Pascalines, Arithmometers & Comptometers

You might recall in an episode a few weeks back, our guest was Jim Warrick of Beacon Technology Partners. Jim is the impresario behind our latest biennial survey of engineers around the world, which we call “The Mind of the Engineer.” The results from the survey tell us a lot about the electronics industry, about engineering as a profession and about engineers themselves.

Here’s my favorite bit from that episode:

JIM WARRICK: We actually did focus groups, which led up to our first questionnaire. It was one of the Boston sessions. We were asking a variety of engineers about what really motivated them and so on, and I was trying to get these engineers to articulate what it felt like if they had been working on a very vexing problem for days or weeks, and then they got the answer. What does that feel like? And of course, of engineers tend to be rather prosaic, so they were saying “It felt great; it’s wonderful.” But one of the individuals happened to be a woman. She was from Russia. She actually had been trained as a physicist and was actually making the graduation over to power electronics. She got so excited trying to answer this question that she started shaking and just uttering things in Russian. And we said, “No, no, no. Please. Please. English. English. We want to understand.” And she held up her hands like this. And she said, “What I feel like? I feel like little god!”

BRIAN SANTO: The reason we’re revisiting that earlier episode is because EE Times is now offering access to the full Mind of the Engineer survey report, complete with data and insights. The report is an invaluable source of information about engineers, the engineering profession, with plenty of gold nuggets to mine about the global high-tech industry. The web address is rather long, so I’m not going to repeat it here, but if you go to the web page of this podcast episode, at eetimes.com/podcasts, we’ll have a handy link waiting for you, along with a link to that earlier podcast, titled “I Feel Like Little God.”

We really do have to get t-shirts made with that…

And now the moment you don’t have to wait for because it already happened. Just about every week, we like to celebrate the anniversaries of watershed events in technology history. And if we can’t find a great event, we’ll settle for a diverting curiosity – which is what we’re doing this time around.  Today we are going to set our Wayback machine to…

October 11, 1887. That’s the day that Dorr Eugene Felt applied for a patent for an improvement on a machine he had first filed a patent for just a few months earlier. This thing was something I had never heard of before today, even though the device was supposedly used well into the 1990s. In the patent, Felt referred to it as an “adding machine,” but it quickly became known as a “comptometer.” Felt’s comptometer is considered the first commercially successful, key-driven, mechanical calculator.

That distinction includes several qualifications, and that’s because it was hardly the first mechanical adding machine. You could argue that that was the abacus, which goes back thousands of years, although they’re now carefully referred to as “calculating tools.” The modern ancestor of Felt’s comptometer — and imagine air quotes around the word “modern” — was the Pascaline, a mechanical adding machine that Blaise Pascal began building in the 1640s. Though there were some contemporary designs, Pascal’s adding machine was the only one from the 17th century known to have worked properly. Of the 20 or so Pascalines that Pascal built, nine are known to still exist. Most are held by European museums, and IBM owns one.

Gottfried Liebniz, the co-inventor of the calculus, and the namesake of a rather delicious cookie, expanded upon Pascal’s idea. He built two calculating machines – one in 1694, the other in 1706. Both were considered lost, until 200 years later the first of the two was rediscovered. It had broken and was sent out for repairs that apparently were never performed. Two hundred years later, in 1880, workmen clearing out a university attic found it and sent it back to Hanover, where Liebniz had lived for a while. That machine, called the Stafflewalze, or loosely translated into English as the step reckoner, is considered to be the first functional design to perform all four mathematical functions.

By the 1850s, at least three European inventors had created key-driven adding machines. At that point, the devices were being referred to as “arithmometers.” In the United States, Thomas Hill heard about the European machines and decided to build one of his own, which he completed in 1857. It is considered an impractical design; by one account, when one presses a key, it sends the calculating wheel spinning and it keeps spinning until it stops from inertia. Hill, by the way, would later service as president of Harvard University.

Other American inventors tried building arithmometers, all with limited success, until Dorr Felt came along. Felt was working in a machine shop in Chicago, and was inspired by the works of a planer – a machine used for shaping various materials. He decided to try to build one, but he couldn’t afford metal parts, so he built his first adding machine model using elastic bands, meat skewers, string, staples and a macaroni box. By the way, that macaroni box adder is in the collection of the Smithsonian.

Hey, did somebody mention venture capital? Felt traded a future interest in his invention for a workbench where he could work, but later he borrowed $800 from a cousin. He bought back the original interest and used the leftover money to buy parts. Later, he would be introduced to a machine shop owner named Robert Tarrant, who would eventually invest roughly $5,000 of his own money in Felt’s endeavor. Felt used that money to pay back his cousin, perfect his machine, and start filing patents in 1887. That first patent was filed in July, and the next on October 11th. In November, the Felt & Tarrant Mfg. Company was established. By the end of the year, Felt & Tarrant had built eight machines. Priced at $400 each, the first four went to the U.S. Treasury Department, and the remaining four went to local businessmen.

As a side note, Felt’s comptometers were just complicated enough to require skilled operators. Felt & Tarrant did the training themselves. The average accountant at the time was capable of adding long columns of four-digit numbers. Felt’s comptometer could far exceed that, and when used by a skilled operator, could add much faster than people could in their heads. By one contemporary assessment, it could multiply eight to ten times faster than anyone using pencil and paper.

This is purportedly the sound of someone operating one of the early models of the Felt comptometer:

SOUND EFFECTS: Comptometer

BRIAN SANTO: Felt died in 1930 at the age of 68. His company, Felt & Tarrant Manufacturing, started making electric models in the 1930s. The company went public in 1947, and in 1957 it changed its name to the Comptometer Corporation. Four years after that, it merged with the Victor Adding Company.

The last comptometers were made in the 1960s by a British company that had purchased manufacturing rights. The Victor Adding Company still exists, though it’s now known as Victor Technologies. The company makes digital calculators, which are echoes of the old comptometers that are themselves nearly obsolete themselves now.

That’s it for the Weekly Briefing for the week ending October 16th. Thank you for listening. The Weekly Briefing is available on iTunes, Android, Stitcher and Spotify, but if you go to us via our web site at www.eetimes.com/podcasts, you’ll find a transcript along with links to the stories we mentioned.

This podcast is Produced by AspenCore Studio. It was Engineered by Taylor Marvin and Greg McRae at Coupe Studios. The Segment Producer was Kaitie Huss.

I’m Brian Santo. See you next week.