Contact Us

Ditch The Keyboard?

By Maarten van Swaay

Not long ago I wrote about a trend to remove longhand writing from the curriculum in grade schools, and to replace it with ‘keyboarding’. Change now appears to come faster than I can follow: on March 14 I came across an article under the title “Smartphones are reinventing – and ditching – the keyboard“.

Granted, phones were initially intended for speech; in their antique days they came with rotary dials, and later with a small set of numerical keys. As ambitions grew, the phones had to become smaller, yet they had to provide space for a growing set of keys. Then the phones became so ambitious that they gave rise to the joke: “you mean it can make phone calls too?” Today’s phones would put a computer of even a decade ago to shame, with a built-in camera, storage for address lists and pictures, games to keep their owners entertained, and access to the internet, to video, and to email. But we have not yet grown ‘micro-fingers’ with which to manipulate either real or simulated keyboards.


That drives the latest developments: attempts to shrink the traditional keyboard with its 50-odd keys into something with as few as four ‘keys’. Those of course should no longer be called keys; one might call them ‘joy-pads’ as a takeoff from ‘joystick’. But how could one produce ‘written’ text with so few things to touch?

It appears that what we write is rarely as unpredictable as it seems, and some inspection confirms that written text is indeed quite redundant. How else would it be possible to detect typo’s? Yet there is a big difference between detecting a typo, and predicting what the next letter, let alone the next word, should be. But prediction is precisely what these ‘smart keyboards’ attempt to rely on for what they attempt to do.

In the days when ‘the computer’ on a campus filled a large locked room and was surrounded by a priesthood of people tending to it, students complained about being forced to use a ‘student compiler’ for their work, instead of the ‘production compiler’ that the insiders could use. As you may know, people speak a language very different from what tells computers what to do. So programmers write in a ‘source language’ that serves human needs, and an associated compiler then translates that source language into the binary form that computers can handle. Compilers also do much proofreading; they verify that the source text abides by all the rules of the programming language, and they are quite unforgiving, for good reason.

The difference between the two types of compiler does not lie in their quality; it reflects differences in what they were designed to do. A student compiler, more properly called a development compiler, will not dismiss a submitted program on detecting the first error; it will attempt to do a ‘standard fixup’ so that it can continue to read the rest of the code, and return a list of errors, rather than just the first one.

For a production compiler it would be anathema to attempt any ‘fixup’: that would give the compiler authority to guess at the thoughts of the programmer. A production compiler differs from a development compiler in other ways as well, but those need not concern us here. Almost all development compilers are much faster than production compilers, to the great advantage of harried students on a deadline.

The idea that a piece of software can anticipate what we want to type should give us pause: who does the thinking? Granted: what we wish to type may often be predictable, but only when we are typing something prosaic. A deep and informative thought must by its very nature be unexpected and therefore unpredictable. That makes the idea of ‘anticipatory keyboards’ worrisome, if not outright silly.

Many years ago I encountered a simple example of what this anticipation can lead to. I was writing a presentation about ‘progress’, which one hopes can be used to advantage, but which can also be subverted into mischief. Plato’s Phaedros contains a story in which god of invention Theuth introduces king Thamus to writing. Thamus protests, arguing that the written word may replace memory with recollection and thereby give people a false sense of knowledge. I wondered whether Encarta – the digital dictionary available from Microsoft – would have an entry for ‘Thamus’. That proved to be revealing: Encarta insisted that I should have asked for ‘Thames’ and proceeded to display a story about a river through London. Encarta refused to admit that it had no entry for ‘Thamus’. Google knows better, and does return a large number of links to ‘Thamus’ …. One can find a short vignette of the encounter in the book “Technopoly” by Neil Postman.

Far more recently I received an email message from one of my grandchildren with a glaring hiccup in the form of “it’s” that should have been “its”. I stood ready to have some fun with this teacher of English, until I learned that many ‘smart-phones’ are conceited enough to mishandle the apostrophe on a grand scale, and are quite resistant to correction.

So much about the vaunted ‘smart keyboards’! As long as I can think my own words, I also want to write them as I want them.