Total Pageviews

Monday, 16 May 2016

Getting SHIRTY about QWERTY (Part I)

- With apologies to Jimmy Fallon and The Tonight Show crew
St Joseph Herald, Michigan, 1868, six years before the
Sholes & Glidden first appeared on the market.
The reference to its speed is in relationship to
penmanship (handwriting), not sending Morse.
To this day, the QWERTY keyboard configuration remains, to paraphrase Winston Churchill's October 1939 description of the Soviet Union, "a riddle, wrapped in a mystery, inside an enigma". QWERTY will most likely remain so forever. The reason for this, perhaps, is that there seems to be every possibility the sequence of letters which comprises the "Universal" keyboard emerged almost completely at random, a juggling act of sorts. In turn, this might well explain why Christopher Latham Sholes and James Densmore didn't bother to leave behind a definitive explanation of the way in which they settled on QWERTY - maybe they simply couldn't. As desperate as we may be to finally figure QWERTY out, to decode it, there will always be obstacles - such as the absolute fact that we cannot possibly look into the minds of two dead men, from a distance of almost 145 years. These hurdles will remain insurmountable. Thus, every single theory concerning the configuration, no matter how well researched or documented, can be no more than that: a theory - and one based on pure speculation. What we are actually left with is a choice: Is one theory more plausible than another? That's really all there is to it: speculation upon speculation.
Let's just say, for the sake of an argument, that this is what started out as the 40-character four-bank keyboard, in alphabetical order, which Matthew Schwalbach came up with in late 1871, and which Sholes and Densmore then played around with in order to overcome "technical difficulties" (Ernst Martin). FGH stay together, for example, as the only surviving trace of an original alphabetical sequence, albeit on a lower bank. Can such extensive character movements be related to anything other than the efforts of Sholes and Densmore to get their machine working more efficiently? Schwalbach had already achieved the objective of a small, compact four-bank keyboard, to reduce the amount of hand movement by the operator. What remained was to ensure the typebars connected to the keys could operate smoothly. One overall task, but a number of necessary solutions, and much fiddling - the diagram above is not intended to convey an accurate idea of Schwalbach's original layout, simply to indicate the amount of (key)chopping and changing that went on. Were Sholes and Densmore premeditated, or indeed guided, in their decision-making? Based upon the balance of what evidence we do have, that would seem to me to be most unlikely.
A young journalist who worked with me at The Canberra Times and went on to "make it" elsewhere - in spite of little or no encouragement and guidance from his "superiors" - once trailed along on a lunchtime hunt to a large Fyshwick second-hand bookshop. I suspect he later came to place a little too much faith in a chance remark I happened to make as we walked the long aisles, that "If I was ever short of an idea for a column, I could pick up any of these books and find one". For me it still holds true, but good second-hand bookshops (in Canberra at least) seem to be disappearing at a rate far faster than New Zealand glaciers. Using the Internet is not the same thing, because, for one thing, an Internet search presupposes a subject, and risking a random selection is more likely than not to throw up something about which no column in a family newspaper can be written. What's more, old-stagers like me were brought up in a time when non-fiction works - references and textbooks - were almost entirely reliable (properly researched, checked, proofed, legalled, the works). Many of these are online, of course, if one knows how to find them. But today we're much more likely to be directed toward some half-baked thesis or an opinion piece masquerading as an article dealing in facts. Key in "QWERTY" on Google, for example, and see what nonsense comes up on the first page alone (oztypewriter notwithstanding). Give me a book anytime.
In the days when I was a newspaper columnist, there were occasions when I was monetarily stumped for a topic. Happily, my brief was to write on anything I liked. Not even when the subject was sport did I concentrate a whole column on sport. I suspect it might be much more difficult for someone charged with writing each week on nothing but, say, technology. Such a columnist might often relish stumbling across a topic that had just the slightest whiff of technology about it. That thought came to me when I read a piece in The Atlantic from May 3, 2013, written by one Alexis C. Madigral and headed, "The Lies You've Been Told About the QWERTY Keyboard - The QWERTY configuration for typewriters can be traced, actually, to the telegraph." Suspiciously, on the very same day, another article, by Wesley Fenton at TESTED, appeared under the headline, "The QWERTY Keyboard Layout May Have Come from Morse Code. The story of QWERTY as we know? Probably wrong." Madigral and Fenton had, obviously without too much checking of their own, picked up on an article by a Jimmy Stamp, published online earlier on May 13, 2013, by The Smithsonian Magazine. This one was headed "Fact of Fiction? The Legend of the QWERTY Keyboard - What came first: the typist or the keyboard? The answer may surprise you." Both columnists had taken their leads from a paper published more than two years earlier, in March 2011, by Koichi and Motoko Yasuoka of Kyoto University and titled "On the Prehistory of QWERTY". Neither Stamp nor Madrigal offered much in the way of an original thought of their own, they simply followed what the Yasuokas had written, passing it on as gospel. The abstract for the Yasuoka document says, in part, "In this paper we reveal the prehistory of [the] QWERTY keyboard along [with] the history of telegraph apparatus: Morse, Hughes-Phelps and Teletype. The early keyboard of [the] Type-Writer was derived from [the] Hughes-Phelps Printing Telegraph, and it was developed for Morse receivers. The keyboard arrangement very often changed during the development, and accidentally grew into QWERTY among the different requirements. QWERTY was adopted by [the] Teletype in the 1910s, and [the] Teletype was widely used as a computer terminal later."
The Yasuokas go on: "Then we debunk several urban legends about QWERTY". Koichi Yasuoka is very good at debunking, especially any ideas that might run contrary to his own.
One of the golden rules of old-style print journalism was, "Never risk publishing the word of anyone, without first checking out the facts for yourself." If Stamp, Madigral or Fenton had ever heard it, in this case they certainly didn't apply it. Stamp used the Kyoto theory to argue that the user (Morse operators) determined the structure of the QWERTY keyboard. There is simply no evidence to suggest that's true. Nonetheless, Stamp - and this is a guy representing the Smithsonian, remember - promptly adds to what he describes as the "myth and misinformation surrounding the development of QWERTY" by saying " ... right before their machine, dubbed the Sholes & Glidden, went into production, Sholes filed another patent, which included a new keyboard arrangement." Utter rubbish! Eventually Stamp gets to the point: "While it can’t be argued that deal with Remington helped popularize the QWERTY system, its development as a response to mechanical error has been questioned by Kyoto University researchers Koichi Yasuoka and Motoko Yasuoka ... They conclude that the mechanics of the typewriter did not influence the keyboard design ... The Kyoto paper suggests that the typewriter keyboard evolved over several years as a direct result of input provided by ... telegraph operators." Questions, conclusions, NOT facts.
As for Madrigal, he makes no attempt to hide the fact he is lifting whole paragraphs from good ol' "Jimmy Stamp over at Smithsonian". He boldly claims: "The QWERTY keyboard did not spring fully formed from Christopher [sic] Sholes, the first person to file a typewriter patent with the layout. Rather, it formed over time as telegraph operators used the machines to transcribe Morse code." "But the development of the design wasn't accidental or silly: it was complex, evolutionary and quite sensible for Morse operators." The rest is pretty much verbatim from Stamp. Did this guy get paid for copying and pasting Stamp's copy?
Fenton said first quotes that uber reliable source Wikipedia before himself dipping into Stamp's piece. "Smithsonian Mag turned up a lot more in their research." Duh? Their research??? Isn't this Yasuoka research? Oh, hang on, Fenton adds, again lifting Stamp quotes word-for-word, "Kyoto University Researchers Koichi Yasuoka and Motoko Yasuoka ... tracked the evolution of the typewriter keyboard alongside a record of its early professional users." Presumably Fenton held his hand out for some pay for this crap, too.
What happened here is far worse than mere plagiarism, however. Stamp, looking for something to hang some thoughts about the KALQ keyboard on, stumbles across the Kyoto paper and, without bothering to question its premise, runs with it. Madrigal and Fenton see Stamp's item and again, without looking too far into the Yasuoka theory, they simply repeat Stamp's unproven assertions. Suddenly, a paper which has been waiting around for two years for someone to pay any attention to it has developed legs. It's up and running. The QWERTY-Morse connection has got currency. A mere theory, pure speculation, enters the realm of accepted wisdom. It tops Google's QWERTY search page. All because of what passes for "journalism", but is actually very sloppy analysis, filled to the brim with unchecked assumptions.
I was - quite fairly and reasonably - put on the spot when Richard Polt asked about this theory in a comment on my blog post on the mysterious QWERTY-advising son-in-law of Amos Densmore, James Densmore or Latham Sholes (take a pick). "I wonder whether you have an opinion on the theory that the S&G keyboard is designed to separate pairs of typebars that frequently follow each other?" Richard inquired.
"In ETCetera #6, Richard Dickerson [Ed: above, creator of the "Dickerson dodecamer"] published an analysis that supports this theory, with the notable exception that the R and E typebars are only two positions apart. Dickerson calls this a mistake. The Yasuokas take it as evidence that the whole theory is wrong, but I am not convinced by their argument about Morse code - partly because, I admit, I just haven't had the patience to follow its intricacies." "Put on the spot" in that, oh dear, was I being asked to question to word of the Yasuokas, to tackle Kyoto wisdom and risk incurring the wrath of Kiochi? I did manage to read right through the Yasuoka paper, and like Richard was not convinced. In my case, far from it. I am not an academic, never have been, never will be. Nevertheless I am very familiar with modern academic processes, which in my humble opinion too often begin with a preconception and then set about manipulating the solid evidence, using it selectively to come to the desired result. That's not proper research, that's simply pushing a chosen wheelbarrow in a predestined direction. 
When its comes to working with what actually happened in the earliest days of the typewriter, I place a great deal of faith in Ernst Martin and Richard Nelson Current, one an out-and-out typewriter enthusiast, a one-time trade journalist and later publisher, the other an esteemed academic and writer. I doubt if anyone has studied typewriter history in more detail over a longer period than Martin. Michael Adler might challenge that, but then Adler allowed himself to reveal a distinct bias against Sholes - his agenda, a little like that of the Yasuokas, was to establish an extended QWERTY prehistory. But in Adler's case, much of it was irrelevant to the actual development of the typewriter we know and love today, thus rendering his work of merely "academic interest". As for getting inside the heads of Sholes and Densmore, nobody comes even vaguely close to Current.
Martin tell us this: "Among the first typewriter builders were some who believed that their keyboard [should] emulate a piano. They had overlooked the fact that typewriter keys must be individually hit in quick succession, and that it would be advantageous to keep the keyboard small, to accommodate the keys in a confined space to avoid the hands having to make large movements. Schwalbach recognised this need early on and persuaded Sholes to [employ] a four-row keyboard, which we have today.
"The world believes Sholes placed the characters in the same sequence as in the American printer's case - Sholes was a printer - [but this] is wrong. It is true that Sholes and Densmore worked out [QWERTY] together. Sholes' [earlier] experimental machines had a single-row keyboard with long and short intervening keys like the piano and the characters placed alphabetically, namely A-M long, N-Z short. Schwalbach distributed the keys over four rows, but the keyboard was originally in alphbetical order. The model shown in the first Remington catalogue [has in the second bank] ADFGHJKLM side by side. That the alphabetical order was not quite retained was due to technical difficulties, which had to overcome by Sholes."
NEXT: The views of Current, Schwalbach, Roby, Porter, Wheeler, Edison, Davidson and most tellingly, Louis Sholes. 


Richard P said...

I'm glad that a responsible journalist—there aren't so many!—is researching this topic that is of daily relevance to millions, no, billions of people.

Nonsense does spread easily on the Internet. On this Instagram photo:
a Writing Ball is described as "1940s. A typewriter designed to conserve the metal needed for the war effort during World War II."

Rob Bowker said...

Funny. Only last week was someone asking me "Why QWERTY?" I had to say that it was just "because", though there are many variants on the theme (AZERTY, QWERTZ etc) and other layouts quite different.

Steve K said...

They should've asked Mr Turing to make sense of it!

Bill M said...

Great article Robert.
I took note of "Never risk publishing the word of anyone, without first checking out the facts for yourself" My news writing professor to a T. Too bad modern news liberals do not know how to properly report the news. Ever notice how good reporters and writers have not existed for at least a decade or two?