Talk:Chorded keyboard

So there are no clear WPMs. Are there any opinions by notable users about speed relative to regular keyboards? - Omegatron 22:53, May 8, 2005 (UTC)

Omegatron: With Spiffchorder, Greg Priest-Dorman typed 55wpm even after months of using another chorder. "So, I put my seven key chorder back on (after not using it all those same months) and did 55 wpm. " OjM (talk) 12:22, 21 July 2009 (UTC)

For example, each finger might control one key which corresponds to one bit in a byte, so that one to eight fingers can enter any character in the ASCII set if the user can remember the binary codes.

Aren't there only 7 bits in ASCII?—Trevor Caira 13:12, 28 July 2005 (UTC)

In 1980, I purchased a Microwriter MW4 - I learned to type about as fast as I could on a conventional keyboard. However, when I was at the Microwriter stand at a computer show in London, the salesman and I were chatting about the device. He seemed to have been nervously fiddling with the keyboard as we talked - but at the end of the conversation, he casually rewound the device and showed me both sides of our entire conversation scrolling across the screen. So he was able to type at normal talking speeds. I don't think I know anyone who could type that fast on a conventional keyboard.

However, one would not want to use one finger for each bit of a 7 bit ASCII character - the Microwriter uses 5 buttons for normal typing and one extra thumb button to act as a kind of shift/ctrl/alt key. 5 bits gives you 32 combinations (one of which is no-buttons-pressed). So you can get all 26 letters of the alphabet, plus 5 others. If I recall correctly, the Microwriter uses one for 'SPACE', another for 'BACKSPACE' and the remaining three for '.', ',' and 'RETURN'. Tapping the second thumb button with other key combinations makes the next character a capital or a number or a control character or a symbol (@#$%^&*(), etc).

I decided to write a proper article about the Microwriter - and added a photograph of my Microwriter - which sadly no longer works because the battery voltage regulator has blown.

It's definitely a cool interface - and as a 1980's predecessor of the PDA, it was easily a decade ahead of it's time.

SteveBaker 02:57, 20 January 2006 (UTC)

Chordon

I redirected Chordon here, but I felt some of the text might be useful to integrate here:

Chordon is a concept used with some chording keyboards (e.g. GKOS). It is a string of chords where common keys between two consecutive characters (i.e. chords) are not released.

Often, the detection of a combination of keys, representing a character to be entered on a chording keyboard, requires that all keys must be released before and after the chord (= combination of keys). This somewhat slows down the typing and may easily lead to wrong or missing characters because the gaps between chords may disappear while typing fast.

There are several methods to overcome this. One is to decide on the character when a key (any key) of a chord is released first time. This greatly improves the reliability. Another method is the Chordon technique where a more complicated detection of key combinations is used, allowing overlapping of chords during complete words and not requiring the release of keys that are common to consecutive characters.

The Chordon detection algorithm can not rely on absolute lengths of key presses because typing speeds can vary a lot. Instead, the average length of a single character must be estimated and the structure of the string of overlapping chords (e.g. a word or a syllable) is interpreted as a whole.

The advantage with Chordon detection is that overlapping is fully allowed (less errors, all characters are still detected) and, because there is no need to release all keys between characters, typing becomes partly parallel resulting in higher possible typing speeds.

Cheers. Ëvilphoenix ^Burn! 10:03, August 9, 2005 (UTC)

I do not understand the Chordon technique. It is not fully described above. However the read-chord-on-first-release algorithm described by Bequaert et al. in expired US patent 4042777 does provide excellent character recognition in the presence of chord overlap. The Bequaert algorithm toggles between two modes, chord forming mode and chord releasing mode. The first key press puts the keyboard into forming mode, The keyboard remains in forming mode through any additional key presses. The first key release puts the keyboard into releasing mode where it remains until any new key press returns it to forming mode. The intended chord is taken to be the one existing just before the entry into releasing mode, i.e., just before the first key release. A little thought shows that any keys needed by adjacent characters need not be released. The resulting feel of the keyboard is very natural and it's simple to program. Bequaert's method is what I use in the Chordite. --- John McKown —Preceding unsigned comment added by 24.251.186.204 (talk) 21:57, 27 February 2011 (UTC)

Microwriter

The Microwriter has SIX buttons - not five, I corrected the article (and added a photo as proof!).

Doug Engelbart?

 "Doug Engelbart, inventor of the computer mouse, may have invented chord keyboards."

...but then we have already said that the stenotype machine is a chording keyboard and that it was invented in 1858. So I don't think Doug gets the credit and I'm removing this statement. SteveBaker 16:07, 14 March 2006 (UTC)

I rewrote the history section, and I hope I've clarified Engelbart's contribution. Rbean 08:03, 26 March 2006 (UTC)

Chorded keyset redirect, and use at PARC

I've redirected Chorded keyset here. That page included this text, which was added yesterday:

A five key chord keyset was standard issue with the Xerox Alto in the early days at Xerox PARC. The advantage was that multiple keys could be struck at the same time, allowing 31 differnet combinations. There was an editer, ugh, which catered to use with this keyset. It allowed one to use the mouse with one hand, and the keyset with the other to input text. The chord keyset was mapped to allow the letters plus some escapes for use with upper case and numbers and other characters. There were folks who could enter text as fast with this as with a standard keyboard, and they had simultaneous mouse capability without moving their hands.

The chord keyset was also heavily used in playing Maze War, where the keys were mapped to the common movement patterns.

I'd like to see a reference for this-- everything I've read says PARC only used the mouse, not the keyset. I've also never heard of the "ugh" editor. Most of Engelbart's contemporaries seem to have dismissed the keyset as unusable (despite a few enthusiasts who became highly proficient with it) Rbean 08:03, 26 March 2006 (UTC)

The Digibarn has a page about this[1].--Jecel 16:53, 7 May 2006 (UTC)

Joystick Descendants

Wouldn't the Playstation controller and its friends fall under this category? They're not designed for entering text, but they can be used that way, like for naming characters and saved games. And, as a game controller, "chorded" combinations of buttons are standard. --Badmuthahubbard 09:45, 24 July 2006 (UTC)

I would say not. Game controllers do indeed use 'chords' - but not for entering text. Essentially, every game I've played had you wiggle the joystick left/right or up/down to select a letter - then hit a single button to 'enter' it. Hence it's not a chorded keyboard because it's not a keyboard. With true chorded keyboards (such as the Microwriter) you type text by pressing one chord for each letter...well, roughly...you might need to press some kind of escape code to get into or out of uppercase or to shift between numbers, letters and punctuation - but essentially, there is one chord per character you type. SteveBaker 20:14, 24 July 2006 (UTC)

I wasn't sure, it says "characters or commands" and some of the commands used by chords on video game controllers are pretty precise; not ASCII, though, to my knowledge. At any rate, I bet folks with video game experience would excel with these machines.Badmuthahubbard 06:56, 28 August 2006 (UTC)

Touch-Typing

"On the other hand, the failure of touch typing to penetrate the world after a century of availability leads buyers to question their ability to remember the chordings necessary." Failure of touch typing? In industries relying on typing or data entry, it is the norm.Badmuthahubbard 06:56, 28 August 2006 (UTC)

I am a computer programmer - and I've been using keyboards for 35 years without any kind of training in their use. I still can't touch-type although I do use more than two fingers to type. My son was taught to touch-type in school since about 8 years old (an odd quirk of US public education!) - and he's very good at it. I deduce that touch-typing has to be taught - you won't just pick it up from use. That's also true of chording - you can't just pick it up without cracking open a manual and consciously learning it. The significant difference is that with a QWERTY keyboard, ANYONE - even someone who has never used a keyboard before - can just walk up to it and have immediate success. They can see the letters and instantly understand that you press the button for the letter you want. Yes, they are very slow compared to touch-typists - but they can reliably do it. That is utterly not the case with chording keyboards. Faced with the six buttons on a Microwriter, people have no clue what it's function is - let alone how to drive it.

I have (on several occasions) attempted to force myself to touch-type (after all, I completely understand the concept) - and even after several hours of effort, I'm still a lot slower than with my usual 'ad hoc' six-fingered approach. However, I picked up the chords for my Microwriter well enough to type reasonably quickly on the two hour train ride from the Microwriter offices to my home. (My first Microwriter was faulty - and rather than post it back, I took the train to the factory and swapped it out there and then!) So I conclude that (based on a sample of one), chording is easier to learn for those motivated to do so than touch-typing - but the key factor is that you don't need to touch-type to make use of a QWERTY keyboard.

What upsets me the most is that I have this keyboard with 112 keys on it - and they couldn't put six more on there to allow one handed chording data entry! If they did that - then enough people seeking to type faster would use chording - and gradually we would find keyboards that ONLY have chording. SteveBaker 14:14, 28 August 2006 (UTC)

I also am a self-taught typist. I can type without looking at the keyboard, and when I try to "touch type" it is much slower. But you can type faster with a Microwriter than a regular keyboard? Interesting. — Omegatron 14:29, 28 August 2006 (UTC)

Yeah - I can type on a regular keyboard at pretty reasonable speeds without looking - but probably not as fast as if I learned the 'correct' touch-typing technique - but for most uses, I don't need to be able to type faster than I can think! My Microwriter ability was once much, much faster than my regular keyboard speed - I could enter text at 'dictation' speeds if I didn't try to fix typo's as I went along. But my ancient Microwriter's power regulator crapped out years ago and I can't recharge the batteries anymore. Without practice I'm pretty sure I could hardly remember the chords anymore. I'm left-handed and I always believed that I could do better if there had ever been a left-handed version of the Microwriter. The manufacturer always claimed that they hadn't seen any significant difference in speed with left-handers though - so maybe not.

My experiences with the Microwriter (over a couple of years before it died) were that it was great for word processing, documentation and stuff like email where it's just English text. But it really stank for programming because you need far more wierd punctuation like '<', '#' '{' and such that are quite hard to get to on a Microwriter. One unexpected benefit was that it left one hand free for a mouse - which is really handy. Also, it was around in an era before portable computers, PDA's or anything like that. The idea that you could write documents on this little machine while you were away from the office then download them to your PC later was probably more revolutionary than the chording keyboard.

I'd dearly love to have a Microwriter keyboard as a regular PC peripheral though. I guess I could probably build one with the remains of my non-functioning Microwriter. The software to decode the key switches ought to be pretty simple. SteveBaker 22:44, 28 August 2006 (UTC)

Or just disassemble a regular (wireless) keyboard and map the key switches to the keyboard's lines... — Omegatron 16:41, 2 September 2006 (UTC)

If you are typing without looking at the keyboard, that shouldn't "lead buyers to question their ability to remember the chordings necessary." Whether it's "correct" touch-typing, it's typing by touch. I think the article implies that typing without looking at the keyboard is not very widespread, and that this carries over into demand for chorded keyboards. --Badmuthahubbard 07:50, 26 September 2006 (UTC)

The demand for chorded keyboards arises mainly from wearable or at least pocketable systems, not desktops. John McKown

External links

User:81.64.235.149 - please don't keep adding links back to external sites unless they add information that is not contained in the article. Wikipedia does not exist as a link-farm. The official guidance is to not link externally unless there is content that cannot be in the article itself for some reason.

Tiki® puts gold under your fingertips — a 6 keys mixed system for digital mobile convergence was linking to a site that contains absolutely no actual information about their keyboard - or anyone elses. It doesn't add any content whatever to this page - it's a pure advert for that company.

SteveBaker 23:32, 19 October 2006 (UTC)

"Weasel words"

User:Flewis has politely requested that I provide an informative edit summary for my edit.
I have tagged several instances of usages such as "some question X" and "others maintain Y" with the {{who}} tag. This seems to me a very straightforward case of "weasel words", which we should be trying to avoid (generally by citing to a reliable source). We should eliminate these "some say", "others say" constructions by citing to specific sources saying specific things. -- 201.53.7.16 (talk) 15:38, 25 September 2008 (UTC)

See User talk:201.53.7.16 - I've placed the {weasel} template at the top of the article --Flewis^(talk) 07:24, 26 September 2008 (UTC)

Thanks. Seems logical. -- 201.53.7.16 (talk) 14:34, 26 September 2008 (UTC) —Preceding unsigned comment added by 201.53.7.16 (talk)

Give me an example

I DO wish this article, along with about a million others like it in WP, was not so damn snobby about giving clear examples of the procedures it refers to. I don't mind the "weasel words" in the article that much. After all, what is the guy to write "Mr Kafoops and his friends aver that...". What I do find irritating is that in all this verbiage, there is not a SINGLE example of how one might type an alphabetical character, and then a word. Couldn't someone, ANYONE, just give a step-by-step account of how one types the word "weasel" for example by chording? Myles325a (talk) 02:41, 25 October 2008 (UTC)