin on the ground. Edward Packard had already tried to get the first of what would become the Choose Your Own Adventure line of books published, but had been rejected by every publisher he had turned to, and would have to wait years more to see his idea in print. A group of scruffy wargamers in Wisconsin were toying around with the systems that would become Dungeons and Dragons, but, again, their work was years from publication. Wargames and other simulation games certainly had an experiential component, implicitly inviting their players to imagine the events they simulated unfolding in their imagination, but said events unfolded from the perspective of a god on high rather than that of an individual player in storyworld. In the world of computers, there was some ongoing work into computer-generated narrative among artificial intelligence researchers, but these were not really interactive narrative, but rather self-contained stories that the computer generated beforehand based on a set of input data and played out for an audience.

Yet The Oregon Trail opens by telling us, “Your family of five will cover the 2040 mile Oregon Trail in 5-6 months – if you make it alive. You had saved $900 to spend for the trip, and you’ve just paid $200 for a wagon.” It’s dropping us into a storyworld, and inviting us to take a role there and decide what happens next. Was there a computer program before this that so obviously wanted to make a story with (as opposed to for) us? I don’t know of it if there was.

So, I set off on a quixotic quest to experience The Oregon Trail in as close to its original form as I could manage. More on that next time.