Path: utzoo!attcan!utgpu!craig From: craig@gpu.utcs.utoronto.ca (Craig Hubley) Newsgroups: comp.cog-eng Subject: Re: one-finger keyboard Message-ID: <1989Oct16.142423.16103@gpu.utcs.utoronto.ca> Date: 16 Oct 89 18:24:23 GMT References: <1989Oct6.221013.8269@agate.berkeley.edu> <1259@cbnewsj.ATT.COM> <783@cogsci.ucsd.EDU> <1989Oct16.012114.23142@agate.berkeley.edu> Reply-To: craig@gpu.utcs.UUCP (Craig Hubley) Organization: University of Toronto Computing Services Lines: 150 Checksum: 35518 >I can envision a couple of applications for a one-fingered keyboard: By 'one-fingered keyboard' perhaps you mean 'a single-switch device whose serial output is assembled into characters' ? Or do you mean 'a many-switch device, each of whose switches can be reached quickly by one finger' ? A protocol for the former (Morse Code) already exists and many people (anyone with an amateur radio license) knows it. For the latter, work on photocopier layouts, TV and appliance control panels, etc., may be of some interest. >1. (as Ms. Foley suggested) Handicapped people may not have the > use of individual fingers, or they may need to type with a stick > held in their teeth or attached around their heads, or whatever. > They, then, are performing "one-fingered" typing. I think this is a narrow way to look at their problem. While they have a significant degree of physical expressive power, it simply isn't in the usual places (i.e. fingers, hands) that keyboards are designed to gather their input from. If they have most of their physical expressivity in their neck and head movement, then a set of head-mounted motion sensors would do a far more reliable job of distinguishing motions than a single-switch device, once it was calibrated to the movements available to that individual. Unlike keyboard-style devices, it could then be programmed to ignore some characteristic involuntary movements. Although it might not be as expressive as a 100-switch keyboard that can be typed on at 100 wpm, it would provide a lot more affordance to that individual. The software to translate this into words and arbitrary ASCII characters could be modular and reside on a portable system. Plugging the output into a keyboard jack would be all that was necessary to use any given computer. Not every machine would need to be equipped with these 'new' keyboards. Of course, this is a vertical solution for each individual, and not a horizontal solution to provide a lesser degree of access to many such individuals. >2. Hand-held computers / memory aids / phone dialers. In all of the > examples of these I have seen, the size of the device and the size > of the keys are such that the user typically holds the device in > one hand, and pokes at the keys with the index finger of the other > hand. A particular drawback to these (potentially very useful) > pocket-sized devices is that the QWERTY keyboard, while well-known, > is not at all suited to the task. Some of the most common letter > combinations require jumping across the entire keyboard. Also, the > size of the QWERTY keyboard is effectively 3 rows of 10 keys, making > it difficult to arrange a keyboard on anything that will fit in a > pocket. It's been recognized for a long time that the keyboard and screen are the clunky leftovers that restrict portable computers from being really portable. If our computer keyboards were fully interchangeable between machines, as they could easily be with some standards (they all generate ASCII anyway, plus some quirky control codes that are often encoded as sequences of ASCII), we could have abandoned QWERTY for Dvorak or whatever long ago, as each individual chose to do so, instead of waiting for some massive sea change. There are already chord keyboards to provide effective one-hand typing, which is 1/4 a solution. Freeing that other hand is another 1/4, and keeping the screen in front of your eyes without restricting movement is the last 1/2. Reflection Technologies in MA somewhere has come out with a small (1" x 1.2" x 3", four ounces, I think) CGA display that is head- mounted a couple of inches from the eye and looks like a full-size screen to that eye. They are selling starter kits for $495. This may solve this half of the problem, but really we need input devices that do not require us to hold something physically in either hand. >3. I imagine that there would be situations where someone with restricted > finger dexterity (in particular, if they're wearing heavy gloves or > the like) may need to enter data into a keyboard. An astronaut on > a spacewalk, for example, may need to type something while wearing > a spacesuit. He doesn't 'need to type something', he needs to 'express something in written form'. One may be able to select from options on a menu (which requires a screen unless you have a trained user & pie menus), translate from voice (difficult reliably today), or a number of other things. Since this user is already wearing gloves, why not rig up Dataglove -type devices, (soon to be available cheap from Mattel for the REST OF US), and just pick up the finger movements as if he was typing (a mode switch would have to precede anything interpreted as typing)... if you can type, then without a keyboard you can still express what you're saying as typed sequences, with a high degree of certainty that you got it right, even with no feedback at all. Since this user wears a helmet, having a display incorporated in it could be an optimal solution. I don't think the one-fingered keyboard is really optimal to solve any of these problems, particularly those where you have the luxury to define an individual-oriented solution. >I should say by way of disclaimer that I am not an engineer, and have >very little experience in ergonomics, per se. I am a mathematician, and >am particularly interested in mathematical approaches to language-related >problems. My approach in this problem has been to find an arrangement of >the keys which would minimize the expected key-to-key distance traveled >in the course of typing "normal" text. This sort of analysis was done by Dvorak and others designing alternate keyboards. Engelbart's chord keyboard referred to a lot of this research. However, key-to-key distance is not the only factor - the dexterity of each finger alone and in quick combination with others is important, too. Ideally you would map the most common chunks of ASCII (e, a, the, sh, and...) to the easiest finger combinations, not necessarily just one per letter - if you were willing to introduce language dependencies, in this case on English. But there are actually already language dependencies in QWERTY, even though it was designed to deliberately make English hard to type quickly so as not to jam up a mechanical keyboard - try to type 'wz', a common combination in Polish, quickly - the fourth and fifth (little) finger do not like to go in different directions in quick succession. >The purpose of my original posting was to see if anyone had already >addressed the problem, solicit suggestions from you who know more about >human factors than I, and to try and brainstorm up some other uses for >the one-fingered keyboard when I find one I'm happy with. The most appropriate use of a one-fingered keyboard, I think, is as a default keyboard for use when nobody is carrying around an input device with them. For situations (shopping malls, say) where this is common, it makes sense to ensure that the device is optimal - but if nobody is going to be typing in English but only responses to menu prompts... >Thanks to all of you who have replied, either by email or on the net. >Any further replies are still welcomed and encourages. Thanks. You have mine. Further lowest-common-denominator input devices are not my idea of a good solution to any of the problems you mention, although they may provide a default solution in the abscence of other options, they should be considered temporary and should be designed in a modular way to be replaced. If I have a QWERTY board, a dataglove, a head-motion- sensor, a voice recognizer, etc., that I can carry with me, then I would like to plug it in and ignore the one-fingered keyboard. If I don't, then I will very much appreciate the effort put in to make the OFK work well. But only until I learn something more expressive - very few people like to use Morse Code... at that point my new expertise must be supported also, or I will be frustrated. If I can type 100 wpm, or paint with my toes, I will not be happy with an OFK, even if the 'average' user loves it. Unless we make these devices adaptable for the non-average, we will be buying ourselves a mediocre world, built for someone's arbitrary idea of the lowest common denominator. Not to deride your idea, I think it's worthwhile, but only as a default, and only in conjunction with some more general protocol for hooking up alternative input devices. As I mentioned, the NeXT 100 MHz I/O cable, and the Apple Desktop Bus are examples of approaches that make this possible. This may sound funny, but I would love your one-fingered keyboard on a Mac SE (with a Desktop Bus interface), and despise it in a shopping-mall terminal (without such a jack). The difference is whether I have a choice. Craig Hubley -- Craig Hubley ------------------------------------- Craig Hubley & Associates "Lead, follow, or get out of the way" craig@gpu.utcs.utoronto.ca ------------------------------------- craig@gpu.utcs.toronto.edu mnetor!utgpu!craig@uunet.UU.NET {allegra,bnr-vpa,decvax,mnetor!utcsri}!utgpu!craig craig@utorgpu.bitnet