Chapter from “The Engelbart Hypothesis: Dialogs with Douglas Engelbart” by Valerie Landau and Eileen Clegg in conversation with Douglas C.Engelbart.

I believe that the complexity of the problems facing mankind is growing faster than our ability to solve them. Finding ways to augment our intellect is both a necessary and a desirable goal. The mouse was just a tiny piece of a much larger project aimed at augmenting human intellect.

It was 1964 and I was working at SRI. I envisioned problem-solvers using computer-aided workstations to augment their efforts. They required the ability to interact with information displays using some sort of device to move around the screen. We were looking for the best and most effective device to point to and select the information displayed.

By the time I invented the mouse, I had already spent a dozen years exploring ways for people to increase their capability to solve complex problems. In the early 1960s, several devices were in use: the light pen, joysticks, and others. We analyzed the various characteristics of other pointing/input devices before the invention of the mouse. We made a grid, similar to the Periodic Table of the Elements. We laid out the grid in rows and columns with the characteristics that define each group of devices. And just as the periodic table’s rules have led to the discovery of certain previously unknown elements, this grid ultimately defined the desirable characteristics of a device that didn’t yet exist.2 That device was the mouse.

We approached NASA in the early 1960s3 and said, “Let’s do a study to determine, once and for all, what is the most effective selecting and pointing device.” With NASA funding, we developed a set of simple tasks and timed a group of volunteers to complete the tasks with the various devices. For example, the computer generated an object in a random position on the screen and a cursor somewhere else. We timed how long it took each user to move the cursor to the object. It quickly became clear that the mouse outperformed all the others.

The light pen required the user to pick up the pointer and reach across the screen. After several tests, it became evident that it was very tiresome. We also developed a knee-based pointing device.4

A fellow named Bill English built the world’s first mouse. Bill was an extremely effective guy. I had sketched out the idea fairly quickly in a little notebook and I gave it to Bill to build. He went home and carved a piece of mahogany and built the world’s first mouse.5 I couldn’t have done it without Bill. But the patent attorney at SRI did not agree that Bill English should share the patent. The original mouse had the cord in front, but we quickly moved it to the back end to get it out of the way. It was a simple mechanical device with two perpendicularly mounted discs on the bottom. You could tilt or rock the mouse to draw perfectly straight horizontal or vertical lines. Or you could give the mouse a push and lift it off the desk, and watch the cursor continue moving while the disc was spinning.

No one can remember who coined the term, “the mouse.” It just looked like a mouse with a tail, and we all called it that in the lab. I thought it would be called a “servo-control unit” or something like that. But the name “mouse” just took.

We also developed a chording keyset to work in an orchestrated way with the mouse and keyboard to get maximum flexibility and efficiency. The keyset allows users to type the entire alphabet and numbers, as well as key commands, in conjunction with the three-button mouse. The concurrent use of mouse and keyset also provides considerable gains in speed and flexibility for modifying document structure. For example, if the author perceives that Statement 2b really belongs in Section 3, following Statement 3c, she/he can execute the necessary move command in a very quick, deft manner. In order to move an entire section, which I called a “Branch,” using the keyset, the user strikes “m” and then “b” for “Move Branch.”

Meanwhile, the mouse hand is positioning the cursor anywhere in the text line of Statement 2b. So in order to move an entire section or branch, it only takes two chord strokes.


The chording system of typing is similar to playing notes and chords on an instrument. Each finger is assigned a number, so when you press a combination of fingers simultaneously, the numbers are added together. Letters are also given numbers, so, for example, 1=A and 2=B, so pressing those simultaneously results in a “C.”

_______________________________________________________________________Professor Bill Buxton conducted and researched a serious study of text input devices in 2002. He concluded his study with the following: “Based on the literature and personal experience, I believe that chording keyboards have an important role to play in human-computer interaction … Major improvements in methods for keyed input will only be achieved through a radical change from current practice.
This is true for both novices and experts. The use of chord keyboards as an alternative means for keyed input is still underdeveloped. This is, I feel, due to the range and complexity of the issues affecting their performance. To change this situation, research must investigate appropriate applications as much as technical issues. In my mind, the issue is not if chord keyboards can be effective, but where and how?”
2 Bill English recalls, “The mouse development was one of the very early engineering projects. I don’t remember anything about the ‘grid’ that defined the characteristics of pointing devices. We simply looked around at all of the display pointing devices that were in use, and conducted the experiment.”

3 “It was 1962 when we proposed the NASA experiments, and the final report was issued in July, 1963.” Bill English recalls.

4 According to Bill English, “The foot has very poor fine control, and the foot control was the poorest of any we tried.

5  Bill English commented, “I did not carve the box. I did the engineering work necessary to take Doug’s sketch to a working device that held the orthogonal wheels and the potentiometers to transmit position to a computer. After the device was built by the SRI machine shop, a box to hold it was carved, by a draftsman who worked near our project. The original mouse had the cord in front and only one button, but we quickly moved the cord to the back to get it out of the way, and realized that there was space for more buttons that could greatly increase the functionality of the mouse.”


2 Responses to “Engelbart on The Mouse and Keyset”

  1. Rob Stephenson created an adapter that works with an IC board we programmed to make the original keysets work with a modern laptop via USB

  2. Lesley Says:

    An Englebart mouse, keyset and line-processor is up for auction on eBay under the title: “Englebart Mouse; *Very rare* Engineering Prototype”


    Question: is Rob’s adapter design available anywhere?

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: