Picture of Instructable: Keyboard System
In honor of Instructables.com, this will be a collaboration. My goal is to arrive at a finished design, as a product of the collaboration. Several steps in this instructable are reserved for posting specific ideas, programming logic, flowcharts, or other ideas related to that particular step, which advance the original concept toward completion.

As we already know, this instructable will serve as a record of any development of the seed concept, as it takes place. There are four optional paths to expand upon, within this instructable (Steps 4 thru 7). Each of these optional steps can be developed independently. Each does not need to be analogous to any other. They can be appended on a first come first serve basis.

I am hoping to decide upon what would be considered (by the group) as the most favorable programming method for the finished product. For now, I am asking that the simple requirements presented here are observed.

I have initially supplied these optional steps as clean slates, so that anyone may start from scratch, if they so choose, as long as the specific operational requirements in a particular step are met.

There are no limitations as to how much relative credit any particular contributor would be entitled to.

NOTE: As is the case with any forum, one should avoid editing any post, after a comment or response has been made to it.

Step 1: Keyboard Fundementals

I've added this step to collect information relevant to any and all of the following control options. This is for discussion of hardware/firmware combinations for detecting key presses, switch denouncing options, software development for the alternate layout display, etc.

Keep in mind that it is best if the different control options are easily usable with the remainder of the system. This is to allow easy testing of different options, and provide for the possibility of others adding different control schemes.

This step added by nah.

try 3d printing it

RockmanDash2 years ago
I think a good idea to add to this is a curved, possibly even rounded hand keyboard so it would be more comfortable
Voxel8 years ago
how about the index finger has vowels, since you need those for every single word, and the index finger seems to be more nimble. also i do not like all of those smaller keys stuck on the bottom of the thumb, that part of the thumb can't possibly push those small buttons, as far as i know
Scrupulous (author)  Voxel8 years ago
Hey, thanks for the input, Voxel. You have a good thought with the vowels. With those other keys below the thumb, they're just control keys. And, let's face it, everyone has to look down at the keyboard, and change their "home row" positioning, for most other keys except the letter and number (and ,.) keys, right?
Uh, no. As touch typist, I don't have to look down to find the Insert, Home, Page Up, Delete, End, Page Down, Ctrl, Alt, and arrow keys. So no, I don't look down at the key board a lot.
Scrupulous (author)  hammer98766 years ago
C'mon, let's be realistic, now...we all know that the average person will never hit any of those keys on a standard keyboard without looking (except maybe the CTRL). That's hundreds of millions of us.
Woo hoo! I am unique! According to Scrupulous, I am the last of a dying breed: The touch typist!
Scrupulous (author)  hammer98766 years ago
Yaaaayyyyyyyyyyy! (I still don't know what a touch typist is...)
Wait. You are designing a new keyboard and you don't know what a touch typist is?

From http://www.sectorsoftware.demon.co.uk/typist.htm
"Unless you learn to type faster without looking at the keyboard or at you hands then you are the most inefficient part of your computer system and are wasting part of your life." (Emphasis mine.)

Have you ever had a lesson in typing? Have you ever noticed the little bumps under your index fingers when they are sitting on their home keys? Do you know what the home keys are?

If you are using a qwerty keyboard, you home keys for the left had are "A-S-D-F' and for your right hand they are "J-K-L-;". Qwerty keyboards generally have bumps on the "F" and the "J" key. They are there so that your sensitive index fingertips can find their home keys.

Take some online lessons. Improve your typing skills.
Scrupulous (author)  hammer98766 years ago
Noticed 'em??! Heck, that's were I've been placing my PINKIES all along! Maybe that explains why I type so slowwwwww.

But, all "F'n'J"ing aside...I think the majority of us rely on those special keys, we just don't take as much pride in it. I've thought for a while, though notwithstanding pricing concerns, that every key should be "texturized" in some unique and readily identifiable way...hence, Raised Letter Keys:

P.S. have you ever noticed that "werf-jiop" would be a more ergonomically-correct home row?
http://en.wikipedia.org/wiki/File:KB_United_States_Dvorak.svg this layout works much better, but I don't want to spend time learning it.
That would be a great idea, but there is a problem here. Feeling requires some form of feeling, for example consider braille. Because the letter is a concept as a picture not a feel, it's completely a different thing and also requires having your mind remember it, and a letter isn't exactly the best texture, something similar to braille would be, but how many people would be using this that know braille? :P
Actually, I do the same thing. Most touch-typists can reach for ANY key on the keyboard without a problem. I personally reach for every key, without looking and relocate without looking. Perhaps from years of playing MMORPGs that frequently use these keys, but that aside, the true touch typist doesn't look at all.
OK, I admit that I sometimes (er, frequently) hit the wrong non-alpha-numeric keys, but I can hit the Backspace key easily still without looking. It is odd, though, that the lettering on my s-d-c keys are the ones wearing off, not my Backspace key. :-)
I guess perhaps the best thing about this keyboard would be that you would never HAVE to reposition because it would be so familiar of a stance (your actual hands layout!)
Scrupulous (author)  pyrofyr6 years ago
THAT, my friend, is possibly the c-r-u-x crux of this project...
Scrupulous (author)  Scrupulous6 years ago
How about the vowels on the index finger, and the rest arranged alphabetically on the remaining outer digits. That would give the alternating finger action for almost every word. I starting to shy away from the thumb being used for letters. There was some sense to that with the standard qwerty. Though, I believe the thumb is still highly under-utilized, now.
This looks like a good idea, but it looks like one would frequently mistype, whereas I can type one handed very easily(I used it to type this post).
mr.origami6 years ago
awesome sweet radical too cool 50 STARS******************** bingo should be called boing
Actually, that's only 21 stars.  Late reply i know, but i couldn't resist.
nice to know there's someone else out there who just HAD to count them :P
flashcactus6 years ago
I think this can also be used as a portable computer input device (mounted onto a glove) I'll make some sketches and send them if somebody wants
Scrupulous (author)  flashcactus6 years ago
Hey, this is open to any ideas... let's see those sketches.
that's how it could be used. i'm going to add another sketch that shows how the glove could look like
Sounds good to me, a keyboard glove is a great idea. You can modify a powerglove for it, huffah, huffah! But laughs aside gentlemen, this is a novel idea. I'm not sure it would fit this instructable, but perhaps in another one, although it's possible it would just be a radical change. Oh and 'finger tapping your palm' doesn't work, it was tested before and kinda sucks. D:
WurdBendur6 years ago
Putting the keys in alphabetical order is a nice idea, but it won't help much unless you're hunting and pecking. Since the layout would have to be relearned anyway, it should be designed for speed and ease of use. You might put the most common keys on the first two fingers, on along the bottom. But also keep in mind that putting all the most common keys together will cause the user strain from having to type long strings with one finger. The best layout would have the two strongest fingers alternating most of the time with other fingers coming in for less common letters. Putting all the vowels on one finger should promote this, but consonants should be distributed more. And common digraphs should be arranged on adjacent keys going left-to-right so you can easily roll your fingers over them (or maybe or top-down if the keys are low enough to slide your fingers over).
Scrupulous (author)  WurdBendur6 years ago
Yes and No.

I follow your thinking on the digraphs and such. However, from a marketing standpoint, this would have to appear simple to learn, otherwise it would be a bust. (Case-in-point: Data Hands)

So, having at least the consonants arranged in alphabetical order is as good as any layout (especially if they can digraph with a row of vowels) and it will be learner-friendly.
What if all the vowels were on the thumb so you'd tend to alternate between finger and thumb? The thumb is the strongest digit, so it should be able to handle the greater load, and the other fingers would share the consonants. As long as there aren't a lot of common sequences on one finger, it should be fine. There are exactly enough keys to fit all the consonants (including Y) on the fingers, and then the four other vowels (plus one more key) on the thumb. The escape can go there, where it isn't so easy to hit accidentally. Or the vowels (including Y) can go on the thumb, and the escape key can fit on a finger like it already is.
jcramer7 years ago
I'd like to point out that if done right, you could allow anyone to re-arrange the letters to their liking much like with a real keyboard. I'd be in favor of some research into the most ergonomic layout as some have suggested, and then declaring the decided one as the default. I personally would like to see maybe some variation on the one-handed DVORAK layout. Another idea is to put some kind of mouse pointing device into the setup. Maybe a trackball, or pointing stick from a game controller as one commenter suggested. I really like the idea that digital enigma had about strapping two of these on your hands, however it would really need a slick design to make it practical. Maybe where it hovered over the top of your hands when not in use, then you stretch out your hand and it unlatches a release that swings it down under your hands.
Scrupulous (author)  jcramer7 years ago
Coming back to this after quite a while, it immediately seems to me that maybe the most practical thing to do is this: 1. Have two "hand" controllers, where each one has half of all the letter keys (like the way those full-sized ergonomic keyboards physically split the keyboard into two separate "halves") and we keep the standard QWERTY layout. 2. Have each controller operate as a mouse as well, so that either hand can operate the mouse action (for lefties), and so that each hand can control separate functions of video gaming action. 3. Have the NumLock key somewhere on the left hand controller, so that the right hand instantly turns into a number pad (like a ten-key). Anything else?
Scrupulous (author)  Scrupulous7 years ago
I forgot to mention that the controllers would need some sort of bracing that your wrists could slide into, or some design that would stabilize the palm to the controller.
Scrupulous (author)  Scrupulous6 years ago
On second thought, that wouldn't be necessary at all...just a well-designed raised area in the palm (possibly secured by the base of the thumb) that allows you to move the device, and type at the same time! (I like this over the trackball approach, which would probably require moving the fingers out of typing position, just to move the mouse pointer.)
Having the thumb kind of fill into a 'groove' on it, could allow this to work, perhaps the thumb be 'gloved' and slip into an enclosure. Then you mostly move it with your thumb/palm, and you could have a button on the top of the thumb thing, and when that button is held down it changes it into "Mouse mode" in which case the first button on the pointer/middle finger become "Mouse1/2" buttons instead of whatever letter they were 8D
Scrupulous (author)  jcramer6 years ago
Right. I think we all agree that a default layout could be learned by anyone, and that no single layout will please everyone. I'm liking the idea of also making it a big optical mouse, more and more. The movement of the hand itself would have be be pretty smooth (maybe on rollers). But, there wouldn't need to be much movement...I set my optical mouse to max pointer speed, so that I barely have to move it. Imagine pointing the cursor into a text input window and typing in it, without moving your hands an inch... By the way, I'm of the camp that thinks trackballs suck eggs. I could be wrong, though. A nice big baseball-sized trackball in the palm area might just do the trick.
dombeef6 years ago
Can I collaborate?
Scrupulous (author)  dombeef6 years ago
Of course you can.
Scrupulous (author)  dombeef6 years ago
dombeef dombeef6 years ago
But will you
Scrupulous (author)  dombeef6 years ago
Of course!
Today? it says no colaboratoring
Scrupulous (author)  dombeef6 years ago
Sorry, it's been a while...where does it say that?
I dont know? But let me guess, go to my user name and it might say for me to colaborate on one of your instructables
dombeef dombeef6 years ago
Ok now I know - Ok go to Author Options-Edit-share-Collaborate-enable collaboration- add/remove collaborators -add dombeef
Scrupulous (author)  dombeef6 years ago
It won't let me do it...probably because the instructable has already been published (is my guess). Just feel free to add suggestions, imagery, and/or developments to any one of these fine steps. Yaayyyyyyyyyy!
Are you sure? I tested an instructable and it work fine
Voxel6 years ago
What if the plastic that housed the keyboard could have vibration sensors, so that when you wanted to space, you could kinda "hit" the whole assembly, not necessarily the keys. Spacing is so common, that it would be kinda nice if it didn't have to be delegated to the thumbs alone. Just a thought. Not exactly practical. I don't know much about the Dvorak keyboard layout, and i have never used it, but it is interesting to look at the usage of such an effective and efficient keyboard layout, and how it fell to obscurity because of the stifling monopoly of the far less efficient qwerty layout.
pyrofyr Voxel6 years ago
I see a few problems with this. 1)People enjoy keys, it's a typical concept around forever. Now we have 'touchscreens', but even then, mechanical keys are pretty much as good as it gets ATM 2)It seems too much trouble to have this 'vibration' key, also since you can hit it anywhere, it's a bit... risky. Maybe something that you can hit in with your palm, but it seems uncomfortable. This kind of keyboard would need to go through several (as in many, many) prototypes before an RC is worked out.
pyrofyr6 years ago
I just have to ask, because maybe due to the endeavor you happen to know. How would one go about creating an 'input device' or a different kind of keyboard? I've been looking into this for a long time, I know it's part of electronics, but I can't find anything specifically on the sciences behind a keyboard, game controller or what not, and I've been attempting to work on one, and finding the information on the electronics portion seems near impossible at the moment. So can anyone shed any light on this?
Scrupulous (author)  pyrofyr6 years ago
I would think that a hack is the most direct way. Take a dremel to a regular old keyboard. Preserve all the wiring, and rig up new buttons on a piece of wood...
Well, wihle it sounds like a good idea I have 2 problems. 1)I don't have a dremel (you mean a standard crafting dremel, not a 'serious business, I can cut drywall dremel' right?) I'm also guessing you mean 'rotary tool' as dremel is just a brand, although I know rotary tools are typically labeled 'dremels' for some reason (most popular brand most likely, but still) 2)Well, the main idea is that although this does allow me to use a cheapy standard PS/2 keyboard and turn it into it, it also means that I need the main chip at the top right corner (is usually there) which I'm guessing is the micro controller (or whatever they call that doodad when it's on a keyboard) and then the membrane comes next. The other problem is setting up the membrane correctly, and making two sheets. Now the way the keyboard normally works is that you have wires that don't touch, and wires on top that make them touch. When the wires connect they send the signal, so realistically I would need to tell exactly where which ones go, and work from there. Seems really tough without some knowledge behind it, hence the need for some kind of reference. I was considering sending in an inquiry to a company such as IBM that creates keyboards, however I'm not sure how smart that would be. The first idea I'd have is to prototype something similar to this. My next idea would be to attempt to make two of these, the second one would be more spider shaped of a hand (imagine holding a mouse but with a finger in each direction. And each finger would have a purpose (the ones that CAN, at least 3 fingers should be dedicated to helping movement), and those might do things, for example... Pointer finger - Mouse1 (as a normal click w/mouse) , an extended reach allows you to tap Mouse5 Middle finger - Mouse2 (as a normal click w/mouse) Ring finger - small button at tip, so that you can move leisurely but if you want can quickly extend to reach a button (Mouse4) Thumb - Has a button (Mouse3) but it's a lot 'tougher' to click and is smaller so as to not accidentally hit it while moving the mouse I'm still considering it. I had a few ideas to toss around, I'd just like to prototype it. I have also been very interested in working out my own gaming controller, I realize it would be a lot tougher, however I also very much so like gaming and would like to work it out a bit.
pyrofyr6 years ago
It's been a while since I've read this, but I suppose I'm still very interested in the development procedure of this, and mostly the ergonomics themselves. I agree with most here that the ergonomics definitely need to be re-thought, however I believe that they should be thought out perhaps right before a prototype is ready, as more things such as design need to be fleshed out a bit more. One thing that I can definitely suggest is two rows for every finger, and the reason I say this is because it's easier to have one finger TIP hit 4 keys that are aligned 2x2 than it is to have them aligned 4x1 vertically (or horizontally). I'll try and work out a picture of what I mean and post it later. I still love the layout, was hoping this would get done, but it seems to have been quite some time, and I don't see much advancement.
Scrupulous (author)  pyrofyr6 years ago
The 2x2 arrangement is a thought. I, for one, would have to kick it around a bit...
pyrofyr6 years ago
This is a great idea, great! However, there are a few things that just need to be addressed. First of all, I think it would be too tough to have this work 'as is' for both left-handed and right-handed people, it should be able to work for either easier. Secondly, I think something like this would work a lot better with either switchable stickers for each key, or something similar to LCD-Keys, so that they can be swapped easier. The problem with that however would be the price. As far as the actual working both ways goes, I believe having an extra digit out on the other side could work, it would mean that the pinky would do double the work, but this isn't always a bad thing. I like it the way it is now, and I'm not sure that the extra digit would help for anything more than swapping sides though. I think another thing that should be done is that the mouse should be eradicated in this build, the second hand could be designed around a mouse, but with a similar 'hand' design, with the difference being a trackball either in the middle, on a convenient fingertip(pointer), OR have the whole thing be a mouse (laser on bottom). The reasoning for this would be that any keys that change layout (Ctrl, Alt, Shift, and 2 "Alternate Layout" buttons) can all be placed on the fingertips for easy use. I must ask however, does anyone here have any experience with circuitry? I'd be interested in trying to make a prototype of this, but I don't know my foot from my hand in circuitry, especially of input devices, although I'd love to learn.
Scrupulous (author)  pyrofyr6 years ago
I cant remember if this was mentioned before but...going from left to right, I think the entire layout (whatever it winds up being) should be reversed when used with the other hand.
shooby7 years ago
I think this was going in the right direction (have you seen 'Data Hands'?), but your design is largely constrained by the Instructables logo. If your design could be better by straying from the instructables hand logo shape, then do so, otherwise its just a display piece.
Scrupulous (author)  shooby6 years ago
The design doesn't need to be constrained to the original shape, shooby. I did that as a gesture, and to serve as a starting point. But, I think the true ergonomically optimal shape is something that will barely resemble an actual hand.

The fingers, for example, swing naturally on an arc around the second knuckle. If you'll notice, each fingers covers the length of an area easily big enough for 8-10 keys (not that actual finger movement is absolutely necessary).

I hadn't ever seen those "Data Hands" but I am disgusted by them. It took thecompany half of the user manual to get to a diagram that even resembles something remotely understandable, let alone effective. The people who invested in that product were obviously smoking some sort of petroleum by-product.:)
I3uckwheat6 years ago
you could also make this into a mouse
i want one its better then the cheap walmart one i am using
seems quite awkward to use and get used to. I think ill stick to my keyboard for now.
A good name7 years ago
The human hand is not agile enough for this. I can personally type about 50-70 WPM at leisure, and I'm only 13. The fastest I've ever typed was about 144 WPM (That was REALLY rare though, only happened twice) and so I figure that the keyboard is just perfect as is. Seriously, it's okay as is.
good ideas here keep pushing i think we all agree the KB could use a update! i would enjoy a 2 peicer that can strap to the wrists and allow type at any arm postion! always typing (or waiting to type) holding my arms in front of me makes my arms feel zomie-bish!
sweavo8 years ago
ETAION are the most frequently used letters in English. Ergonomics would dictate that the most frequently-used letters and pairs of letters should be placed most conveniently, e.g. the digraph TH is very common, and the trigraph ING. So something like Index finger: EIN Ring finger : HO Third finger: TR Little Finger: Z Q ...where the first letter is in the finger's most relaxed position, would seem to make sense. Some study would be needed on the most used characters, pairs of characters, and trigraphs in a given language (say EN-US as used on the 'net) and the key layout be generated for each locale based on that data.
Scrupulous (author)  sweavo8 years ago
According to Guiness (circa 1992), the most words ever typed in a single minute was 216. This was done back in 1946 by some poor punt who may have not yet realized that there might be more to her own life than typing dictation at a sustained rate of almost 4 words a second. One could surmise that the value of the things one had to say (or type) would be inversely proportional to the time one spent on increasing one's typing speed, naturally. We may never need to type 216 WPM again. Why would we? What would be the true need? I don't deny that some favorable way to alternate regions for the letters that are included with common groupings would be desired. Yet, I must insist that there would be no single arrangement that would rightly satisfy everyone. The moment you complete a thorough and realistic study on the current needs for letter placement (and probably before you have time to summarize it effectively), a new set of needs arises in that language as a result of cultural changes. The reason why the QWERTY layout has been so prevalent is because it specifically does not take into account the considerations which agreeably make the most sense. It just is. We have all been forced to learn it. And, we see that our left pinky can do some amazing things. Ironically, it is the left hand that must type the three most common letters, many of the most common groupings, etc. Take the English alphabet, for instance. It is in no particular order, itself. Sure, it's alphabetized. But, what does that mean? Are all of the vowels grouped in some convenient way? Apparently not (unless I've captured it with my layout). Are they arranged in order of their frequency? Nope. They're just in alphabetical order, for what it's worth, which is constantly minced by the construction of ordinary words and the corresponding frequency of use for all the letters. In a sense, the alphabet becomes partially useless the minute we go learn some other arrangement. And, I say that fact is worth something. We've all had to learn the alphabet, too, even before we ever learned to type (in most cases). Now, count them. That's two (2) too different ways to memorize the English alphabet. Why? The answer is Why Not. But, the question is still Why. So, given that we could all conceivably do 216 amazing things, using all of our fingers, in any given minute, we might conclude that it isn't really important how the letters are arranged, as long as they are all there. Right? Now, when I suggest an arbitrarily alphabetical order on a keyboard, it sends some people into a tizzy. That's fine, and perfectly understandable. I would suggest, though, that their issue lies with the QWERTY layout, itself, not mine. And, with the unique functionality that was developed on this instructable after it was originally posted (and within the first 100 or so comments), you may find that ergonomic arrangement with respect to letter frequency has forever become even less of a legitimate consideration. I encourage anyone to discover why. I know it's a lot of reading. But trust me, it's kind of interesting. Sweavo, you are obviously smart enough to offer useful thinking on this. I'm just using your comment as a springboard for the invitation that I must present every so often to dig into this thread and see what has been developed already. And, thanks for the well-stated advice.
Haha, I'm only in elementary school and I once typed 144 words per minute... I can type at almost a constant of 114 with no errors... I however lose a lot to backspacing.
Interesting rant :-) You're right that qwerty does not pander to the user's idea of convenience. IIRC qwerty comes from mechanical constraints in a typewriter - the letters were arranged to stop the type bars from jamming as they left or returned to the type basket. The ordering of alphabet is arbitrary, sure, but there is only one constraint on how to order an alphabet: it must be consistent. Here we have a physical input device so it seems to me clear that anything we know about the hand should be brought to bear on the problem. If I hold my hand up and wiggle my index finger 20 times, I can do it faster than I can count to 20. However, if I try to pulse the palm, as if I'm pressing with the part where the index finger joins the hand proper, then I am a lot slower, and I mess up twice before reaching 20 clean presses. I know where I'd rather have the E key :-)
sjrSpike sweavo7 years ago
Having skipped around quite a bit, and admittedly not reading the whole bit, I apologize if I repeat anyone else's ideas: I have two suggestions about your layout: 1 -- make the 'hand' with a snap-off cover so that users can interchange the key layout themselves, whenever they need to -- this would automatically require a program that pops up an onscreen keyboard and lets the user press a key and mouse-click to the onscreen keyboard to set the 'value', but would not be any more challenging than many of the PDA character recognition programs currently in use. 2 -- Instead of a single row of buttons down each finger, put in two rows; this should give more than enough room to move all those fidly keys up off the palm/lower hand, as well as not have to extend quite so far down each finger as to induce cramping, and still be quite usable from a reach standpoint. I envision that the keys' platform could be set in such a way that the platform itself, instead of the keys, tilted each row in slightly towards the middle of the 'finger' so that the 'rearrange-it' system in #1 could still work. It would also give the opportunity to arrange keys like 't' and 'h' next to each other so that a slight rocking of the finger was enough to type both.
A good name7 years ago
People couldn't adapt to this. It's to hard to use individual areas of your fingers to press keys... alternatively you could probably make two hands... maybe even a foot for space :P Anyways,I'm guessing it would take way to long to learn to note use the tips of your fingers.
Hey it may have already been said.....Now we can have carpel finger....Or some other repetitive finger stress injury...heheheh
Scrupulous (author)  digitaltripper8 years ago
Yeah, to go along with my other "mouse-claw" hand. I'm just lookin' out for the workers comper. ; )

Actually, the key layout got abbreviated so that all the letters are within one key of home row. If you're curious as to how that can be done, and you have a bunch of free time (who doesn't?), read through this baby. Who knows, you might have something to add.
I have an idea how bout we make it cure like this to make it fit the better
Thanks,(at least i fit in with all you guys) something i thought up.
even though this is a hand keyboard it might be good idea to think about making it a type of game pad/usb controller with some kind of stick like the kind on the psp on the bottom with rubber on the tip for a joystick input and sliders so it can move on a pad or something
Scrupulous (author)  pokefreak5017 years ago
What you're saying sounds cool. I'm just not sure I'm picturing it completely. Is there any way you could post a sketch?
Oops sorry for taking so long to reply but here is a quick sketch (I had only 5 min to draw it) the big black circle is the PSP style stick and the little hollow ones are the sliders.
hand joystick.bmp
we could curve it and put a joystick in the middle then make software to make an on screen keyboard and then the sliders could be the mice and left and right click. hows that?
i dont get it... it just says "poop" for every step!
nvm my last comment, they updated it now...
add a mouse to the palm of the keyboard
Scrupulous (author)  potbellypiglord8 years ago
Yeah, that thought keeps coming up. You mean like a trackball in the palm area, right? It almost seems like the right thing to do. I mean left. Now, if you're talking about a big old mouse-like device that rolls (or glides) around on the desk...I suppose you could do that, too. You'd have to make it pretty lightweight, otherwise your left forearm would become excessively huge.
Maybe you could make a second hand for the mouse With 2 mouse buttons and 3 shortcut buttons
Scrupulous (author)  ipod killer8 years ago
Yes, very good. Well, does anyone make a mouse with additional buttons, say, for the thumb?
actually, ya. ive seen [ergonomic] optical mice w/ a thumb button for back and a ring finger button for forward
Darkshot7 years ago
weird..........but cool
aceman 5698 years ago
I don't know about the whole not having all the keys in one layout thing. It could be very irksome
tasty. I'm new and i credit this idea with peaking my intrest
Scrupulous (author)  the fat hobbit8 years ago
Hey fat hobbit, thanks for the vote of confidence. Now if you could just hit that (+) button with your mouse pointer...up near the top of the page...that'd be great. BTW, you're still welcome to contribute anything that hasn't already been contributed.
Phill8 years ago
Now; I suggest making a keyboard/mouse. Going back to type, then going back to a mouse is way more annoying than anything ever conceived by man.
Scrupulous (author)  Phill8 years ago

Do you mean like a palmball? or a keyboard on wheels -- er..ahh, roller(s)?

Your right about the back-and-forth thing. I do it all day, every day. It's second only to pushing a phone button 3 times to get the letter S. I guess kids just love it.

I interpreted this as a mouse on the bottom of the keyboard. Although I hate mice because they get stuck a lot, optical isn't so bad, and better than a trackball/etc only for DRAWING in my opinion. But Mice need lots of mousepad space to move around on. Also I'm often annoyed when a straight line "staircases" when I release the mouse button. Thank God for "Undo"! But I think like a good optical mouse on the bottom of a keyboard!
I really don't know what I'm talking about, it's just a huge bottleneck, ya know?
I don't mean to be a downer, but what if you are a lefty?
You're not a downer. And, I am a lefty.
Yokumguy8 years ago
Sound cool. What happens if you need to hold down 2 different letters from different layouts at once? You could use a sticky keys sort of thing.
nah Yokumguy8 years ago
The keys are decoded in real-time(more or less), so you can't hold down keys from different layouts. Any keys that are depressed (before or after changing) will be decoded based on the current layout. I'm not sure I follow what you mean about using "a sticky keys sort of thing."
Yokumguy nah8 years ago
A sticky keys sort of thing: On windows (yes, I am a mac user) there is a function where you press shift 5 time and you then can press each caractor individually instead of holding them down.
nah Yokumguy8 years ago
Oh yea, now I remember. I hate sticky keys so I always turn it off, haha. Right now I'd like to get something that works. Getting the baseline functionality going right now is not going to be simple, so I'd like to get that done, and then consider little complications that add minor functionality. Also, it will (hopefully) be fairly easy to create/edit custom layouts, so having the keys you need on one layout shouldn't be a major problem.
dev45 nah8 years ago
Sorry guys I'm going to be busy the next couple months I'm truing colleges so the prototype is going to the back burner for a little bit
Scrupulous (author)  Yokumguy8 years ago
Thanks. Yokumguy. As far as needing to hold down 2 keys from separate layouts, I can't think of when you would need to do that. Can you give me an example?
you know it would be kinda hard to press enter with your layout i would suggest you have like 10 different fingers (on the keyboard) and have them not on the palm
Yokumguy8 years ago
I have noticed that many games on the internet use the keys WSDA instead of arrow keys. You would need to place those keys into a good spot, or you could make a gamers layout. You could even have a user settable layout so the user can choose where they go.
it looks alittle hard to use
nah8 years ago
I ran into an extension for Firefox that, while not directly keyboard related, is interesting in that it is an alternate input system. It is based around using mouse gestures keyed to common tasks (e.g. load home page, close tab, etc.) I think it is interesting if you're looking at different and better input systems. I don't know of any page for this other than the download page, so I'm afraid you'll just have to check it out for yourselves. clicky
Mepain8 years ago
i think that the bottons should be orange, the hand should have the pattern (with the lines and whatnot) and the bump that sticks up should have the word "Instructables" on it. i think that this is a great idea!! not much else for me to say here! =D
Scrupulous (author)  Mepain8 years ago
I can see it now! Thanks, mepain.
now that i think of it, instead of making the whole button orange, maybe just make the letters, lettering, and #'s orange.
your welcome!
Mr.Grinch8 years ago
From my experience in programming. it doesn't sound like you need to adapt the screen fading to every program for them to fade out, just each operating system, Microsoft, Linux, and Mac that says to fade the whole screen and open the application- "alt_keys" for example. Also if you need help with configuring the keys to work correctly as stated by "Nah", you should contact Microsoft and Mac and Linux and talk to the programmers of the operating systems to figure it out. My blackberry has something like that, pushing "sym" pops up a screen and "A" turns into "]" and "E" to "{" the calculator works the same way. it all depends on how the operating system is configured.
Scrupulous (author)  Mr.Grinch8 years ago
Yeah, we figured that would be one of the better ways to do it. Of course, emulation is another way, etc. If you know any contacts at one or more of these companies, it may be helpful. One of the main ideas on this is to free one hand to use the mouse full-time.
pascha8 years ago
I really, really like the idea of this, and I have simplifying suggestion and wonder if you have considered adding a SHIFT-LOCK key rather than adding all the complexity of using the mouse scroll wheel. Also, what percentage of PC users have scroll wheels? 90+%? Mac users? near zero, I'd assume. My other thought is to supply two of them and each would have a switch or shift key allowing one to be letters and the other to be numbers. One right-handed and the other left to suit users.
Scrupulous (author)  pascha8 years ago
I'm glad you like it, pascha. We've since moved the wheel over to the keyboard side, and considered some type of toggled lever instead. We've also agreed that the mouse might never go away, and that it may not be a good thing to take away any of its capability. There is a diagram of a more recent physical layout, further down this thread. (We all agree that the thumb row could be more horizontal.) Letters and numbers...not bad. We talked about making this modular with physically removable fingers and keys. That was the extent of it. Thanks for the feedback. I know there is a lot to read already. But, it takes you from the starting point shown in the intro, to something even more useful. People have made particular criticisms, which indicate that they haven't read the entire thing, and that they have no idea what the current status is, or why the interest has gone on this long. Lately, we've been talking about using two adjacent keys to create an intermediate character.
CBL8 years ago
First, I want to say that I think this is a really neat idea. Kudos to its brain-parent and the brain-village raising it. This may not be the best forum for this thought, but here goes. . . . 1) The keyboard should have a place to rest the heel of the hand, unless it is absolutely necessary to use the heel for the space bar. 2) It's difficult to have vertical motion (or at least precise vertical motion) with the thumb. That's why the thumb is usually reserved for large keys on the bottom of a keyboard -- space bar, numeric 0, etc. A shift key, space bar, ctrl, and maybe a wheel, can be added to the side of a slightly thicker keyboard (like the side button on a mouse), so the thumb can be utilized more effectively. 3) I agree that alternating fingers makes for quicker typing. Rather than using an alphabetic system for ordering the layout, a frequency system (etaoinrsh -- it varies based on written, spoken, etc., but they're all pretty close) would keep certain fingers from cramping up. 4) To replace the thumb keys (left open by suggestion 2), and to shorten the height. . . . My forefinger, for example, can only reach 4 inches comfortably -- the height of my keyboard from space bar to numbers. That's 5 keys for that column. Extra columns of keys can be added for the forefinger and pinky. On the QWERTY keyboard, we typically reach with them anyway. These suggestions leave about 30 keys open for the fingers, and maybe 4 (in 2 rows of 2 along the side) or 5 plus a wheel for the thumb. Another 'reach' key (or 2?), such as parentheses, slashes, semicolon, etc. can be added to the top of each finger. That makes the 40 keys on the layout above.
Scrupulous (author)  CBL8 years ago
Very good.

Yeah, I'm not sold on the exact location of the space "bar" either. But, the hump in the palm area represents a support, which can be heightened as needed. I like your thought on side keys. We've agreed that the existing thumb keys can be placed in more of a horizontal line, if nothing else. Even an arc that follows the lateral path of the thumb, would provide a swath of keys on roughly the same plane as the finger keys.

Key placement has been the great debate. And, yes, ergonomics definitely come into play, so as to avoid a grossly overdeveloped left forearm. But let's face it. I think most people could stand to beef up their left arm at least a little bit. ;>)

Thanks for the feedback. You've helped me arrive at an interesting notion...
forgive me if i sound somewhat skeptical, but what is the point? to make money or be famous? on the other hand it DOES look like a cool idea...
Other hand-- hahaha (not really that much)

Its like that FrogPad, it will never catch on. Even the dvorak keyboard hasn't caught on, and it requires no extra equipment.
Scrupulous (author)  Lithium Rain8 years ago
This project has morphed into something pretty interesting. You would need to read through the discussion, to see how that materialized here (quite possibly for the first time).

What you see, with the original idea, has taken on more capability. That much was fully expected, because of the sheer number of intelligent people on this forum. This project was posted, in part, out of appreciation for this site, and is turning out to be a benchmark, of sorts, in collaboration. I think others would agree.

So, the point is that we'd all still be in the dark ages, if everyone was too consumed by such things as money and fame. We're all just making the world a slightly better place, one stroke at a time. And, the Instructables site here serves as an effective way of legitimately documenting and archiving the process.

By the way, you are more than welcome to help... ;>)
A fair question, if somewhat odd all things considered. To be fair, this is a site dedicated to mods, hacks, and general do-it-yourself type things. In some ways, this is the epitome of that, some one had an idea for a device that cannot be bought, so it's being made. Now, for a bit more of a "real" answer, if you will. For me, this is, in no small part, a learning experience. I can't say that I've ever worked with USB at this level before, nor drivers. This gives me a project to learn on, and these are marketable skills for me. In some ways, that's worth it's weight in gold. It's also a good chance to explore alternate interfaces. Let's face it, the QWERTY keyboard is getting long in the tooth, and isn't really that well suited for many applications, especially mobile computing. Now, granted, this is not the end-all, be-all design either, but I think it's a good place to start. What's more, after doing this I will have solved a number of basic problems in dealing with alphanumeric input. What that means is, in the future when I need to have input of that type for some project I will have a good starting point and a basis to build off of. I also find things like this interesting. Solving problems, learning about systems, and getting to see other peoples ideas are all things that I enjoy. I do realize that this is the hallmark of dorkdom, if you will, but I don't much care as it make me happy. So, in short, why do it? Because I enjoy it and am learning at the same time.
zachninme8 years ago
You shouldn't do it alphabetically, you should use either a dvorak-system (Alternating fingers is faster) or at least, use etaoinshrdlu... (letter frequencies, and I know those 12 by heart)
usuck8 years ago
thats awesome but the fingering would be a bit uncomfortable, imagine typing the word "at" or other words that could possibly sprain your wrist
nah8 years ago
Ok, I think I've got the code so that it will "work." I'm not positive if I have some of the USB stuff right, but I feel like testing is the next step at this point. I wish I could put one together and test it myself, but I simply can't right now. I'm hoping that a few folks and proto one of these and see what's broken. I've included a parts list and a .bmp of the circuit this time, as well as a new keyboard layout file for windows. If some one wants to make a layout file for Linux or OS X, let me know and I'll do what I can to help you.

If you don't have an AVR programmer, I'd suggest getting a Butterfly and using ButtLoad. At the bottom of that page there's a link that has the stuff to make it USB easily. Probably the cheapest and easiest way to get a USB AVR programmer.

I'm going to hold off on adding the rest of the functionality until we can get it working at this stage. There's a few things in how an endpoint is setup that I'm not sure if it's correct, so we'll just have to see what happens there. If it works, it will only have a single layout and the LEDs wont do anything.

Oh, and for the record, I guarantee nothing about this and am not liable if it damages anything you own, or you.
dev45 nah8 years ago
Nah I looked at the print, seams easy enough to put together but I do not have a AVR programmer so that mite take a little bit of time but I have to order the chip any way. I have a 12MHz clock any way but what is the ISP header exactly? And is there a map of the key matrix to help me wire this thing up?
nah dev458 years ago
I'd get a programmer that is ISP(In System Programming) capable, then you don't have to worry about taking things apart to re-flash the chip. A lot of programmers wont have sockets for all the different packages, so being able to solder the chips down and then flash them gives you a lot more flexibility. On that note, the ISP header is just the connector for the programmer. I believe that it's just a 2x6 set of header pins that are .1 in. pitch. There is also a 10 pin version that some programmers use, the two aren't pin compatible (obviously.) If you need or want to use the 10 pin version, look at the specs for Atmel's STK500, it has the pin out and it's not really any more complicated. I've attached a layout for the keys that I hope makes sense, let me know if it doesn't.
dev45 nah8 years ago
Nah Now I get what ISP is thank you. I am looking at trying to make a buttload but I'm having trouble getting the plans but I will work on that well I am waiting for the chips, when I get some one with a card to order them. Any tips on making one. I may just be looking in the wrong file but I'm locking for the map for the columns and rows like for (g) what column is connected to witch row to make the letter? I see the layout for the fingers and such but not the matrix map.
nah dev458 years ago
The ButtLoad isn't really that bad to make. If you download the v2.0 from AVRFreaks and look in the "support" folder there is a connection diagram that shows the connections for both the Atmel (six pin) and Kanda (ten pin) headers along with level shifting resistors. I would suggest building any cables so that you can unhook them from the Butterfly, then you can easily experiment with it as well as using it as a programmer. When it's all said and done, making the cables is the real physical work to make the ButtLoad. I haven't gotten a chance to make a USB version, that's a relatively recent advent, so I'm not sure of the particulars on that one. It looked fairly well documented, and easy, though. I thought I had uploaded another file with my last comment, and it turns out that I didn't. Sorry about that. This time there should be a file called "key_caps_matrix.txt" attached. That should be what you need.
dev45 nah8 years ago
Yes that's what I was locking for thank you. I'm still working on getting dad to order me some chips because I don't haven't a card just no need till now.
nah8 years ago
Ok, after a bit of back-tracking and some reworking, I've got another preview of the code for you guys. The USBTiny library doesn't (didn't) directly support HID devices, so I'm having to make some additions as I go. I also realized that the Mega48 didn't have enough inputs. I haven't settled on which model, in particular, will be best to go with, so it may not be consistent between the schematic and the code, etc. and may change at some point. This should give you some idea of what's going on, and if anyone wants to take a crack at this and make changes, feel free to. I'll merge any additions in once they've been gone over.
VIRON nah8 years ago
The code looks good. Although I'm not familiar with AVR (never used one
in a project), does this prototype board give you any useful ideas as to
what controller to use? (40 pin ATmega16-16PI or AT90S8535)
40 pin AVR USB protoboard
Sorry I don't have software to read your schematic, in case it's similar.
I'm not trying to advertize the board; just showing a "typical" example.
nah VIRON8 years ago
It looks to be fairly similar. If you want to check the schematics, I used the latest version of Eagle and included any non-standard libraries that I might have used parts from (I'm not sure if I mentioned that before.) Most of the reason I haven't picked exactly which controller to go with is to make sure the code will fit and it will have enough I/O pins. I don't think that will be a problem, but I'm hedging my bets on this one.
VIRON nah8 years ago
I think the 40 pin chips have 32 i/o's instead of 23. Also the chip and the board is similar to what's inside typical computer keyboards, for practical purposes. I'm amazed at your speed in the actual work on the Makeing. I think I've only contributed experience-based knowledge and ideas, and my arrogant passion against excessive complexity in general. Dealing with the USB port would have annoyed me for a long time.
nah VIRON8 years ago
The 40 pin chips have 32 or 35 I/O's, but they also normally have xtal1, xtal2, and reset on their own pins, which is nice. On that note, I looked, and probably the best bet is to go with the Atmega324p. It's a step up from the 164p, but I'm concerned about cramming all the code onto the smaller chip. I checked, and the main hex file is just under 10K right now, and growing. The 16k of the 164p would probably do, but I'm not positive. There's a fair amount left to add, and it's only $1 more for the 324p. I'll try to get a parts list together in the next few days. It's amazing how simple keyboards really are when you open them up. Depending on the package you went with for the micro, I bet you could get the whole circuit onto something smaller than a business card. I wouldn't discount experience too much. I'm sorely lacking in that department, and it shows at times. I think simpler is better as well, but sometimes, but when complex is the only way to get what I want, well, it'll do.
VIRON nah8 years ago
A wireless qwerty the size of a business card (or wallet calculator) is Makeable, especially using IR beam, but that's too small for a hand.
Tiga8 years ago

i think you should scrap the mouse. have 2 of those hand pods, and incorporate an eye-tracking function. instead of "point and click", it'll be "look and click", hell if we're going to be breaking boundaries, why don't we have a blink function with the eye-tracking? 1 blink = one click, 2 blinks equals double-click. wouldn't that be innovative? theres already people incorporating eye tracking mods into their video games.
VIRON Tiga8 years ago
Stephen Hawking (Famous Paralyzed Astronomer) has that, I think.
Actually, this project is somewhat set on the hand shape.
There are a lot of patented handboards that don't look like hands.
Wolfking Handboard
More Handboards
Scrupulous (author)  VIRON8 years ago
Let's remember that we've all probably been "programmed" into looking at buttons in a simple way. The first typewriter must have been a marvel of modern engineering at the time. The linkages, the concern for leverage, speed, and so on, only got better with the electric typer. Somehow, we got stuck on the idea that the space between buttons was off limits. We know now that it may just be the next frontier. I honestly couldn't tell you what the cap on WPM's would be on this. I could not rule out the possibility, though, that this could easily exceed everything we have ever tried. In fact, I have not seen anything like this in existence. Yes, it relies on old-hat chording approaches. But, it does so in a way that allows typing to be more of an analog process than a digital one. It does about as much with the English language as possible, given the nature of English itself. This optimizes the way we learn, and among other things, avoids confusing that process. I can go on, but I don't thinks it's necessary. A long time ago, I realized that there is almost literally no end to what we can do with true electronic input, using our fingers. And, I'm talking about some of the tangents that have even been alluded to on this thread alone. You guys seem to grasp the notion that we simply don't know what is possible, until we get there. And you've been pretty darn impressive, in the process. Viron, I have come across patents for some of the devices you linked. (It's usually interesting to see the difference between the appearance of the actual product and the patent specification that it's derived from.) But, as you know, none of these do what has been fleshing out here. And, I would not be surprised if much more develops from this. You seem to be very valuable in this, as I said much earlier on this thread, for bouncing ideas around. At this point, I want to throw in another vote of confidence in the ideas that have been progressing. As easy as it would be to get distracted, it would be just as worthwhile to follow-through on this. Keep up the good work guys!
nah Tiga8 years ago
There has been research done into the general use of eye-tracking systems. The short of it is that it isn't easy to do right now. Resolving precisely where a person is looking on a normal monitor would be a non-trivial task. If you want to take on that project, I'm sure it would be both interesting and rewarding, as well as do-able. I don't have the time, skills, nor desire to do that right now.
nah8 years ago
Ok, so I've looked at options for implementing the USB portion, and here's what I've found: There are two main ways to go about this, either using a software stack, or a USB enable micro. The software stack will probably be a little cheaper and parts will be easier to obtain, but it wont be fully USB compliant (which is not to say that it wont work, it just takes a few short cuts). There are two main options for libraries that I see, Ob-Dev's, the one I was working with, or USBtiny. Ob-Dev's has a more restrictive license to go with it, USBtiny may require some tweaking (I'm not sure). The USB enabled micros implement some of the USB functionality in hardware, though how much I'm not quite sure (there's a lot of reading). There are two options here, as well, the Atmel AT90USB646, and one of the PIC18F4xxxx series. The AVR is more expensive ($11 Vs. about $7 I think), and Digikey says it has an 8 week lead time. The PIC architecture doesn't really like C, and I haven't found a good, open source, or even cheap, compiler for the 18F series. Right now, I'm leaning towards using the USBtiny implementation. It has a less restrictive license, cheaper, and lets me use a high level language I like and am familiar with. Does anyone know of a reason not to do this?
VIRON nah8 years ago
Looks and sounds great. I don't know of a reason not to, except to wonder how long will USB 1.0 be supported, and if Vista dropped it already.
nah VIRON8 years ago
USB 1.1 is still supported as far as I know. If anything, Windows will continue to support it after it's useful lifetime. In truth, this project will probably be long since forgotten by the time USB 1.1 support is dropped. The way that the USB standard is written, and their "it complies" stickers handed out, will help to ensure I think. My understanding is that any High speed host has to support full speed, and any full speed has to support low speed.
nah8 years ago
Ok, I've been working on the code some, and I'll post what I have. It still has a long way to go, but there's a first whack at reading in and decoding the keys. It wont really "do" anything right now, but if you want to test out the key-decoding and can think up a good way to do it, that would be great. Or if you just want to go over the code and call me out on any stupid mistakes or things that could be done better, that would be good as well. Things to do on the key-decoding front: 1) Handle overrun events (more than 6 logical keys pressed at once). 2) Handle keys which should produce "shifted" characters. 3) Rework key code enum/keymap structure. There's going to be some problems, I'm sure, but I think this is a fair start. I'll try to keep y'all posted as to what I'm doing and how it's all coming along, probably put up new code every few days for those interested.
VIRON nah8 years ago
Mouse Wheels very likely work just like the wheels that read the mouse ball, two IR sensors detecting holes in the wheels and incidentally making gray code, which the chip translates into a code for the direction it "stepped" in Gray code which works like this... 00<->10<->11<->01<->00 I can't say "I AGREE" to the AVRUSB code. I'd look for generic USB keyboards and put a chip between the handpad and the keyboard chip, but wait, don't they have cheap legacy adapters for USB to PS/2? Again, maybe there is a USB keyboard so generic that it has orphan VID/PID numbers. Totally annoying that they replaced the "free" plugs with ones that need to be registered and ID-numbered. Anyway, either way, the computer should think it has a generic keyboard. How to program the user-defined key preferences? It's cheaper and easier than you might think to have talking firmware. In programming mode it says Push A, Push B, etc. and you push whatever feels like an A and B to you. Just need an 8-pin flash ram and a piezo-speaker, can be gotten for $2 or less. I'll find an example. Key matrix may be confusing with 3 or more keys but there is a solution, might not even be a problem if the controller has tristate pins. If more than 2 keys (excluding shifts) don't need pushing, it helps to scan the shift/ctrl/alt keys first. I'm not sure if the USB code is different than the ps/2 and AT code, but there were some weird keycodes having many extra bits. It will be a challenge to tap the USB port for the codes if they are unpublished.
nah VIRON8 years ago
I thought mouse wheels were optically encoded, but wasn't sure. Gray code wont be too hard to decode, but if we can't find a ready source for the wheel assemblies, it might be a trick to construct them.

There are silicon implementations of USB stacks. They run around $4US from Digikey. There are also micros with the USB stacks built in, the AVRs are about $12~$14 I think, and the PICs are about $7~$10. One of those options would be nice for a couple of reasons, such as adding the CRC(timing is too tight for it to be done in software) and leaving the uC with more "free cycles". I don't have a problem using some other option if no one else does.

I can't say I was overly excited by the license they put forth, but I don't know of any other free USB stack for the AVR. Also, it was going to be puslished here anyway, so I wasn't too worried about it. If I might ask, what don't you like about it in particular?

The VID/PID thing is annoying, if you're willing to take the risk you can just slap in a random set (or one belonging to some podunk company you'll never buy anything from.) They really wont become important to us until a driver is written. There are some people/companies who will sell you a block of 10 under their VID at around 25 Euro, I think. I'd have to look again, it's been a while since I checked.

Microsoft has a reasonably decent key mapper that can be had for free. I'm of the opinion that it will work well enough until we can get a driver with a UI specific to this project written up. I forget what it's called, but it's available in their "downloads" section.

Unfortunately, ghost keys may be a very real problem because of the chording (two physical keys -> one logical key). Software blocking, like most modern keyboards use, may be a bit limiting because of this. The other option is to use diodes. At about $.02 a shot, I think I'd go with the diodes.
VIRON nah8 years ago
I've seen what you mean where two keys can be one, even a different one, (not sure if this is what you meant, but where I observed it), if you decode with diodes, and have 6 rows, pushing a key from rows 1 and 6 will make a code for row 7. Such a keyboard has loads of trouble with more than one key being pushed. If you have diodes on every key,the problem goes away but uses too many diodes. I've seen keyboards handle multiple keys in different ways, such as a pair of CD4051's and counting to 0x3Fh and storing all the thusly electrically isolated rows/columns that are detected as connected, this works with 80 keys with a 4017 and 4051. (I tried to post that before but didn't see it post, but since you mentioned diodes ... seems too synchronistic.) Anyway, the problem is solvable, but the ghost key effect I had in mind was for example on a number pad, pushing 125 might read as 1245 because the 4 is shorted by the other keys in some methods of scanning a matrix, especially if only rows are scanned and all columns read at the same time. I would not only like to avoid EULA-like terms of using USB, but also have the keyboard work without a driver on a non-MS system like most keyboards already do. I use small ps/2 and AT keyboards for things that run on batteries and have nothing else in common with PC's. It would be a waste if I couldn't recycle PC keyboards for projects anymore. Yes it's a good idea to use a uC with a USB port and hardware CRC, how about the option to read the registry for your own computer's keyboard and enter it into the handpad? Otherwise I'd be looking for imported el-cheapo keyboards, and that might not even be necessary because keyboards may just already be just too generic to need drivers or ID, especially if they still use ancient PC-AT PS/2 codes. The mouse wheel has a lot less resolution than the mouse ball, and I'd say that because of that it's easily Makeable. It could have a brush-grounded half-circle of copper, and two contacts 90 degrees apart from the axle. That also would make the gray code. Mine feels detented, so I'd guess it was a switch like that instead of optical. It could also be a cam-gear and two buttons.
nah VIRON8 years ago
I think we're talking about the same thing with the ghosting problem and just not understanding each other. As for working with out a driver, I'm writing it up so that it will look like a standard HID keyboard. I've also written the report descriptor so that it should match the keyboard boot protocol. In English: it should work with any and every modern computer with no drivers. There are some caveats, however, which is why a software key mapper is helpful. The input report (data going to the computer) of the keyboard will be 8 bytes long, byte 1 is a bit field that contains the current setting of all modifier keys (Left and Right ctrl, shft, alt, and GUI) byte 2 is padding, the remainder are scancodes for currently depressed logical keys. By standard if there are more than 6 non modifier keys depressed all those fields are set to ERROR_OVERFLOW. All of that is a bit convoluted and technical, but what it gets at is this: if you want an '!' key, you have to send the scan code for they '1' keyboard key and have at least 1 shft modifier bit set in that report. That will also make any other keys in the report be shifted. This can be fixed in firmware, but it wont be fun. The other option is to use key mapping software to set the keys to have primary, shifted, etc. states as you'd like, and make the firmware simpler. What can I say, I'm lazy and key mapping software is available for free, so I like that option ;). I find it hard to believe that Linux and OS X don't have key mapping programs available as well. The VID/PID combo for USB devices is used to determine if a specific driver should be loaded for it. I'm on a laptop, so using the PID/VID combo from mine would be a Bad Thing™. If someone else has a generic USB keyboard, it's easy enough to find the VID/PID from it under windows, simply go to Control Panel->System Properties->Hardware->Device Manager, expand the "Human Interface Devices" item and double click on the keyboard. Now go to the Details tab, select "Device Instance ID" from the drop down box, there should be a line of text that starts "HID\VID_XXX(ampersand)ID_XXX" I believe the XXXX parts are the VID/PID for the item in hex. (So, the comments box here eats everything after ampersands, didn't know that before) I see what you're saying about the wheel encoder. I can't put one together myself right now, but I'll work on the assumption that it will encode as above, and pull the line to ground (unless someone knows a reason not to).
VIRON nah8 years ago
The two things that bug me the most are:
The complexity and proprietaryness of USB keyboards, and
The consequential excess of code required to deal with it.

I'm sure this situation doesn't help the fact that some **** politician
changed the daylight saving time schedule, as if all the clock chips
in the world are going to know, and not "bug out" like it's Y2K.

It is easier than all of this to reprogram a ps/2 keyboard's controller to
type directly onto a monitor without the PC, AND ALSO
to Say what key was pressed, as a toy for babies!
(If this is unbelievable, consider that the Odyssey2 and (Intellvision?)
1970's home game systems used the same i8048 controller as ps/2, and
not only had a keyboard but graphics and sound also!)
I mean, it's easy to make a handpad or any other keyboard, but
apparently a big headache just to put a silly USB wire on it!
VIRON VIRON8 years ago
Show and Tell, sort of, about some of my simple homemade keyboards.
With simple microcontrollers, One talks,
and a pair of them have displays and chat with each other.
(it's slow so you can see detail. one of the chatters has a weak screen.)
Keyboards Video
VIRON VIRON8 years ago
Speech Synthesizer
This uses samples of the voice of an obsolete robotic speech chip,
to make a PIC talk, but any other chip can do it too.
nah8 years ago
I've started to work on the code to scan the keys and generate codes. I'm planning on making it follow the latest map that Scrupulous posted, use the smallest AVR possible, and use Ob-Dev's software USB stack. This first run wont include the alts yet. I have an idea on how I'd like to do that, but I want to get the basics working first, then obfuscate it all a bit. On that note, there are a few questions to be answered. First, do we want the number are of keyboard to use the two-key combination system? Second, does anyone know if mouse wheel assemblies are available anywhere? If so, what type of connection and output do they have? If not, the closest thing I know is available is a touch sensitive strip (Quantum Research part # QT411, Digikey has it available in small quantities). Finally, I don't suppose someone out there has a VID/PID combo they wouldn't mind donating to this? I know I'd be grateful for it.
dev458 years ago
I ran this by my friend And he said I would suggest A software key mapping for this device. I'd recommend that you consider it for the prototype especially. You've had some discussions about support for multiple languages and Layouts. Doing this in hardware (even firmware) is possible, but alternate Layouts and languages are much easier to achieve with minimal software. This Device will be (pretty arbitrarily) assigning multiple simultaneous keys Presses to specific key codes in order to make it work with existing PCs. While you could do it in hardware, you are going to be messing around with Alternate layouts as you experiment with it. This whole project is a Relatively new game that combines frequency of character usage (in various Languages) on keyboards with ergonomics and limits in dexterity for the Average and non-average users. As such, I think you will be trying a bunch Of different combinations to balance learning and dexterity requirements Before you come up with a layout that is good for English. If you start in With other languages or character sets, the game starts again. Re-wiring Software is a lot easier than re-wiring hardware (even firmware). By the way, I don't necessarily mean a "driver" on the PC when I say Software. In this case, I would consider putting a microprocessor inside the Device with a protocol for downloading the mapping software over the USB Port. My two cents worth for the PS2 vs. USB debate tends to agree with "nah". Most New computers don't have the PS2 ports anymore. Everything has gone USB-2. I'd bite the bullet on that one. Now I have some CPU's and microprocessors if some one would like to work with me to come up with a prototype. If you can help with programming Email me at devin_coggins45@hotmail.com With "Keyboard" in the subject
Scrupulous (author)  dev458 years ago
Well, it looks like there would be at least two ways to put lipstick on this pig.

Your friend brought up some very good points. Remember, manufacturing and pricing would also play a big part in this, at some point.

I like the driver-less point of view. You could just click on an icon to run your keyboard program, as soon as your machine boots up (if that's what he's thinking). USB...this would make it very portable, yes?

Either way, could someone just carefully gut a keyboard for some juicy parts?
Gutting a keyboard is not hard but I do not have the skills to do the reprogramming needed. If we use an easy to get USB keyboard as a base I can write up how to get the hardware up then some one can work on the programming. Then all of us can test it by making the hard ware then loading the software. If I come up with a USB keyboard with the fist map hard wired could some one reprogram the chip to get the maps and multiple key thing to work?
nah dev458 years ago
I've been looking into the firmware some already. I was thinking to base it around an AVR and code it in C. Speaking of which, there's a good project here which shows some of what we will need to do. It's a good primer, if not exactly what we are after. Sadly, I'm not in a good position to work on hardware right now. I'm out of country until almost July, and I don't have any of my tools with me.

I'll also see about doing a design that can be built from scratch. The hardest part about this circuit will be building the key-matrix, I think, and we'll have to do that even using a gutted keyboard.
dev45 nah8 years ago
Nah if you can design it I can make it. Where are you located? I'm in southern Ontario Canada
nah dev458 years ago
Right now, I'm in the U.K. but I'm from the South in the U.S.
dev45 nah8 years ago
Ok then mailing is going to be hard till you get home.
Scrupulous (author) 8 years ago
Okay, this example is a bit closer to where we are at right now. I am showing this for reference only. (It would take a while to re-do the 3D models and the update the top views to reflect the changes.) For complete details on how we have arrived at this, one would need to read through the entire instructable at their leisure.

We have already agreed that the thumb line would be more perpendicular to the finger lines. This may put the ENTER key on the outside of the thumb line. I have added a scroll-wheel button to this image, on the outside of the thumb line, as well.

All of the keys may be tailored for ergonomics and kinematics. Again the !?,. keys were squeezed in late, but should be relatively the same size as the other keys. And, the finger lines can be adjusted slightly to establish a efficient layout, in terms of accuracy, dexterity, and comfort. I would love to say that the home row is AEIOU, but I don't believe that is the case.

This is merely one suggested arrangement for the primary layout. The alphabet is shown using the paired-letter functionality, and requires only 13 letter keys for the entire alphabet. This functionality may be appropriate for letters only. In this case, each of the other characters would have its own key. Of course, the @123 and plus keys would also need to provide the paired-letter functionality with the letters below them.
13-key alphabet.jpg
nah Scrupulous8 years ago
Looking at this reminded me of something I was thinking of. Do we have a list of what characters/keys we want to be available? Which ones should never change? How many custom slots do we want at a minimum? If we settle on these things it may help to flesh out the key placement, number of alts for the layout, etc. Similarly, what languages are we aiming at? Is English the only real concern for this version and other languages accommodated as they can through remapped keys? Should we consider how to make it simpler to use with other languages, esp. those like Chinese and Japanese? Along a slightly different note, what operating systems are we looking at supporting. I've been looking mostly into software on the Windows front, but what about OS X and Linux. How important is portability? I know all that's a lot to consider, but I think it will help us better work out how this should look, feel, and operate if we have at least some ideas about these things.
Scrupulous (author)  nah8 years ago
These are all good questions. I have added a blank layout here, which has most of the keys that would need to stay on each layout. I would stick with English for now, since our main challenge appears to be getting two keys to work as one without generating two characters. Other languages are good to consider. But at this point that may be biting off more than we can chew. I imagine other languages could be added easily as alternate user-programmed layouts, regardless of what key functionality worked best with it (standard or paired). This can be independent of the operating system. Yet, windows is probably suited for a working model. If another OS is more favorable to you, then that is fine with me. Portability would be a definite plus. I know this is a lot to take in. It can be really easy to get overwhelmed. But, the main thing we are concerned with is getting one layout to work, while allowing it to accept one or more alternate layouts. Remember, those other layouts can be labeled on the keys, too. The layouts themselves are very flexible. You can print out a few blanks and try a couple. As I see it, the really tough part is being able to assign the paired-letter functionality to a specific subset of keys (for any given layout). Regardless though, sometimes it's best to just take a step back, a give it a rest for a while. More often than not, you come back with better ideas...or just plain solutions. ;>)
blank layout.jpg
Any thoughts on how to do the two keys to a letter? In trials on a paper key map I found that the end keys are easer but I have long skinny fingers Arrow keys would be good.
Scrupulous (author)  dev458 years ago
I had kicked around the idea of a two-position key.

It could act on a single shaft axis.

Or, it could be something like a hat switch, where pressing it gets one character, and pressing-and-joggling it gets another character. The "key" would naturally center itself, when you release it.

Now is a good time for me to disclose the idea of pressing any kind of key, for a particular character, and holding that same key to get a different character. The simple act of holding a typical key down is usually thought of in the "electric typing" world as repeating a character.

What I am revealing is, in one example, a way to capitalize a letter merely by holding its key in the pressed position. This is done without the SHIFT key or CAPS LOCK key. Tapping a typical letter key like K gives you a lower-case k. Pressing that same key and holding it gives you an upper-case K. (Where holding a letter key down would normally cause that letter to repeat over and over, now it simply generates the upper-case of that letter. This can be applied to non-alphabet characters as well.)

Another example of this idea is a kind of smart-correct function, where the user presses-and-holds a particular button and the cursor returns to the position immediately following the last prior instance of that held character, while erasing everything after it. The user then continues typing normally, et cetera.

I am not necessarily suggesting that we incorporate this idea into the current project. I am simply publicizing it now because, as the game goes, my hand is admittedly forced. I wouldn't mind including it, since it's out of the bag now. But, I understand that it adds more to the process.

Scrupulous (author)  Scrupulous8 years ago
Oh yeah, arrows keys would be good. Earlier in this instructable, we talked about flush-mounting them in the central palm area. But, if that turns out to be a roller, then anywhere they'd fit would be good. Who knows...maybe a roller arrow? (I'll stop ;>)
nah Scrupulous8 years ago
I don't know that I see a need for dedicated arrow keys. If one of the alts has them in a reasonable place, I think that's good enough. I say this because I don't use them when I'm typing, don't know of anyone who does, and can't think of a good reason to, especially if you have your other hand on the mouse. There is something to be said for limiting the complexity of the keyboard. To be honest, I think I would cut back on the number of keys from what your example above shows.
Scrupulous (author)  nah8 years ago
I agree. I almost never use the arrow keys. I think some traditionalists would say that you need them for BIOS setup screens (as well as the F keys). Then again, nothing about this need be traditional, necessarily.

As far as the number of keys, whatever works. We might as well plan on as many as can fit on the fingers, using rectangular keys. Standard row spacing is about .75 inch, on center. For our purposes, that would probably be closer to half an inch. I could also do without anything unecessary. We'd want the primary layout to accommodate as much of normal typing as it can, though.

I know that's a lot of hardware for a prototype. But, we could get away with only the eighteen required keys to make the whole alphabet, with the paired letter functionality.
nah Scrupulous8 years ago
It's not so much that I don't think there should be arrow keys or F keys, on the contrary, I think that they must be included so that you don't have to have a separate keyboard in case you need to reinstall Windows, or mess with the BIOS, or what ever. I just don't think it needs to be on the primary layout. This is something I had mentioned a little before, as there is a way around it. If every alt generates a separate set of scancodes, and this is handled all by the keyboard hardware/firmware, then you can have access to those keys when you need them. The drivers would simply be "value add" functions, like the on-screen display, changing how the alts work, and remapping the keys. The characters I was thinking should be on the primary layout were the alphabet, space, enter, esc, tab, shift, ctrl, alt, period, question mark, single and double quotes, and comma. Those are, for the most part, the keys I use regularly and with out breaking the cadence of my typing. In my mind, anything less than that would not make a keyboard I would use day to day, and I would carefully select any additional keys so that it filled out the keyboard to a comfortable level. Of course, the hard part is knowing what "a comfortable level" is, especially before having tried out one of these.
Scrupulous (author)  nah8 years ago
I am right with you on those keys. And I also agree that comfort dictates the number of keys.

One thing I have a particular preference for is a number "pad" on the primary layout. If it can fit, I say throw it in. I have never appreciated ever having had to push a NUMLOCK key. But, since it seems you would take the lead on the hardware/firmware approach, whatever way you think is best is okay.

Yeah, it's gotten to the point where I would love to try one of these out.
dev45 nah8 years ago
I was thinking arrow keys for people that do gaming [nah] you are right. The hardware/firmware can do all the work but I'll need help with the firmware if I could edit it could made with out relays This also opens all operating systems from ms-dos to Linux.
nah dev458 years ago
If the keys are rectangular I think it will be easier to press two at a time, as opposed to being round. As for the actual switch, there are many options. Probably the best options are SMT tactile switches, 3 layer membrane switches (like in standard keyboards), or capacitive switches. Each has advantages and disadvantages. Getting one scancode out of having two keys pressed is, in principle, easy. You simply scan the key-matrix to see what keys are pressed and store this in an array. Once you've scanned them all, cycle through the array looking for pairs and singles, generating the scan codes as you check and send them to the computer. There's a bit of hand waving in that description, and there will be some tricks in optimizing it, but nothing insurmountable, I don't think.
Are you using {c} and {e} to make d?
Scrupulous (author)  dev458 years ago

Pressing C and @ make D (in this hypothetical layout).

What you're suggesting may work, too. But, it would seem to be more difficult to learn, no?
nah8 years ago
I've been thinking about the keys themselves and was wondering what people thought of using a capacitive touch sensor in place of traditional switches. Quantum Research Group produces a series of ICs designed for just such a task, and they aren't that expensive. A 48 key chip is a bit under $10 from digikey, with 32 and 24 key ICs being somewhat less in single unit quantities. I think it would give the keyboard a really slick look, but I don't know how it would feel to type on. Has anyone tried a keyboard like this before? What do people think about the idea?
dev458 years ago
Ok I'm not a programmer so pleas take this with a grain of salt. I've read this list of ideas and like it vary much but I would consider reducing the software to optional. If you use [nah] idea of a wheel on the keyboard this would work nicely plus it will allow you to use a hardware method of changing layouts using a microchip or something of that nature. This allows you to use hardware making it easer to get to work with all programs and operating systems and save you from plugging in a normal keyboard for BIOS work. Also I'm concerned with how much you can send over PS2 but then you can go to USB And a second point I'm like to mention is in nah's lay out it requires pressing two keys with one finger would it not work better horizontally, or were the keys close enough to easily hit both together. If you have a key map the hardware can be made fairly easily. Just my two cents let me know what you think. When it's finished I'm vary interested in trying one.
nah dev458 years ago
The image I posted was just an example of general placement. The shape of the keys, exact location, and type of switch would need to be adjusted to make it easier to press two keys at once. That shouldn't be a real problem, on my laptop's keyboard it is easy enough to press two keys at once, even with my pinky, and it was designed to make that less likely. PS2, while it is simpler, is a legacy port. It's not really a good option to develop new hardware for because it may not be on any given computer, and is not on any laptop I know of. USB is a bit harder to work with, but there are implementations available that are software only for micro controllers, or there are silicon implementations. I've been looking into drivers under Windows, and it looks like if it were USB, it could simply work as an HID keyboard. A relatively simpler filter could be written that would allow keys to be remapped, and communication for the display to take place. This is better than using the standard Microsoft mapping system, as it could be made to handle more advanced hotkeys and macros. This could be done for a PS2 keyboard, but it isn't really that much more work to use USB.
dev45 nah8 years ago
Nah you have a good point but at current ps2 and the keyboards for the old 386s. I'm just wiring to a keyboard card so the same thing applies for USB I do see that it would de easier to map key by software rather than hardware but I'd like to get it working first as for a driver the bios will pick it up all ps2 keyboards are the same as far as the computer is concerned except the net boards witch need a windows driver. (Ok this is a little over explanation sorry in advance) I like the key idea it just took me a little time to get it. But the implementation on a prototype is hard if not implemented in the software witch leads back to usb. I'm going to run this by a programmer friend.
Scrupulous (author)  dev458 years ago
Thanks for your input dev45, Yeah, we have been in the process here of narrowing down the primary layout to a good key map to start with on a piece of hardware. I think nah presented his layout for discussion purposes. I not sure how you would put the pairs horizontally, because you need the next key(s) to continue the functionality, down to the end of a given finger line. The idea already has been to put the keys close enough to easily hit both together. If you get a chance try to read through the whole instructable. I know it is getting long, but it is also pretty interesting. If you would like to contribute, then maybe you are the right person to throw a working model together. If you say that's it is much easier to keep the wheel on the board, then let's go with that.
dev458 years ago
Ok I dug out my ps2 keyboard that I mapped last year as part of a different project.

I'm looking at a key map, do we want it in the shape of a hand or can the thumb, first and ring fingers be widened at the end for more comfort ?

If you would like to know how I'm going to try this see the link.

it may be for a flight simulator but it should work for this to.

Flight simulator keyboard hack
Scrupulous (author)  dev458 years ago
For accuracy, all the fingers may be widened toward their ends so that larger and larger keys can be used, the further away you get from the palm. I'm not sure what you mean by widening those three fingers specifically for comfort. The housing, itself, would not need to have fingers shaped into it, necessarily. Although, there has been talk of relocating entire fingers to make it usable for either hand. But, to make a physical prototype, you could use any convenient shape for the board, as long as the finger lines extend radially from the palm, and as long as the letter keys are arranged in alphabetical sequences. I suppose you could try horizontal (left-to-right) sequences. But, that may just eliminate the thumb from the functionality, or at least require more than one finger for certain characters...probably not as good of an idea. I want to thank you in advance for any work you do decide to contribute, though. I can say that I would consider a mechanically working model a substantial part of the overall development, even if you left the programming up to someone else.
Sorry. I must admit I have trouble expressing my self clearly I'm good in any subject other than language. I meant to say a widening at the tips of the fingers for two keys on the first row this seams to feel comfortable on the three fingers I mention particularly on the thumb. As for software if it goes as I plan the computer will see it a a standard 101key ps2 keyboard. Then its yours if you would like the on screen stuff. Just need a lot of relays if any one knows a good place to get logic microchips pleas let me know it would make it easier. I may have relays to cover 2 key maps probably not any more.
dev458 years ago
It's a lot of relays. let me finish my college applications and I'll see if I can mock up a prototype and wiring diagram.
Scrupulous (author) 8 years ago
If anyone hasn't read Step 7, please make sure you do that. (I removed the image from that step, because the final product hasn't been established here yet.)
Scrupulous (author)  Scrupulous8 years ago
Note: Step 7 has now become Step 8, by the insertion of a new step 1 (Sorry guys, last one. ;>)
nah8 years ago
I think you might should reevaluate the ergonomics of the layout, but I suppose that might be best done after having a working model. I would suggest one change which could effect the coding and circuits heavily. I think using the mouse scroll wheel would be a mistake. That particular doohickey is already used extensively, and well. Trying to change it's use to controlling the keyboard would be, well, wasteful. As an alternative I would suggest either a scroll wheel or hat-switch built where the thumb of the typing hand can easily reach it. For example place it on the outside edge of the thumb portion of the board. Add to this the use of indicators on the board for the different modes (i.e. an LED at the tip of each finger which switches on for a given mode) and I think you end up with something much more usable. Two other quick things to think about. Having the alternative layouts on screen may seem nice, but I suspect it would quickly become irritating. A the very least it would need to be optional. Also, having modifier keys where only one finger can reach them is problematic at best (think of capitalizing a 'w', or using ctr-alt-something else)
Scrupulous (author)  nah8 years ago
Very good input.

Yes, it would be best done with a working model. A few key points on that (honestly, no pun intended):

1 The thumb line can be more horizontal, no problem at all.
2 The keys shown in the model are shown primarily for relative positioning. They may be shaped and located to facilitate the kinematics of each finger, to the hilt.
3 Believe me when I say that this is pretty close to an optimum physical arrangement, given particular constraints which cannot be deviated from (in terms of intellectual property). I will continue to steer this development accordingly. Please don't assume I am being stubborn, as many of my considerations (expressed or otherwise) are absolutely necessary.

Having said that, allow me to insist that we divorce ourselves from some of the standardized ways of looking at keyboards in general. Our 101-keys are a grandfathered exercise in arcane thinking. Nothing about a staggered grid-like or matrix-type layout is either natural or intuitive to the human hand (or the human mind). I'll let you go from there.

As for your other suggestions, I accept your reasoning on the mouse issue. I also embrace your approach at putting a wheel on the keyboard side. As you say, outside the keys, right about where the big knuckle bulge on the thumb is, is where I would put it. I see the wheel turning on an axis that is perpendicular to the thumb itself (the wheel is parallel to the thumb). A hat switch may be even better, as I believe it can act as a button also, and centers itself to neutral position. I would need this to be treated in the same fashion as the wheel button found on mice. And I would expect it to comply with one of the four available options (Steps 3 thru 6)

I like the LED approach in conjunction with the screen aspect. There is no problem with marking each key with more than one set, each of which can be color-coded or position-coded with respect to the LED color, or to the relative location of a particular LED (active layout indicator). But, I could not limit this project from using the screen aspect. One reason is that this allows for eventual user-programming and the need for additional labeling scenarios that it may entail. There are others.

However, I don't have a problem with making the screen aspect optional. If it poses an undue burden on you to involve the screen aspect, nah, in regards to your further contribution, then go ahead and dedicate one of the option steps to developing your LED approach. I would much rather you included the screen aspect along with it, if you could.

One thought on the final comment in your post. Let's be honest, in the real world, we all take a look at the modifier keys (as you called them) from time to time, when we are typing. We all take our hands off of the home row and do funny things with the "wrong" fingers every once in a while, in performing certain program functions, and/or to use the mouse at the same time. This whole project is partly intended to eliminate that unrelenting ineffectiveness, or at least mitigate it.

Trust me, there are ways to input characters efficiently, and by providing new standardized typing methods, that don't even resemble the way we were all taught with a typewriter. If you use your thumb to press the small w and your index finger to press the capital W, for instance, it wouldn't be the end of the world.

Believe it or not, the human hand and mind are capable of much more than is required by the routine of the QWERTY-style keyboard, or any bound derivation of it.

By the way, I hope you can stay with this, because it sounds like you could be a key contributor. Again, no pun intended!
Scrupulous (author)  Scrupulous8 years ago
Note: Steps 3 thru 6 have now become Steps 4 thru 7, by the insertion of a new step 1
Here's a whole nother idea with the two hands idea. One of the hands is just letters. And the ball in the middle should roll so you hand hoes up and down more smoothly. On the other hand would be a mouse, and keys such as space, enter, shift, control, etc.
Scrupulous (author)  Weissensteinburg8 years ago
I like the idea of a rolling palm support. It leads right into the idea of a trackball on the keyboard side.

What you're suggesting with the mouse and keys is essentially an expanded form of what I am presenting here. At the moment, I'm not interested in a second keyboard of any kind, on the mouse side. This project may develop somehow into that, however, but I need to insist on the four options made available in this instructable (Steps 3 thru 6). Start with one of those, and go from there.

You're thinking is not much of a stretch from any one of those options. Typical controls, such as SHIFT, CTRL, ALT, et cetera, can be regarded as alternate keyboard layouts in the steps. If you can make it fit within that context, then have at it.
Scrupulous (author)  Scrupulous8 years ago
Note: Steps 3 thru 6 have now become Steps 4 thru 7, by the insertion of a new step 1
I said that wrong...I meant it like you said, just adding to those ideas
I understand completely. You're half right about not knowing which layout is being used. It's a bit of a pain on public/shared computers, because XP is stupid, and has separate language settings for each window, so you have to change settings everytime you open a new window, else you're typing jibberish. On my own computer its ok, I've set dvorak to default. Only annoying thing is, the cut-copy-paste hotkeys have moved wildly across the keyboard, no longer are they nicely in a row close to the ctrl key. I've pretty much forsaken qwerty, but I can type both independently. So as long as the language settings are right, I'm fine.

And in light of this new development... Cool! I saw something using the same "key in-between keys" design a while back. They used it for the world's smallest keyboard, (smallest back then) I can't seem to find a link for more info cause it's probably old-school technology now. I think they based it around a touchscreen panel, with software to work out which key was pressed. Of course, button design will have to be CRITICAL for this to work.

When I first looked at your diagram, I saw six rows, and I immediately thought. Ouch. That must be a pain to type with. This is definitely better. Less movement = Less Ouch.

Though I don't believe that the alphabetic layout will be easier to learn than any other. Personally, I reckon learning to type is kind of like conditioning. In that I think "b", and my finger moves a certain direction to B, and I do this so many times that the letter becomes associated with the movement. So with practise, typing becomes nothing more than muscle memory.

Irrelevant of the layout, to become naturally familiar with the keyboard, you will have to spend time consciously thinking about key positions while typing, thus causing you to slow. I think that it's this process of actually thinking that slows you down. For about two weeks, I knew where every letter was, but sometimes I'd get a mind blank, and have to think "B? Where's B... Oh. It's here."

I simply think that to learn a new keyboard layout will be extremely painful from the words-per-minute side of things, irrespective of how intuitive it is.

Scrupulous (author)  infiniteregress8 years ago
I'm with you on those thoughts. I also agree that the button design would be very important. It seems like the edges of adjacent keys would need to match up well with each other. The central portions of the buttons may be either lower or higher than the edges. Maybe not. Hey, let me know if you come across that reference to that "key in-between keys" design.
Scrupulous (author) 8 years ago
Hey nah, this is a great topic to include somewhere in this collaboration. And, I don't have a problem with adding step, per se. Just be careful to observe some of the effects of inserting steps. In this case, everywhere that I've mentioned Steps 3 thru 6 for the four optional paths would now need to be edited also. (But, I'd rather not have that done.) Maybe this topic could be announced on the intro page, and be directed toward one of the other existing steps.
nah Scrupulous8 years ago
D'oh! Sorry about botching the numbering, I didn't even think of that. I assumed it would pass it by you before posting, but it didn't seem to. I also wasn't really sure what was considered good etiquette on these things, didn't mean to over step my bounds or anything. I'm not sure I follow about the topic being "direct toward one of the other existing steps." The only reason I added it was to have a place for things that were applicable to all the other options, and thus did not default to fitting under any other individual step.
Scrupulous (author)  nah8 years ago
That's alright. Let's keep this step right here. (I'll go back and edit the references to other steps accordingly.) I will also edit your step to mention that it was added-by-edit, by you.
I'd like to add something to this, that's VERY important for this layout. DVORAK. To those unversed in the art of dvorak, it's an alternative keyboard layout to QWERTY, which is something like 60% more efficient or something. You should all do some research on it, because this dvorak guy spent like 3 years on the perfect keyboard layout according to the letter patterns and frequencies of the english language. It's a very good layout. I use it when I'm typing up long reports and stuff. Much more comfortable. You can do some analysis of your own on the frequency of letters pretty simply... I think it was said somewhere that in dvorak's research, he found it to be easier/faster if people typed with alternating hands. Also, its easier to "roll" your fingers from the outside in, rather than from the inside. ie. tap successively using your pinky, ring, middle, pointer, rather than the other way round. I felt this was URGENT. because i have very long fingers, and reaching down is a big pain on normal keyboards. (which is why i switched to dvorak!) if anyone's wondering, took me about a month to learn to type with dvorak. feels real sweet now...
Oh, I didnt realise dvorak had been mentioned. but yeah. Did a bit of research, and found, from dubious sources. ie, wikipedia. that the letters from highest to lowest frequency in the english language are: etaoinshrdlcumwfgypbvkjxqz which is mostly useless. using this for a first draft layout, lets go with from top to bottom, pinky column: nelf ring column: stcgvx middle column: hauykq pointer colmun: rompjz thumb column: diwb the second row is the home row, right? so the most used keys go there. starting from the pinky, cause it's easier to roll inwards. then we continue along in order of frequency, assigning them to the top row, then to the lower rows. I realise that the thumb is less dextrous than the other fingers, and that certain couplets occur more frequently, and this will probably be factored in at a later date. cheers
Scrupulous (author)  infiniteregress8 years ago
I take your word for it, that the Dvorak is a better layout than the QWERTY. Yours is the only personal account that I have ever heard. The increase in alternating sides makes sense to me. Maybe that can be thought of in terms of alternating fingers somehow, in a like manner.

I'm not sure I see how rolling the hand inward is easier than rolling the hand outward. To me, outward seems more powerful. And, it may just be a personal thing.

That's the inherence of vouching for a particular layout. Everyone has their own take on it. It would be very difficult to experimentally resolve the value of any layout over the QWERTY, because it is so well established. Asking someone to learn some other layout after they have learned QWERTY, gives the the new layout an unfair disadvantage. It would also be tough to find a true control group (people who have not yet learned to type) that is large enough to be statistically useful. For all practical purposes, QWERTY is the monopoly.

The challenge would be to present something much different, and better. As favorable as the Dvorak may be, it's no surprise that it has never taken hold to speak of. But, the benefits of the Dvorak can certainly be applied to this project, analogously, with respect to letter frequencies and strength/comfort/dexterity.

My primary layout, for example, is arranged to take advantage of mental memorization that has already been accomplished. This actually gives it a leg to stand on, especially if you were to go and expect an employer, a writer, a programmer, or whoever, to seriously consider the plausibility of switching to something else permanently. There would have to be something in it for most people.

I'm sure you are aware that your suggested layout is one of many possibilities. That's where having the option to customize a layout, and having it display on the screen when that layout is active, comes into play. I wouldn't be interested in selecting something other than the ordered scheme I've presented as the primary layout, for one because the chances of anything else finding a commercial niche drops way off.

I want to encourage you, though, to develop your approach here as much as you need, because it wouldn't necessarily affect the programming. I hope you also stay on board with this because your input will probably prove to be very valuable. I have something to disclose shortly that relates to some of your considerations, and may very well be to everyone's liking.
Scrupulous (author)  Scrupulous8 years ago
First, the reason why I arranged the primary layout alphabetically and by vowel is to give a new user something to start with, a way to ease the transition from a QWERTY board, so that productivity isn't sacrificed to an unreasonable extent. I would imagine that one of the ongoing difficulties with the Dvorak board is that sometimes you forget which board you are using. That is specifically one of the reasons why I am distancing this design from any form of a grid-type layout, which includes both QWERTY and Dvorak. This departure is where the actual hand design comes in.

Ironically, the reason why I didn't immediately balk at your first draft is because you've managed to keep an even number of letters with each finger. Your comments and input on Dvorak also got me thinking about something that might just work for everyone. Hear me out, please.

When people say the English alphabet, they can use a sort of 1-2 cadence. It goes something like a-b, c-d ... s-t, u-v, w-x, y-z. Now, if you think of those 26 letters as 13 pairs, you can cut the number of keys needed by half.

So, let say the pinky has A at the bottom and C right above it. To get B, you would press both A and C. And, across the board, by finger you have AC, EG, IKM, OQS, UWY. To type the "missing" second letter of a given pair, you simply hit between its two adjacent letters (pressing both).

There can be a punctuation key, or a number, or whatever, located one key beyond the last letter key on each finger, so that its paired letter may be generated also. This requires a maximum of 18 keys to access the entire alphabet, without the use of controls or alternate character sets.

If home row is CGKQW, this means that, with one hand, the furthest a letter will ever need to be is a distance of 1.5 keys away. This is fairly remarkable, considering the fact that a standard two-handed board cannot do any better.

Now, you might say big deal, why not just make the keys much smaller, and use all of the letters? Well that is a much different situation. You would be requiring human fingers to be constrained beyond their limitations. With the flexibility of using multiple adjacent keys (which normally results in typing errors), you are giving your fingers the freedom they have always needed.
nah Scrupulous8 years ago
That is a very interesting concept. I've played with it, and found that with 18 keys you can get the whole alphabet, enter, space, and period and have two split-key combinations left over with 2,4,5,4,3 keys for each respective finger from left to right (I find my pinky has a more limited range than my other finger, so I only put two keys there, but this may not be the best way in general). There are some problems with the layout I mocked up (I've attached it for reference) but this does shows promise. I think for it to work best, alphabetical order may need to be tweaked or abandoned. I say this because you end up with "e" not being on a physical key, but it is the most used letter. It might be worth while to consider placing keys by their frequency of occurrence, and not alphabetical order, if for no other reason to see how the two compare. I admit this may make it harder to learn, but the long term gains in efficiency may out weigh this. Which letter is on which key is, however, trivial and should be changeable in software with few problems.
I have the nagging feeling that this discussion would better be carried out in the keyboard layout page... so I'm posting a reply there. cheers.
dataphool8 years ago
Twenty years ago, IBM developed a keyboard meant for an operator with only one hand. That is, the keys were ergodynamically arranged for an amputee so that she could keyboard / type nearly as fast as someone with two hands. This keyboard was expensive but worked quite well. I think you are trying to reinvent a pencil. Or a mouse to government specifications that resembles an elephant.
Scrupulous (author)  dataphool8 years ago
If it is US patent #4,042,777 you are referring to, that has very little to do with what we are developing here. But, thanks for your concern.
nah8 years ago
I'm not sure if this is the best place to put this, but it makes sense to me so, here we go. I was thinking about what all the alternate keyboard layout system entailed, and realized that some care needed to be taken in it's implementation. A potential problem is using keys before an OS (and thus drivers) has been loaded. An example of this is accessing the BIOS, which may require a function key to be pressed (i.e. F2, which I think is fairly common one.) Unfortunately some systems have been known to require nearly random key combinations. There are two solutions I can think of to this: 1) Have the keyboard default to generating the F-keys and a few other select scancodes. 2) Provide for the alternate layouts to be accessible with out the need for drivers. This would require that the keyboard be able to switch layouts prior to any drivers being loaded, as well as the standard scancodes for all keys being generated . The first option is technically simpler, but poses the problem of selecting the right keys. The second option is more difficult but ensures compatibility. I would say that the second option is the better, but it does preclude the possibility of using the mouse or some other device as the only way of cycling through layouts. If there are any other thoughts or comments on this, I would like to hear them.
Scrupulous (author)  nah8 years ago
If the layout switch is integrated with the keyboard, all required keys would be accessible before any drivers are loaded (if it can emulate a 101-key), correct? In this case, the additional programming could then detect the initial position of that on-board switch, and provide an appropriate screen display if necessary. With your key labeling scenario, additional actions may not even be required once the driver is loaded. But, maybe a way to allow a second device to interject with the layout controlling would be good. Is that what you're thinking? By the way, the screen layout wouldn't need to be so big. And the previous display wouldn't need to be lightened anywhere but behind the hand. (I know, that might be the last thing you want to concern yourself with.)
nah Scrupulous8 years ago
That's part of it, but in someways it gets a bit more convoluted. I've attached the USB document that gives the standard scancodes for different keys (page 54) as well as if they are present on "boot" keyboards. The scancodes of our keyboard need to at least partly match that standard. The convolution comes in if we try to used the keyboard HID. They scancodes provided are based on physical location of the keys, not so much what character(s) they are associated with. What's more, every key always sends the same code, regardless of if a modifier key is being held down.

I'll give an example which will, hopefully, help things make sense. Lets suppose that one of the alternate layouts has a key designated as "/" and one as "?". For a normal American system you would need to have 0x38 send for the "/", but for "?" you need to set either left or right shift (0xE1and 0xE5 respectivley) then send 0x38, then unset shift. Also, if you hold the shift key, and press the "/" key, you will get a"?".

There is something of a solution, and that's write custom drivers for the keyboard layout but use the same scancodes as the USB standard. I think this can be made to work in a pinch sans drivers, but the keys wont be labeled anywhere near correctly, and may behave oddly. It may also be possible to initialize the keyboard with HID drivers and set to work (though not optimally) with them if custom drivers aren't present. I'm not well enough versed in USB to know if this is, or is not, correct. I hope someone who knows more can set me straight.
Scrupulous (author)  nah8 years ago
I really appreciate the progress you have already made. If you can get this to work, you can consider yourself a part-owner. But, to complicate things even more, could you take a look at my recent response to infiniteregress, when you get a chance? It details a fairly unique strategy.
VIRON8 years ago
A 4x2 intuitive code: Note 4x2 is also ok for nonintuitive codes, (morse,braille, binary-hex,bcd, etc...) it may use 8 buttons or only 4, even if all 26 are installed. (I have nothing against DVORAK except this keypad is the wrong shape, and in such cases I prefer ABC order, as on a 5x5+1 keypad) Either try to draw all 26 letters or find a font that does so more or less in a way that may appear on a calculator, the letters M, W, and X will be the most "trouble". Other letters may be writable as either upper or lower case or both, sometimes even a third way. Entering a letter would be pushing first the four buttons for the vertical segments. A nice arrangement may be in this order fingerwise and simultaneous. upperleft,lowerleft,upperright,lowerright. Second, key the 3 horizontal segments PLUS a disambiguator. The meaning of the disambiguator is the presence of diagonals. It changes O to Q, P to R, U to W, n to m, H to X, t to k, and so on, in most cases the horizontals indicate where the diagonals appear. Capital Letter "I" may need disambiguity without diagonals, because this is so far defined as a 4x2 system where the possibility of 4x1 definitions not yet considered, so at least 1 key must be pushed. A nice order for the second set of 4 would be: Top Middle Bottom Diagonals-disambiguator. This is an optional 4 button 4 finger alpha code that may be used by anyone who finds it conveniently intuitive.
VIRON VIRON8 years ago
Clarification of the words-on-a-calculator concept:
Scrupulous (author)  VIRON8 years ago
I should have put my last reply to you here. Press on!
VIRON8 years ago
I don't think I saw this here but it popped into my head, since I make keyboards and control panels, some very unusual. Since I've made this one a long time ago it's got prior art in the public domain, consider... oh how do I explain this... Like once I made a calculator with only 7 buttons, arranged the same way that the lights on the display are arranged, and numbers and functions, most of which are obvious, were entered by making the symbol with the buttons. What I am suggesting is what might have been in "ancient" shorthand script, some intuitive encoding of the way a letter looks. I can see a relatively vague pattern of that sort in morse code which only has one button, but what I mean is, let's see, how would you describe the alphabet to a brain in a jar? OQDPRA have one hole in them, B has two, so maybe all of these could be assigned to one button and one finger. How many lines horizontally and vertically perhaps? H1:THL V1:TILPEDFJB V2:HUN V3 or V4:MW Ok, this doesn't seem intuitive enough. 5 buttons can do the whole alphabet, but what's a good pattern? We need 5 traits that we can describe a unique letter with. -made with one curve? SZVILMNOPUWDJC -made with 2? QTX...no good, the best set would have at least 5 letters per finger, have 4 obvious traits and 1 disambiguator I imagine. Perhaps we draw the letter like I made numbers on my calculator... Maybe a really small (Rubik's) font with 9 pixels/buttons? _A_ AAA A_A Telephone code? 1__, 2ABC, 3DEF 4GHI, 5JKL ,6MNO 7PQRS, 8TUV, 9WXYZ,0___ Maybe I'll do a quick shorthand research. Stenography? would be somehow made very easy like this, probably not like Braille. Should be easier than Palm Pilot Grafitti.
VIRON VIRON8 years ago
If you look closely at my avatar the red lines are in my weird calculator code.
Scrupulous (author)  VIRON8 years ago
Great minds think alike, apparently. This is something I had thrown together a while back. I can't remember why I didn't have the X, and I think the I needs to be on the left side. The Y and the Z could stand to be transposed. But, I made sure not to replace numbers. Naturally, this is a single-case font.
SevenSeg Font.jpg
Scrupulous (author)  VIRON8 years ago
Your logic is impressive. This makes me think of American Sign Language, and how every letter is symbolizes by a consistent shape. I don't know enough about Braille, Stenography, or [Gregg] Shorthand to know how they might apply.

I would definitely vote against the phone code as many times as I could. To me, that is the antithesis of human intelligence. ;>)

In my research, I came across a US patent for a five button scheme. This guy/gal had arranged the buttons in a half-circle and, from what I remember, devised a semi-intuitive graffiti-type alphabet with the 32 possible permutations. The letters that the user was supposed envision were each a skewed product of connecting the dots made by the pressed buttons (drawn through a central dot, which was not a button). It was a little too cryptic to be of any use, IMHO. But, it's something to keep in mind.

It occured to me after reading your post, however, that you can use the speech sounds of spoken letters to establish a scheme. For instance, B and P sound very similar, except for breath control. K and CH and CK sound identical, according to a few simple rules. U and W are related by their initial mouth action, and differ by total number of actions, or so on.

With this, it seems clear that we are still very much in the brainstorming phase in some ways, while certain programming aspects are being worked out at the same time. This is turning out to be a very interesting collaboration.
Entity_X8 years ago
K, hope this ends up in the right place. AND, sorry I didn't read all previous comment so expect some redundency here. Couldn't help myself, I got too excited about the prospects for this project. might want to build a quick working model a submit a patent before some "unscrupulous" rip off artist locks it down so they can profit from a crappy knockoff. a the current rate of decline patent law may forbid public discourse over existing patented items of course they would make sure they could steal our ideas. sorry i seem to have gone off on a tangent there. 1. Would it work to integrate a trakball in the palm position. I think a two hand model might work better, that way: a. could be adapted for right, left, or ambidextrous persons. b. one hand could integrate tracball while other could integrate a joystick control. c. would allow for a design where the the keys would be reachable with significantly improved hand stasis. 2. thumb keys should be placed in line with the natural thumb movement arc. I mean on the side rather than top.
Scrupulous (author)  Entity_X8 years ago
Hey EX, thanks for your input. The enthusiasm is well-appreciated, too. When you get a chance, briefly read thru all of the steps, then the comments. Just about everything you mentioned has been addressed, and the remainder of this project is simply waiting for interested people to stake a fair claim of the ownership. So far, we are leaning toward a device that can be switched from left to right, by detaching one or all of the fingers and relocating them. A trackball sounds good, even if its "dummy" and only serves to articulate a rolling support for the palm. It is important to leave one hand free for using a mouse full-time, or some other device. The primary layout accommodates probably more than 95 percent of normally typed English. And, alternate layouts are free to arrange other character sets appropriately. Some control on a second peripheral device should be reserved for activating alternate layouts of this keyboard. That has been initially assigned to a middle scroll-wheel button on a mouse. There is talk of a hat switch of some kind to be placed on this keyboard, itself, to serve the same function. The screen layout is necessary for additional user-programmed labeling. If I've left anything out, it can be found among the comments, of course. At this point, though, I think more focus could be directed toward the programming.
Scrupulous (author)  Scrupulous8 years ago
By the way, I don't think the potential use for this device as one big "digitally fuzzy" analog controller taking full advantage of the hand shape, which could be applied to game POVs, 3D-CADD, virtual reality, etc., has even been recognized (except by me) much less explored.
MerleCorey8 years ago
didn't read this one...just commenting on the fact that it is a LEFT hand...makes me wonder why the right hand is too busy to type...
Playing the piano, using the mouse, or dare I say it, pocket pool, take your pick.
Seems to me like the "fingers" of the keyboard would need to be closer together. I'm trying to imagine spending hours with my fingers spread apart like that. Basically, the user need to be able to reach nearly if not all the keys without moving their palm.
Scrupulous (author)  bill.schuller8 years ago
Actually, it's pretty close. Watch your own hand, as you curl and extend your fingers. They naturally extend radially, at least mine do. As I responded earlier, the thumb may be better off along a more horizontal path, but not by much. I specifically shaped the keyboard for a first impression, and that may not change. I'm not sure I agree with you that the palm should remain stationary, because that simply isn't the case with many users. However, there is a hump for the palm in the model, which can be used to support the hand as it slides about comfortably.
Maybe you can submit a full scale, top-down drawing as a PDF so we can print it out and see how it might work with our hands?
Scrupulous (author)  bill.schuller8 years ago
Here you go... Keep in mind that any layout would take a fair amount of practice and conditioning.
alphanum8 years ago
I think you might be overestimating the dexerity of a single finger along a single line. I've used the alphagrip in which many fingers need to press just 2 linear buttons. See what I mean:


I find that quite difficult for many of my weaker fingers.
Scrupulous (author)  alphanum8 years ago
Heh, I would never over-estimate the dexterity of any finger along any line. ;>) I'm still amazed that anyone can type numbers without looking.

I took a look at that link and, man! I find it quite difficult just looking at that thing. Nothing about that device looks intuitive, or ergonomically useful at all. You'd have to keep looking at the underside until you memorize the layout, if you ever do...There doesn't appear to be any sense or consistency...There is a disclaimer for CTS...I hope you kept your receipt. ;>)
It's not particularly bad. There are stickers for the front that tell you what keys are where. Ergo wise, I think it feels quite good - your fingers are always completely at rest someplace secure, and your wrists are at a natural angle.
wombat78 years ago
if this is to be a one handed keyboard, then it will be necessary to produce both a right and left hand dominant configuration.
alphanum8 years ago
As far as the electronics goes, there's some GPLed microcontroller code and circut designs that let you build USB Keyboard devices using only an inexpensive microcontroller and a few other minor parts. I've personally built several devices with these designs and I can vouch for them:


Of course, the design would have to be extended a bit to accomidate the number of keys you're looking for. But because the source code is in C it's really easy to modify.
Scrupulous (author)  alphanum8 years ago
Excellent! That definitely seems like one way to go about this.
VIRON8 years ago
Nice. The i8048 chip commonly used in keyboards can handle this I think. If not maybe I could write an instructable to Make a keyboard controller. To all:What is missing from a regular keyboard that this can have? One thing, There could be a virtual mouse in the palm. Perhaps the thumb should unplug and move to the other side so ambidextrous people like me can use two at the same time. (Or will they be available in pairs?) What I really mean is... Do you want to make both leftys and righteys, (risking shortage or surplus) or be able to flip it over, which would be harder I think, than moving the thumb pad. How about User ReProgrammable/ Rearrangeable alpha keys? It might be too late to standardize, with many alternative devices. If implemented virtually (like a touch screen or tablet) one size would fit all, but some kind of movable bumps (like rubber "feet") maybe desired.
Scrupulous (author)  VIRON8 years ago
Viron, I'd say you are off to a great start. I like your thinking on this...

virtual mouse in the palm - Yes! (I was originally thinking flush-mounted arrow keys.)

unpluggable thumb - Why not? Then you'd probably make each finger unpluggable. That would take care of the surplus issue.

user reprogrammable/rearrangeable - Eww, that's been done many, many times.

implemented virtually - Yeah. And, somehow, I see a comfortable niche with the tablet market.

movable bumps - I like it! Picture this...a layout of pads, one for each fingertip and each knuckle. That's 3 points on each finger, and at least 15 points on the entire hand. Slide your hand up a tad, to another set, and you have 15 more. Slide it down, and you have a total of 45 "keys" to work with. Your hand stays flat the whole time, as it would only require slight pressure.

I think it would be cool if you could pull up the keys and rearrange them. Even if its been done..what harm could an extra feature do?
Scrupulous (author)  Weissensteinburg8 years ago
Yeah, I agree.
VIRON VIRON8 years ago
I apologize for failing to notice and read the additional steps. (too late to delete and edit?)
Scrupulous (author)  VIRON8 years ago
That's alright. Yeah, technically now, since I've responded, it would be too late to edit or delete, if you want to faithfully preserve the development. Actually, a forum like this may be more valuable if editing or deleting after the fact was not even possible. I'm not about to complain, because these guys already seem to have plenty of their ducks in a row. But anyway, it looks like you would be a productive contributor to this collaboration. I say run with this, if you like.
This is a great idea, but you should rearrange the letters. For instance, a is one of the most used keys, and you want us to use our weakest finger for it. The most common letters hould be for the index, and least for pinky, and distributed between that.
Scrupulous (author)  Weissensteinburg8 years ago
Thanks for the compliment. If I'm not mistaken, we are already using our weakest finger for A. Not only that but, 4 out of the 8 most common letters are on the index finger in this primary layout. Also, if you consider BFJPV as home row, almost two-thirds of the 15 most common letters are within one key. There will always be some trade-off. But, none of it would outweigh the simplicity and benefit of being able to memorize the keys alphabetically, as well as find any vowel without really thinking about it. I mean, most people cannot mentally picture the whole QWERTY alphabet at a moment's notice, even after decades of practice. Regardless though, this isn't intended to be a "Dvorak" attempt at overthrowing the 101-key. One of the main utilities of this system is to relieve the need for removing your hand from the keyboard to use the mouse. The ability to use the mouse while typing at a productive speed is not only huge, it is the next logical step. I use the shape of a hand, because it is simply more intuitive than anything else there is.
Try typing Every, or a sentance with the planned keyboard. also try plotting out the keys in a more usable format, They are querty for a reason. Try looking at the most commonly used letters and try to see if you can make the keyboard :3. Good luck.
Scrupulous (author)  ARVash8 years ago
A little keyboard history for you: Back in the day of early typewriters, keyboards were actually extremely ergonomic, people learned to type so fast that they jammed the typewriters. So they rearranged the letters to make it harder to type. But now with computers, it would be fine.
wombat78 years ago
looks cool. how about turning the whole thing 90 deg., so they would be in a "two hands holding a sphere" position? or tilting them less than 90 deg.; some slant so they are not flat on the table? just a thought, thank you for asking for input.
Scrupulous (author)  wombat78 years ago
Yeah, tilting the whole "keyplane" slightly counter-clockwise, or also slightly back, would add some comfort. You couldn't go too far because your hand needs to rest on something. You'd also be introducing mfg concerns, add'l storage and shipping constraints, etc.

Completely sideways would be okay, too. But, then you can't see the keys. And, although this primary layout is most intuitive, there'd be no sense in adding minor ergonomic improvements at the expense of visibility and overall utility, as well as the above considerations.

A convex shape, which is what I think you are describing, seems visually appealing, but really it's not practical. A concave layout is actually more effective. You'll notice that many keyboards are concave, at least along the "Y" axis.

Thanks for the input, though.
cartufer8 years ago
perhaps the opposite hand could be the alternate keyboard layout
Scrupulous (author)  cartufer8 years ago
Hmm, I don't know. It certainly isn't the thrust of this instructable. I'm open to ideas, though. But here, one of the main objectives is to free one hand up to use the mouse (or some other pointing device) full-time.
mikesty8 years ago
Agreed, and the "power keys" such as space, control, alt, shift, enter - the fun stuff - should definitely be controlled by the thumb.
Scrupulous (author)  mikesty8 years ago
The space button is situated to be hit by the outer heel of the palm. The other keys you listed are all situated to be hit by the thumb. (I agree that the thumbs are vastly under-utilized in the QWERTY layout.)
Ushanka8 years ago
Like what Weissensteinburg and Trans_Am said, you ought to move the keys around. It would be easiest if every key is struck by the fingertips; consequently it is too difficult to reliably and comfortably hit the keys at the base of the fingers (Underneath A, E, I, and O, in your diagram. It's too hard to curl my fingers that far everytime I want to hit a key. Similarly, I'd move the Enter button to an easier-to-reach place as well.
Scrupulous (author)  Ushanka8 years ago

As I mentioned earlier, the home row would be BFJPV.

Also, I would agree about the punctuation keys (!?.,). They were added to the model late in the process. They would actually need to be similar in size to the other character keys.

That might help you visualize it better. By the way, most people actually lift their fingers off of the keyboard when typing.

Trans_Am8 years ago
How about mving the thumb keys to the side of the thunb, so it would be a less awkward, more comfertable movement for your thumb, and more powerful as well?
Scrupulous (author)  Trans_Am8 years ago
The most comfortable line for the thumb is actually more horizontal than the one in the model. Fittingly, it resembles the "Instructables" thumb more closely than it does mine. (I used my own hand for the model.) Now, a lot can be done to refine the shape of the keys. This would have taken me much more time to model. Besides, it might be better left for a physical prototype. But, all the keys need to act roughly on the same plane. It's not that big of a deal. If you will notice, when many people type, they normally lift some of their fingers as much as an inch above the keyboard.