Touch-Typing, the Underrated Superpower

I recently came across the article 10-Finger-Schreiben: Die völlig unterschätzte Superkraft (“Touch-typing: the totally underrated superpower”; if you don’t read German, you can go directly to the piece that inspired it, The best life hack for 2018 that isn’t on any life hack list). I fully agree with the author when he notes that, given the comparatively small effort you have to put into learning it and the huge positive effect on your productivity, it is probably one of the best investments into personal development you can make since the beginning of the Information Age.

I’d just like to comment on the last part of the article, where the author wonders how long this statement will remain valid. Writing on smartphones and tablets, the ever growing importance of Emoji in daily communication, and the trend toward voice control are, according to the author, some of the changes that endanger the advantages of touch-typing. He concludes somewhate fatalistically: “The dynamics of these developments are hard to estimate, and it is as yet open when keyboards will also retreat from the working world. At some point it will happen.”

I think the reports of the death of the keyboard have been greatly exaggerated. Since decades, the keyboard has been declared dead, and yet it’s alive and well. For example, in the 1990s, “pen computing” was all the rage, and many people were sure that the “natural” pen input would shortly replace “unnatural” keyboard input. In a 1991 article in Byte Magazine, the author states:

In the more distant future, desktop and portable computers may be able to recognize speech, which is the most intuitive interaction of all. Alternatives to keyboards need a lot more horsepower—in the case of speech recognition, perhaps hundreds of millions of instructions per second. Current research in digital signal processors and neural networks may reduce the raw processing power required by today’s brute-force pattern-matching voice-recognition schemes, but, even so, it will be some time—perhaps five to 10 years—before it will be practical to talk to your laptop.

Twenty-seven years later, you can sure talk to your laptop, as well as to your tablet (a pen computer without a pen and without handwriting recognition), and if you have a smartphone, you can not only talk into your phone but also to your phone. But is it practical? If you think about it, neither do hipsters working in coffee shops dictate the business plan for their next startup into their MacBooks nor do professors collaborating on a research proposal dictate it into Google Docs. Voice dictation can be very useful for some people, but it remains a niche application.

You also need to differentiate: tapping a short message to your spouse that your train is delayed is not the same as writing a novel, and instructing a “voice-controlled intelligent personal assistant” to dim the lights is not the same as writing a business report. In other words, there’s a big difference between short messages, device control, and long-form texts. And just as novels still haven’t been replaced by movies, I’m pretty sure other long-form texts—whether project reports, design specifications, or even office memos—won’t go away, and so you need a way to create these texts. You can type in public, but you can’t dictate in public. What’s more, “text input” is only one part of writing: editing and revising using voice control is difficult and hardly efficient. Typing also allows you to type exactly what you want: mathematical or chemical formulas, neologisms, mixed languages, shell commands, code.¹ Keyboards—and thus touch-typing—will remain the best option for entering text into computer.

This is not to say that the push to “voice” will stop. Big Tech loves voice input, because like touchscreen keyboards, they give them full monitoring and full control: what you write is no longer solely determined by the keys you press. Predictive text encourages conformance with training data through convenience. Voice input is only convenient as long as you use words the system understands; you can only write what the system is able to write. Of course the voice recognition system can also be designed not to recognize certain words or expressions; what’s more, it’s technically easy to report users for trying to dictate these words—you don’t get much closer to thought control, every totalitarian’s dream.²

Touch-typing is a totally underrated superpower for sure.

Author: Michael Piotrowski

Computational linguist, computer scientist, professor of digital humanities at the University of Lausanne. View all posts by Michael Piotrowski