Advent of Computing - Episode 188 - Training and Teaching with the Bi-Tran Six

Episode Date: September 1, 2026

How did you learn about computers? And how in depth was your education? For some the answer is the Bi-Tran Six: an educational machine released in 1965. This was one of the first computer trainers, and it represents a very interesting take on design. It's a machine that's small, simple, and can even be disassembled while running. But exactly how was it designed, and what was it used for? What makes the Bi-Tran Six so different from other machines? Like Advent of Computing? Then check out the after show! Adjunct of Computing is now LIVE: YouTube Spotify Apple Podcasts  

Plato has been one of those stories that's stuck in my head for years. In 1960, Don Bitzer at the University of Illinois started experimenting with computers in the classroom. He wanted to find ways to apply these new computer things to the problem of teaching. However, there's a catch. There's a bit of an issue. In 1960, these computer things weren't all they were cracked up to be, at least not yet. Over the course of the next decade and change, Bitzer and his collaborators would start to fix that. Along the way, they pioneered time-sharing, a way to split the power of a computer between multiple users. They made some of the first practical flat-panel plasma displays,

at a time when that technology only existed in straight patent filings. They would end up with fully graphical terminals serving thousands of students. that's not even to mention the software that wound up on these terminals. It would range from actual lessons to some of the first multiplayer games and even some very early first-person shooters. Seriously, it's an amazing and often forgotten part of our past. I can't recommend the book The Friendly Orange Glow enough that chronicles all the gritty details.

Plato is stuck around in my head because it's such a past, powerful story. Bitzer realized that computers weren't quite the right tool for teaching, but they showed promise. He was able to see a path to the future, and he started taking steps on that path. He goes from tech demos on a vacuum tube machine, all the way up to flat panel displays and time sharing. Folk have been trying to use computers in education for a very long time. Plato shows exactly how that can work. Bitser is able to apply digital technology to the task of teaching, to use something new, something fast and interactive as a tool to improve learning outcomes. But that's only one form of computer education. Bitser is teaching with computers.

The flip side is teaching about computers. You know, computer education. How do you tackle that? Is it even different? Would you want to use the latest groundbreaking technology that's never been seen anywhere else? Or would it be better to use something more simple? And does that approach come with its own peril? Welcome back to advent of computing. I'm your host, Sean Hass, and this is episode 188, Training and Teaching with the BITrans 6. Before we get started, I have my usual plug corner. As always, go listen to the after show. Adjunctive computing is still plugging along. It's a great venue for a more casual conversation about what's been missed in an episode, or just things that don't quite fit into my tight scope. It's fun, casual, and available

anywhere you're listening to this podcast right now. And the more timely news, I'm traveling next. week. I'm going to England to go on an archival trip to look up some very important information for a long-term project I've been working on. This trip has been partly funded through Patreon, through money that's been donated by listeners like you. So, first off, thank you. Second off, I'm doing something special while I'm on the road. I'm going to be doing a travel log of my trials and tribulations going through these archives. My plan is to talk about what I'm trying to find, the project I've been working on, and then what I eventually do dig up in pseudo-real time. I'm going to be posting that on Patreon throughout the week that I'm abroad. If that sounds exciting to you,

if you want to get a peek behind the curtain to my exciting long-term research projects, then you can head over to advent ofcomputing.com and click on support the show. That'll take you to Patreon, where for just $1 a month, you can support me doing this podcast and also a few bigger things I have in the works. All right, with all that said, let's start the show proper. Today, well, we're talking about a classic folly of mine. Simply put, I saw an interesting ad for a computer in an old magazine. One thing led to another, and now you get about an hour of me rambling. Today we're going to be talking about the Bitran 6, an early and remarkably small computer trainer. Trainers are one of those forgotten computer form factors, or perhaps computer niches.

A trainer is a simplified computer system that's designed as a tool to teach people about computers. this is a very specific educational niche. There are some computer systems that were meant to be tools for use in schools, as in the computer was used for teaching about something. An example of that is the Play-Dohs system that I talked about in the cold open. Trainers are tools designed to teach about computers themselves. That gives trainers a very particular angle. General-purpose educational computers will be designed to be user-friendly or to otherwise hide their digital details. A trainer is meant to teach you all those details, so you end up with full access to the machine's guts. A trainer's simplicity also leads to very cheap computers.

However, that's plotted on its own access. If you squint, it can look like a trainer would be an attractive entry-level machine, in a way they are. They give neophytes an introduction to computing. But you wouldn't really want to buy a trainer if you wanted to do serious computing. These things are cheap because they're limited. They're limited because they're meant to be easier to understand than, say, a mainframe. Back in the day, there was a very large tradition around computer trainers. There were a bunch of different machines that were produced.

The bi-trans6 has caught me attention for two reasons. First is its age. The trainer was built in 1965. That's well before microprocessors become commercially available, or commercially exist at all. It also predates the first use of integrated circuits in a computer. That's already interesting to me. Somehow we have a cheap and fully programmable computer that doesn't use ICs. Better still, it's a six-bit architecture, so on the numbers, Sounds pretty interesting.

The second reason comes down to the company that manufactured the Bitran. That company, Faberitech, wasn't really known as a computer manufacturer. They were better known for making magnetic core memory. The origin story of FaberTech is interesting and covers a few key details that I wanted to touch on for a while. This episode is going to be a little rangy, so let me lay out the focus. We're going to talk about why someone would create a computer trainer in the 1960s and why someone would buy a trainer. I want to look at how a trainer differs from a more normal computer.

I want to figure out why and how a core memory company might start creating such a specialized machine. Along the way, we have to talk a little bit about manufacturing magnetic core. Local boy makes good. That's a pretty common form of a newspaper article, and one that I don't particularly love. Whenever someone from a small town or even a city to some extent does something exciting, these types of articles flood the papers. I get the excitement, but I don't like them.

They clog up the historical record without giving us really that much information. It's usually that so-and-so from Such-and-such did a one-sentence thing, in this year. Details are few. It's really more a signal than a data point, I'd argue. Merlin Mickelson, local boy from Worthington, Minnesota, was the topic of many such articles in his day. Luckily, one of these puff pieces does give us some detail. In an odd twist, it's from the Reader's Digest. I'm laughing here because I don't think I've ever actually... read the Reader's Digest. I do remember my grandma having shelves of hard-backed volumes of readers' digests, but I never dared to open them. So who is Merlin? Besides a local boy that made

good, of course. The Digest tells that Merlin was a merchant marine and, after a few years of service, joined Remington Rand. Nicholson entered the digital world in 1953. That meant one thing, Univac. EMCC, the company that developed and built Univac, wasn't very financially successful. Over the years, they would be bought and sold to a number of different companies. In 53, they were owned by Remington Rand. And like that, we hit our first mystery. The Digest tells that Mickelson designed and built magnetic core memory systems for Univac while at Remington Rand.

He would do so until 1957. But those years don't really line up that well. There's a bit of a complication with naming here. Univac was originally the name of one computer. But later, there would be other computers that would use. similar names, so Univac gets renamed as Univac 1. That machine was in production in this period, but it used Mercury delay lines for its memory. It didn't use magnetic core. Univac 2 would use magnetic core memory, but the first example of that machine didn't ship until 1958. There's only

one machine that Mickelson could have been slinging core for, at least maybe. It's the Univac 1103A. I know, super specific number. This also isn't even a Univac Univac. Rather, it was based off work done at ERA, another early computer company that was bought by Remington Rand. The 1103A was a version of the 1103. And the A model did use magnetic core memory. It's released to market in 1956, and it would have been under development while Mickelson was at Remington Rand. So I think that's probably the machine that he was doing this work for. Magnetic core in this period is very, very new. It had first been used to store data in 1953.

So by the time Mickelson is involved, the technology is two years old. Hence, why there aren't many options for a Univac with core memory in this time range. Michelson was doing very fiddly work. Magnetic core is composed of, well, cores. These are little ferrite donuts. Those ferrites are then strung up on a grid of enamel-coated copper wire. Once the grid is strung, a final wire called the sense wire is strung through every core in a zig-zag pattern. It's delicate work, closer to weaving than soldering, and you have to manage these little

cores that are millimeters across. It's also highly technical, but in a way that most computerists of the period would not understand. You can't just go out and get working ferrite cores. In order to function, the core needs to have certain magnetic properties. The manufacturing process for cores, especially in this period, had issues.

Batches of cores had yields below 50%, That is, less than half of those ferrite donuts had the correct magnetic properties needed to store information. The concept of yields is very familiar in the era of integrated circuits. You can find endless horror stories about silicon wafers that are reduced to useless rocks, basically, by tiny inconsistencies or failures in manufacturing, or even for no reason at all. But that's still some 20 years in the future from where Mickelson's sitting. Yield and quality control meant something very different in the 1950s. Except, that is, for magnetic core memory.

So core ends up being a pretty specialized device, and its manufacturing is a specialized process. Merlin was ambitious in those days, or at least he was open to a good, opportunity. According to Reader's Digest, Mickelson was approached by Argonne National Laboratory in 1955. The lab wanted Mickelson to manufacture magnetic core memory for them. Argonne would supply the materials, and Mickelson would ship them the completed components. How did this come to pass? Oh, we aren't sure. Mickelson didn't leave behind any interviews, at least none that I can find,

and the article from Reader's Digest does not have citations. It appears to take details from an earlier newspaper article, but that's as far as the trail goes. What we do know is what happened next. Mickelson started his own company. FaberTech was founded in his basement in Minneapolis. From there, it would grow, enough so that Mickelson is able to leave his job at Remington Rand.

And this is where we reach our next big mystery. Magnetic core memory was always, with a few exceptions, made by hand. The pattern of woven wire and the size of the cores simply made automatic machinery infeasible. One often repeated fact is that magnetic core memory was usually manufactured by women. The implication is that manufacturing this memory was similar to weaving or embroidery, skills that more women than men were familiar with. Some later factories would hire unemployed or retired textile workers even. When did that practice start?

The paper trail makes that hard to say. This was industrial labor. You don't get Local Boy Makes Good Articles when a guy scores a factory job, and you definitely don't get those articles when an older woman goes back to work. Liza Stark in a zine titled Hilda Wove All Those Wires Proposes that the first to take up the metal loom was one Hilda Carpenter Carpenter worked on Project Whirlwind and strung the first of its memory planes It's likely she was one of the only people on the project that knew how to use a loom

Stark points out that we know little else about Carpenter and the same goes for many women that built magnetic core memory planes. In the late 1950s, Mickelson would also rely on female labor to grow his business. He started by hiring housewives and retired women in his neighborhood. The lack of early sourcing

means that we don't exactly know why Mickelson made this decision. It may have been simple convenience. It may have come down to skill. One article from the 1960s suggests the latter. As FaberTech grew, they built factories around the country. You can trace their growth through job ads and articles in local papers. One of those articles mentions a new factory in St. Cloud that hired 300 women to weave magnetic core.

A spokesperson is cited as saying they hired women because of the, quote, dexterity and patience required, end quote, in weaving memory. An unspoken part here is financial. A reason that machinery was never really pursued as a way of manufacturing core at scale was cost. It cost more to design, build, manufacture, and run a machine than it did to use human labor. That means female labor specifically. The wage gap has always existed. In a field where low labor cost prevented automation, it was crucial to keep labor cost low.

All of this leads to a gap in the historical record, a really glaring gap. FaberTech may have been one of the first manufacturing outfits to set the trend around female labor and magnetic core memory. But we're left to guess at original intent. What we do know is that the business model was lucrative. Nicholson would become a millionaire within a few years. Fabrotech would make core for Honeywell, GE, CDC, the U.S. military, and even NASA. If an outfit didn't have the capacity to make their own core,

FaberTech could do it for a nominal fee. So how do we go from memory to educational computer? That's the third big mystery and perhaps the most difficult to answer. The force behind the Baitrans 6 was one Dr. Abe Frank. The man is, at least from my current vantage point, a little enigmatic. I've been in communications with the VIP Club who have been helping me track down some information. That club is a group of retired IT pioneers in the Twin Cities area, and a wealth of information for this kind of thing.

As it stands, there's a large trove of documents from Frank sitting at the Charles Babbage Institute's archives. It's a lot of stuff. If I had all the time in the world or lived in Minnesota, I'd be a little inclined to go sift through all those documents. At time of recording, this is the biggest blind spot in the episode's research. I suspect the full story of this transition is hidden in one of the handful of folders marked memos or correspondence.

Let this serve as a challenge to someone listening. Go diving. Please, because I think there is a really interesting story hidden in this bridge. I have a few requests for scans out, and if those shed any light on the story, I'll definitely issue a follow-up. But I'm serious here. If someone listening is interested in this part of the tail, get in touch. I can hook you up with my contacts and you can go crazy. All of this is to say what I'm presenting today is based off public statements,

newspaper articles, and brochures. It's necessarily incomplete, but I think it's close enough for jazz, so to speak. I feel fine presenting this because to get closer to the truth, well, that's going to require going through a bunch of boxes marked memos and correspondence. Frank entered the computer industry at ERA, one of the very first computer companies to be established. This puts him firmly in the pioneering generation. This also put him on a strange track. In 1952, ERA was bought by, Drumroll Please,

Remington Rand. This is a point that's just odd to me. Remington Rand ends up owning two of the earliest computer companies. One day I need to do some more reading on these acquisitions, but for now, that's not important. What matters is that Frank made his way into the Univac sphere. Remington Rand rebadges ERA's machines as Univac. In fact, the machine that I think Mickelson makes core memory for is a derivative of an ERA computer.

There's an oddly direct connection here. Then, in 1963, Frank made his way over to Fabrotech. He was initially hired as staff scientist at FaberTech's new R&D lab. I think it's likely that Mickelson and Frank were acquainted from the Univac days, but that's speculation on my part. And this is where I'm going to speculate some more. I ran to a few newspaper articles that say Mickelson was a proponent of vocational training. One 1964 article paraphrases Mickelson is saying, quote, there is a tremendous need for retraining workers at the vocational school level, end quote. In a 1965 brochure from FaberTech, it brags that all factory workers

attend a NASA-approved soldering school. That sounds like vocational training to me. I don't have any interviews with Mickelson, but I do have this conjecture. He seems to have cared about education. I don't have a nice quote where he says that. This is just reading between the lines.

There is one final piece to my wild speculation. Prior to his tenure at ERA, Dr. Abe Frank was a professor. He taught math at a number of universities. So what happens when you have a company ran by a guy that seems to care about education, and he hires an educator? For one slick brochure, quote, Fabrotech has recently introduced to the educational market, the bi-trans-6 computer education system. The central unit of the system is a modern, inexpensive, general-purpose computer, which is designed exclusively for teaching purposes. end quote.

Oddly enough, the six is always spelled out. It's never the number. So, how did this work out exactly? How did a memory company make a computer? Well, let me give one last piece of speculation. I don't think it was that much of a jump. We know that Frank designed the machine. He had experience from ERA and UNIVAC days.

And ultimately, memory is about one of the most complicated things to manufacture. in this period. At this point, FaberTech wasn't just making memory planes. They were making entire memory systems. These were boxes that just stored data, plain and simple. Some of these systems came in their own giant cabinets with cooling and self-checking and buses and everything. In other words, these are on par with the complexity of a computer.

It would be a short leap to make a whole machine. Besides, they could get a pretty good, good deal on RAM. The machine they produce, Bytrans6, is very unique. It was designed specifically for education, for teaching about computers and programming and electronics. This is the crucial part to remember going forward. The BITrans 6 is announced in 1965. That's very early for computer education. Basic, the most common yardstick for this sort of thing, was first used in 1964. But that doesn't make for the best comparison here. Basic was actually intended as a tool for education, a way to use computers for teaching. The Bi-Trans6 was intended as a computer to teach

about computers. That goal has a huge impact on how the machine was designed. The first is most obvious point is that the bitran is tiny. It's a very, very simple computer and very well documented. The manual that came along with the machine goes all the way down to the circuit level. That's possible thanks to how small the machine is. It's just big enough to be useful, but just small enough to be fully comprehensible. The entire computer fits on just eight circuit boards. Those are crammed behind a tiny front pan. And, in fact, while I'm here, I do have something to say about that front panel. At first, I thought it was styled to look like the blinking lights of a mainframe.

Those panels tended to be covered in lights and switches. Enough that you really can't tell what they all do. BiTram has the same look to it, but there's a purpose behind all the lights. This front panel gives full control and visibility to the innards of the inners of the light. Bytrane, and I mean everything. Bytrane is a pretty basic register-based machine, so all registers are at your fingertips from that front panel. That includes the memory register, which is used during memory access.

It's where the computer formulates addresses to fetch. The panel also includes all the flags and flip-flops and other control bits that are buried deep inside the machine. This is more information than you usually get from a computer. It's enough that you can follow the full execution cycle of Vitran. For instance, you can watch the arithmetic logic unit working. The front panel exposes the accumulator, memory address, and exchange register. Normal machines really only show the accumulator.

During an addition, for instance, an address is put on the memory address register, and that value is read into the exchange register, then added to the accumulator. You can even watch the so-called distribution register to see which step the computer is on. Pair that with a few different level of stepping modes, and you get something really powerful. You can step the machine instruction by instruction, which is normal, but you can also step it through sub-cycles. A computer has these internal steps that are normally hidden to the user. It has to load instructions to code them, set up the logic to execute them, and store data later on. ByTran shows you all those steps.

That's an unusual level of control and visibility. This is backed up by something I've never actually seen before. The circuit boards that make up ByTran are laid out so that they match logic diagrams in the operator's manual. Transistors are grouped into logic gates, and those gates are labeled. Those boards are on rail, so you can pull them out of the machine, but still you can partly pull them out. The board will slide up while the computer is still in operation, and the boards have test points.

You can literally wire up a board and hook it up to an oscilloscope while these gates are being used, while the machine's running. And that's not meant as a diagnostic thing. It's meant as a learning opportunity. The actual instruction set of the machine is, let's just say, not very remarkable. Bytrans6 is a simple load store architecture. It's fully stored program. It basically functions like any other machine of the period.

The key difference is memory. Bytrans6 only has 128 words of memory. How big are those words? Well, as I said, they're six-bit. I don't want to dwell on that too long. This is before the whole standard 8-bit byte was established, and it's a very simple machine, so it's not like you really need to care about practical things. Six bits are enough to teach what bits are.

The simplicity helped keep the computer cheap and small. Brochers brag that the Bitrans6 weighed just under 100 pounds. Well, not while today at the time, that's... almost unimaginable. The computer could fit on a desk and be carried between classrooms. Well, maybe not carried, but at least rolled on a cart. Records are a little scant, but it looks like Bitrans6 initially sold for about $7,000. That's about $70k and current dollars. There really isn't a comparison to make here. The closest may be the deck link, which was a small lab computer that would have been used in the same period.

But that's still a solid $40,000 in 1960s money. Baitran was cheap. It's definitely within range of an educational grant, if not discretionary funding. So we have a small, simple, and cleanly designed computer. It's meant for teaching computer science in a period where computer science is just starting to be a field. Where does it show up, and how is it used? Well, that's the neat part.

we actually have some records about the next step. From yet another glossy brochure, quote, The Bytrans6 is an inexpensive computer, which has been designed exclusively for teaching computer science. It is not a cheap version of a full-scale working computer, and it is not a laboratory mock-up which merely simulates the operation of a computer.

There is nothing like it currently available to the teacher. Essentially, the Bytrans6 is a computer. is a self-motivating functional textbook, which is used to transmit knowledge of computer science to the student through sight, touch, and logical thought." In quote. 1965 is a very interesting time to release an educational computer. The previous year, in 64, the first graduate degrees in computer science are awarded, to Mary Kenneth Keller and Irving C. Tang.

I usually use that as a milestone for when computer science becomes a field. Folk were doing computer science earlier, but there wasn't anyone who could say, oh, well, I'm a doctor of computer science. I know that can sound like a stupid distinction, but it's a very important marker of acceptance. It's a marker of credibility. That doesn't, however, mean the things were well settled in this period.

This is early enough that only some folk are using the term computer science. There are still places to call it automatic data processing. And then we have cybernetics. In 1948, Norbert Viner publishes cybernetics. In it, he presents computers as tools for management. The crucial part here is how computers can be used to gather data, analyze information, and help make decisions. Then the cycle continues.

You form a feedback loop, which allows you to make better and better decisions using the power of computing. In the earliest epoch, this was really just part of the whole growing computer field. By the 1960s, things become more differentiated. Computer science becomes its own field, which is more concerned with the functioning and theory of computation.

Cybernetics is a more applied field. It's concerned with how computers can be used as tools for management and feedback systems. Classically, most people think these fields show up in different places. I mean that geographically and politically. The U.S. is firmly in the computer science camp. The USSR becomes more invested in cybernetics. As does Chile for a time when Ayinde was president. Socialist and communist governments were more important.

interest in cybernetics for a host of reasons, but part of it has to do with managing their economies. At least, that's kind of a thousand mile view. There weren't many American universities that issued degrees in cybernetics. In 1966, the U.S. Army Adjutant General School at Fort Benjamin Harrison started using the BITrans 6 as a teaching tool. They developed and offered what was called the Automatic Data Processing Orientation Course. I'd argue this was the U.S. running its own cybernetics program. You could gin up a pretty spooky story here, right? The U.S. Army, in an attempt to defeat communism, took up the very tools of their hated enemy. They forged a clandestine cybernetics training program to create super soldiers of democracy.

The truth, however, isn't nearly as white-knuckled. Cybernetics was already rampant in the U.S. It's just that it wasn't running wild in academia like computer science was. Instead, it was alive and well in the private sector, just under another name. It was called management science. Some U.S. companies had already started to use computers to automate and aid in management. The U.S. military wanted to take their lead. A 1966 paper on data processing training for Marine Corps officers

analyzed the current state of digital management and looked at how the U.S. military was adopting computers. It had this to say about the private sector. Quote, In the above-average companies, management viewed the computer very differently. It saw the computer system's effort as a major economic resource, to be used in running the business. Further, the computer effort was held no more sacrosanct than any other new corporate activity

and thus was subject to the same management processes. The report continues. It is apparent that superior results from automatic data processing require a basic change and approach. We must demonstrate a willingness to rethink problems of a business in terms of goals, not in the streamlining of existing procedures. Automation requires managerial imagination rather than technical proficiency. Imaginative management in the new world of computers has paid off handsomely, end quote. To reap the true benefit of computers, those machines were integrated into the very management of companies.

They became another cog. Some in the U.S. military wanted to do the same. This would be a huge undertaking. It required computers, but it also required trained operators, and it required that those in decision-making positions knew how to leverage and manage computers. The word isn't used, but this is classic cybernetics. In the model proposed, machines are used as tools to aid in management of an organization, to turn data into decisions and then collect data on those decisions.

This is where By-Tran 6 comes into play. The Armed Forces had a number of training programs, one of which used Fabritech hardware. This was a ground-up kind of course. The U.S. Army developed their own lesson plans and manuals and everything. Personnel would come to Fort Benjamin Harrison for these classes, but that was only one part of the program. By-Tran, you see, could travel.

An article in the Fabritac Inquirer tells all. And yes, that was the name of FaberTech's internal newsletter. The army would take a bi-transix to Germany, Alaska, Okinawa, and even Vietnam. The machine was shipped out in the field to help train personnel on these newfangled computer things. FaberTech called it the quarter million-mile computer. This is a trick that you could clearly not do with a traditional machine, but the bi-tran 6 weighed 98 pounds. That was easy to ship around.

It's also worth noting that this machine couldn't have been used for serious work. That's something that's implied but worth pointing out. As an educational computer, the bi-trans isn't powerful at all. folk would learn the basics on the machine and then transition to some other computer for actual work. One aspect of the bi-trans-6 that makes it so interesting is that it was used in a wide variety of settings. That's kind of the whole point, right? And that's what's so compelling about the quarter-million-mile machine. It's a tool for teaching about computing.

In 65, almost no one knew a darn thing about computing. so there was a very large and very diverse audience for the machine. FaberTech provided their own lesson plans for teaching computer science in schools, and schools created their own programs around Bitran. That led to a wide range of applications and influence. And this is where we get to something a bit odd. So, the most well-known photo of a Bitran, and the one you'll see if you look up By-Tran 6, depicts a nun standing next to the computer, pointing to the machine's circuit boards.

In fact, that's the first photo I ever saw of a Bytrans 6 that wasn't an artistic rendering in an ad. You'll sometimes see the photo circulated with the caption, Did you know, a nun invented basic? Now, of course, that caption is very far, far off. That nun is the earlier mentioned sister Mary Kenneth Keller, one of the first computer scientists with papers to prove it. But she didn't invent basic, even though she was involved in the early years of the language. And the bi-trans-6 didn't run basic. So the photo is almost a complete non-sequitur. So then what's the truth of the matter? What is sister

Kenneth Keller doing with a Baitrans 6? Why? She's teaching. The year after graduation, Sister Kenneth started teaching at Clark College in Iowa. This is why papers are important. It lends a certain legitimacy that lets you do things like teach at colleges. Clark had just established their computer department and Sister Kenneth was brought in to direct it because she had a Ph.D. in computer science. Papers are important. At first, this wasn't much of a department. She had to grow it from basically nothing. The first machine that she had access to at that college was a bi-trans-6. So this popular photo is actually Sister Kenneth with her entire computer science department's budget. The following year, the department was able to get an IBM 1130, giving them an actual computer to work.

work with. That means there was a year where Sister Kenneth was teaching with a Bytran 6 alone. We don't have a lot of information on the exact lessons, but we do know that bytran was used to teach college-level courses and, perhaps, vocational classes. But that's about as far as the paper trail goes for this one. By 1966, there was a real computer on campus. So it's likely the bi-trans-6 had a little less emphasis placed on it. We can also say with certainty that bi-trans-6s were at least used in high schools. But the details are scant. I can find minutes from conferences and some preceding articles that mention discussing how to teach high school courses with the machine. But that's it.

It's more like so-and-so gave a lecture about bi-trans-6s at this conference, but no details. Perhaps I shouldn't be very surprised that sources on high school curriculums in the 60s are a little unavailable, let's say. This is also where I want to take a bit of a tangent to point out something that struck me as strange. The Faber Tech Inquirer. We have three digitized issues. One of them carries a story about one Dr. Donald L. Henderson of Mancato State College. It discusses a certain experiment he was carrying out. Henderson was developing a curriculum to teach computing to high school students, or rather to use computers in high school classes. The college was planning to give a number of local high schools access to computers and this new curriculum. Henderson wanted to test if computers

were useful in high school math classes, specifically if they could be used to teach math and computing in conjunction. The pilot program was to include six schools and start in 1968. I went out searching for the results of that pilot program and found, I found nothing related to By-Tran. Fool that I was, I hadn't read the article closely enough. I know. Me, having issues with reading comprehension. It could never happen. Here's what I missed and what I ran face-first into. Henderson's article isn't about doling out by-trans. It's about terminals. He was actually following the Dartmouth approach. He wrote a grant to establish a computing center and then placed terminals in local high schools. Henderson's secret

curriculum was for terminal-based education. This story actually has more to do with the Oregon Trail than Fabrotech. Text-based educational games like the Oregon Trail or the Sumerian game develop from these kinds of regional computer centers and these kinds of experimental curriculums. the tiny bi-trans-6 does not enter into the picture. Anyway, here's what's so interesting. Someone at FaberTech thought this experiment was exciting enough to put it in their corporate newsletter. The Inquirer isn't just for educational content, and it's not just for bi-trans stuff.

It was targeted at the whole company. I think this speaks to a crucial investment in education within Fabri-Tech. At the same time, this points out a disconnect between the Baitran approach to education and what some educators want. Folk like Henderson were interested in integrating computers into classrooms, in using computers as tools for teaching. That doesn't mean teaching classes about computers. And I think that's understandable, right?

There simply wasn't a framework for teaching about computers in primary or secondary schools. This actually echoes something in the Marine Corps analysis of computer adoption in corporations. Hear me out. That report claims that average companies simply try to use computers to supercharge existing business practices. Above average companies changed business practices, foundationally, to make better use of computers. Many schools are operating as average companies. They want to use computers to supercharge traditional teaching frameworks. Henderson wanted to teach math using a computer.

The creators of the Oregon Trail wanted to teach about, well, a very long hike, using a computer. Getting familiar with a machine was a side effect. Henderson even gives a list of four major objectives for his experiment. Learning about computation is number four. It's nice, but very visibly non-central. BiTrans6 is specifically a machine meant for teaching about computers. That actually hamstrung it. Educators in this period are starting to use software in classrooms.

Bytrans6's limited memory means it can't really run software. That restricted the kinds of things it could be used for in a classroom setting. We're going to close out with something even more tangential that I ran into while reading about by-tran. Well, by-tran everything. Not a lot of sources, so I've hit pretty much all of them at this point. Occasionally, I'll run into a thesis or two. You know, it's what folk have to write about to get a graduate degree. These are, in my opinion, wild documents to read. For many, a thesis is their first published academic work. So there's a rawness to them that I like. In 1969, Guy Evan Rudyick submitted his thesis for his Masters of Education. It's titled, A Computer Technology Game,

time-sharing. And I really like this thesis. Rudjik designed a board game for teaching how computers work to primary school students. The thesis describes all the revisions he went through, all the play testing and how the game actually worked out in the end. It turns out the kids liked it after some revisions. At the start of each game, all players are given an encoding board, a small instruction manual, and a set of wooden blocks for encoding binary. These blocks were a huge stumbling point for the development of this game. The game uses 4-bit numbers, so each block simply has 4 holes.

Initially, Rojjjik used painted golf t's to encode numbers. That was difficult to manage, so he would eventually find these smaller pegs at a toy store that worked better. As the game starts, a program is revealed, like X equals 1 plus 2. Each player is given a pack of instruction cards, and encoding begins. Players race to arrange their cards to form a program that solves the problem given, and then encode that program in binary using encoding blocks and golf t's. This is a race because once your program is prepared and coded, you can start entering it into core memory. This phase takes place in turn order, but if your program isn't encoded, you have to skip your turn.

Once your program is loaded into memory, you move to execution. Players take on the role of the computer, placing down an instruction pointing arrow, and then chunking through instructions. The game part here comes in the form of errors. If a player makes a mistake in execution, any other player can scream error and get an extra turn. So it pays to understand how the computer should operate and to pay attention. This game, which Regic calls time sharing, hits all the highlights.

Programming and coding are two separate tasks, as they would have been in this period. You get to load in a program like you would with a stack of punch cards, and then you take on the guise of the computer itself to make the program run. It proved useful enough as a project for Ruggick to get his degree. So, what's the connection? Why are we talking about this game? When Rujik laid out his design, he wanted to use an existing computer as a basis for the game. Something simple but realistic.

His choice? The Bytrans6. All instructions are based roughly on Fabrice's own machine. This is the type of diffuse influence that I've found in the sources. Educators were inspired or influenced by the BITrans 6. The trail for actual classes and teaching details is thin, I think in part just because of preservation. But we have these touchpoints of the machine's influence.

All right, that does it for a look at the BITAN 6 computer. FaberTech made quite the jump from magnetic core memory to educational computers. The BITRAN would only be the first of their educational machines. In the coming years, FaberTech would release the CompTran, another trainer. But we can see enough from BITAN 6 to establish some patterns. These early trainer computers are unique ideologically. machines aren't normally built for clarity of their circuit boards, but the bi-tran was. It was built from the ground up to teach, so much so that it wasn't very flexible.

The machine only really works in a few types of lesson plans. It can only be used as a machine for teaching computing, not as a computer augmentation for other types of lessons. For all this, I think there's still a mystery here. That bridge between core memory and education still bugs me. If anyone wants to pick up that thread, hit me up, and I can send you details on the Abe Frank collection at CBI. There's a bigger story to be told in that transition, one that I think could tell us something

fascinating about views on computer education in this earliest period. Thanks for listening to Avent of Computing. I'll be back in two weeks with something. We'll see, because I'm on the road yet again. Until then, you can find links to the Patreon. You can find links to everything at advent ofcomputing.com. And as always, have a great rest of your day.