In a world of rapidly changing technologies, few have lasted as long is as unaltered a fashion as the mouse. The party line is that the computer mouse was invente d by Douglas Engelbart in 1964 and that it was a one-button wooden device that had two metal wheels. Those used an analog to digital conversion to input a location to a computer. But there’s a lot more to tell. Englebart had read an article in 1945 called “As We May Think” by Vannevar Bush. He was in the Philippines working as a radio and radar tech. He’d return home,. Get his degree in electrical engineering, then go to Berkeley and get first his masters and then a PhD. Still in electrical engineering. At the time there were a lot of military grants in computing floating around and a Navy grant saw him work on a computer called CALDIC, short for the California Digital Computer. By the time he completed his PhD he was ready to start a computer storage company but ended up at the Stanford Research Institute in 1957. He published a paper in 1962 called Augmenting Human Intellect: A Conceptual Framework. That paper would guide the next decade of his life and help shape nearly everything in computing that came after. Keeping with the theme of “As We May Think” Englebart was all about supplementing what humans could do. The world of computer science had been interested in selecting things on a computer graphically for some time. And Englebart would have a number of devices that he wanted to test in order to find the best possible device for humans to augment their capabilities using a computer. He knew he wanted a graphical system and wanted to be deliberate about every aspect in a very academic fashion. And a key aspect was how people that used the system would interact with it. The keyboard was already a mainstay but he wanted people pointing at things on a screen. While Englebart would invent the mouse, pointing devices certainly weren’t new. Pilots had been using the joystick for some time, but an electrical joystick had been developed at the US Naval Research Laboratory in 1926, with the concept of unmanned aircraft in mind. The Germans would end up building one in 1944 as well. But it was Alan Kotok who brought the joystick to the computer game in the early 1960s to play spacewar on minicomputers. And Ralph Baer brought it into homes in 1967 for an early video game system, the Magnavox Odyssey. Another input device that had come along was the trackball. Ralph Benjamin of the British Royal Navy’s Scientific Service invented the trackball, or ball tracker for radar plotting on the Comprehensive Display System, or CDS. The computers were analog at the time but they could still use the X-Y coordinates from the trackball, which they patented in 1947. Tom Cranston, Fred Longstaff and Kenyon Taylor had seen the CDS trackball and used that as the primary input for DATAR, a radar-driven battlefield visualization computer. The trackball stayed in radar systems into the 60s, when Orbit Instrument Corporation made the X-Y Ball Tracker and then Telefunken turned it upside down to control the TR 440, making an early mouse type of device. The last of the options Englebart decided against was the light pen. Light guns had shown up in the 1930s when engineers realized that a vacuum tube was light-sensitive. You could shoot a beam of light at a tube and it could react. Robert Everett worked with Jay Forrester to develop the light pen, which would allow people to interact with a CRT using light sensing to cause an interrupt on a computer. This would move to the SAGE computer system from there and eek into the IBM mainframes in the 60s. While the technology used to track the coordinates is not even remotely similar, think of this as conceptually similar to the styluses used with tablets and on Wacom tablets today. Paul Morris Fitts had built a model in 1954, now known as Fitts’s Law, to predict the time that’s required to move things on a screen. He defined the target area as a function of the ratio between the distance to the target and the width of the target. If you listen to enough episodes of this podcast, you’ll hear a few names repeatedly. One of those is Claude Shannon. He brought a lot of the math to computing in the 40s and 50s and helped with the Shannon-Hartley Theorum, which defined information transmission rates over a given medium. So these were the main options at Englebart’s disposal to test when he started ARC. But in looking at them, he had another idea. He’d sketched out the mouse in 1961 while sitting in a conference session about computer graphics. Once he had funding he brought in Bill English to build a prototype I n 1963. The first model used two perpendicular wheels attached to potentiometers that tracked movement. It had one button to select things on a screen. It tracked x,y coordinates as had previous devices. NASA funded a study to really dig in and decide which was the best device. He, Bill English, and an extremely talented team, spent two years researching the question, publishing a report in 1965. They really had the blinders off, too. They looked at the DEC Grafacon, joysticks, light pens and even what amounts to a mouse that was knee operated. Two years of what we’d call UX research or User Research today. Few organizations would dedicate that much time to study something. But the result would be patenting the mouse in 1967, an innovation that would last for over 50 years. I’ve heard Engelbart criticized for taking so long to build the oNline System, or NLS, which he showcased at the Mother of All Demos. But it’s worth thinking of his research as academic in nature. It was government funded. And it changed the world. His paper on Computer-Aided Display Controls was seminal. Vietnam caused a lot of those government funded contracts to dry up. From there, Bill English and a number of others from Stanford Research Institute which ARC was a part of, moved to Xerox PARC. English and Jack Hawley iterated and improved the technology of the mouse, ditching the analog to digital converters and over the next few years we’d see some of the most substantial advancements in computing. By 1981, Xerox had shipped the Alto and the Star. But while Xerox would be profitable with their basic research, they would miss something that a candle-clad hippy wouldn’t. In 1979, Xerox let Steve Jobs make three trips to PARC in exchange for the opportunity to buy 100,000 shares of Apple stock pre-IPO. The mouse by then had evolved to a three button mouse that cost $300. It didn’t roll well and had to be used on pretty specific surfaces. Jobs would call Dean Hovey, a co-founder of IDEO and demand they design one that would work on anything including quote “blue jeans.” Oh, and he wanted it to cost $15. And he wanted it to have just one button, which would be an Apple hallmark for the next 30ish years. Hovey-Kelley would move to optical encoder wheels, freeing the tracking ball to move however it needed to and then use injection molded frames. And thus make the mouse affordable. It’s amazing what can happen when you combine all that user research and academic rigor from Englebarts team and engineering advancements documented at Xerox PARC with world-class industrial design. You see this trend played out over and over with the innovations in computing that are built to last. The mouse would ship with the LISA and then with the 1984 Mac. Logitech had shipped a mouse in 1982 for $300. After leaving Xerox, Jack Howley founded a company to sell a mouse for $400 the same year. Microsoft released a mouse for $200 in 1983. But Apple changed the world when Steve Jobs demanded the mouse ship with all Macs. The IBM PC would ;use a mouse and from there it would become ubiquitous in personal computing. Desktops would ship with a mouse. Laptops would have a funny little button that could be used as a mouse when the actual mouse was unavailable. The mouse would ship with extra buttons that could be mapped to additional workflows or macros. And even servers were then outfitted with switches that allowed using a device that switched the keyboard, video, and mouse between them during the rise of large server farms to run the upcoming dot com revolution. Trays would be put into most racks with a single u, or unit of the rack being used to see what you’re working on; especially after Windows or windowing servers started to ship. As various technologies matured, other innovations came along to input devices. The mouse would go optical in 1980 and ship with early Xerox Star computers but what we think of as an optical mouse wouldn’t really ship until 1999 when Microsoft released the IntelliMouse. Some of that tech came to them via Hewlett-Packard through the HP acquisition of DEC and some of those same Digital Research Institute engineers had been brought in from the original mainstreamer of the mouse, PARC when Bob Taylor started DRI. The LED sensor on the muse stuck around. And thus ended the era of the mouse pad, once a hallmark of many a marketing give-away. Finger tracking devices came along in 1969 but were far too expensive to produce at the time. As capacitive sensitive pads, or trackpads came down in price and the technology matured those began to replace the previous mouse-types of devices. The 1982 Apollo computers were the first to ship with a touchpad but it wasn’t until Synaptics launched the TouchPad in 1992 that they began to become common, showing up in 1995 on Apple laptops and then becoming ubiquitous over the coming years. In fact, the IBM Thinkpad and many others shipped laptops with little red nubs in the keyboard for people that didn’t want to use the TouchPad for awhile as well. Some advancements in the mouse didn’t work out. Apple released the hockey puck shaped mouse in 1998, when they released the iMac. It was USB, which replaced the ADB interface. USB lasted. The shape of the mouse didn’t. Apple would go to the monolithic surface mouse in 2000, go wireless in 2003 and then release the Mighty Mouse in 2005. The Mighty Mouse would have a capacitive touch sensor and since people wanted to hear a click would produce that with a little speaker. This also signified the beginning of bluetooth as a means of connecting a mouse. Laptops began to replace desktops for many, and so the mouse itself isn’t as dominant today. And with mobile and tablet computing, resistive touchscreens rose to replace many uses for the mouse. But even today, when I edit these podcasts, I often switch over to a mouse simply because other means of dragging around timelines simply aren’t as graceful. And using a pen, as Englebart’s research from the 60s indicated, simply gets fatiguing. Whether it’s always obvious, we have an underlying story we’re often trying to tell with each of these episodes. We obviously love unbridled innovation and a relentless drive towards a technologically utopian multiverse. But taking a step back during that process and researching what people want means less work and faster adoption. Doug Englebart was a lot of things but one net-new point we’d like to make is that he was possibly the most innovative in harnessing user research to make sure that his innovations would last for decades to come. Today, we’d love to research every button and heat map and track eyeballs. But remembering, as he did, that our job is to augment human intellect, is best done when we make our advances useful, helps to keep us and the forks that occur in technology from us, from having to backtrack decades of work in order to take the next jump forward. We believe in the reach of your innovations. So next time you’re working on a project. Save yourself time, save your code a little cyclomatic complexity, , and save users frustration from having to relearn a whole new thing. And research what you’re going to do first. Because you never know. Something you engineer might end up being touched by nearly every human on the planet the way the mouse has. Thank you Englebart. And thank you to NASA and Bob Roberts from ARPA for funding such important research. And thank you to Xerox PARC, for carrying the torch. And to Steve Jobs for making the mouse accessible to every day humans. As with many an advance in computing, there are a lot of people that deserve a little bit of the credit. And thank you listeners, for joining us for another episode of the history of computing podcast. We’re so lucky to have you. Now stop consuming content and go change the world.
The computer mouse is a ubiquitous device, it's also one of the least changed devices we use with a computer. The mice we use today have only seen small incremental improvements since the first mouse was developed. So how did such a long lasting design take shape, and how did it travel the decades up to now?
Like the show? Then why not head over and support me on Patreon. Perks include early access to future episodes, and stickers: https://www.patreon.com/adventofcomputing
Important dates in this episode:
1961: First Mouse Developed at Engelbart's ARC Lab
1972: Xerox Develops Rollerball Mouse for Alto
1979: Apple LISA Mouse Designed