This is called a chorded keyboard [1]. Like seemingly everything invented in the following 40 years, it was introduced by Douglas Englebart in the Mother of all Demos in 1968.
Englebart had a iOS app that allowed you to chord type as well.
Douglas Englebart is the proof to me that even if you could go back in time, it isn't necessarily true that you could leverage your knowledge of future technology. Poor guy messed up the timeline so much that even his stock market knowledge was useless!
Englebart famously demoed possible ways of interacting with computers that are now ubiquitous decades before they were actually popularized. I made a joke about him only having those ideas because he was a time traveler from the future, but tried to implement them too early. Then, after he already didn't make billions from his ideas, he couldn't use his future knowledge of investment because his demo was still big enough to change which companies got wealthy (explaining why he didn't make billions investing in the companies that actually invented those ideas in his timeline).
I'm sure I remember an Asimov story where a reporter is hiding one of these in his pocket and taking notes unobtrusively, hoping nobody will notice the small hand movements.
I used a nice IBM chord keyboard in the late 70s but can't find a photo. It had the space bar sticking out to the side for your thumb; the other keys didn't just have a divot on their tops, as with a typical keyboard today but also on the faces and corners to encourage you to press multiples. With some practice you could get pretty fast.
I thought it would be cool if I could base a chorded typing system off an actual digital piano and use the MIDI interface to type, and design it such that most words would actually sound nice if the digital piano outputted sound at the same time. I just haven't put in the time to figure out a good efficient system.
Would be kind of neat if I could "hear" a blog post as I was typing it, and maybe eventually even learn to listen to text transcribed as music.
Part of what makes music so wonderful (and universal) is that it encodes information somewhat more ambiguously than written or spoken language. You could probably see comprehension and memory dividends from transitioning to a music-based communication system - especially considering that we're hardwired to process music even more so than we are to process language - but I dunno if I'd want to give up the aforementioned aspects of the human experience.
A strawberry is a berry. I don't care if biologists or botanists or whoever use a narrower technical definition in their research papers. The colloquial definition is just as valid. Prescriptivism-by-authority can take a hike. Yes, I will die on this hill.
I’m not sure which way your email joke is going. Is it poking fun at some HNers who go out of their way not to use common services or devices, or, are you joking that changing your username isn’t worth any effort?
Edit: oh. Likely a 1984 reference. Never mind now. Leaving comment as it is though
It’s because of the two options I proposed. Both being HN-y things.
You sticking to your guns and not using cookies or JS shows me your principles are legit.
Normally and what my comment was referencing, are people who tout one abnormal or specific thing they do or don’t do. Like how awful FB or one of its properties are. While they happily go on every other data and privacy invading site/company.
A lot of what you're saying is hard to take seriously, because it's just plain bizarre. A complete refusal to use email for any purpose is pretty unusual to begin with, and even more so for a participant on a technical forum like this. Publicly complaining about the consequences of that decision is another layer of strange behavior. It seems like you're putting a lot of effort into not communicating effectively, so you really should expect people to have trouble taking your statements at face value.
> A complete refusal to use email for any purpose is pretty unusual to begin with
I don’t know if I would say I “refuse” to use email. I simply opt not to because I am privacy conscious. I also do not browse the web with cookies or JavaScript enabled. If that’s “bizarre” to you well then I guess I am a bizarre person but I deserve a bare minimum of respect just as all humans do.
> Publicly complaining about the consequences of that decision is another layer of strange behavior.
I literally never complained about anything. Someone else initiated a question at me about my communication with dang and I shared the results of that communication. I am not bothered at all by not being able to request a name change nor have I ever expressed resentment towards anyone for not being able to request a name change. My screen name is not very important to me. I feel like I am being unfairly judged here.
It is weird to not have watched any TV in decades. However the coworking community I run online funnily enough has some of the youngest members (early 20s) who do not get any references I make. They also seemingly watch no tv at all. Limited YouTube as well so not like that’s replacing it for them.
It hasn’t been a second thought of that being something to point out.
However I guess in an in person situation, it may come up more as a “thin”
It’s not a non-sequitor nor a joke, the point is to highlight the danger of linguistic relativism. If we dismiss the idea that words must follow some standard meanings, things we value can be gradually shifted into weapons of oppression.
You're seriously overthinking this. Descriptivism isn't some plot to twist our cherished language, it just describes the language as it is currently used by speakers.
I believe there needs to be serious limits on the ability of “descriptivists” to condone divergent meanings of words. Within reason it is fine to describe new slightly expanded uses of words but wholesale accepting that word meanings can change arbitrarily in principle seems harmful to the maintenance of the integrity and quality of our language. For example, many people dislike HOAs but neighborhoods with HOAs maintain a certain level of quality, while neighborhoods without HOAs have quality levels across the board, including very bad.
It doesn’t matter who decides, it only matters that a decision has been made and that is it respected and enforced. Generally academics and/or intellectuals edit dictionaries
> Well, digits; a thumb is not technically a finger:)
Since you used the word "technically": actually the thumb is considered a finger in anatomical use (which is about as "technical" as you can get) just as the hallux (in popular jargon, "big toe") is anatomically considered a toe, despite having its own name.
In a prior life in pharmaceutical development we had a drug program in which this was specifically important.
Not only that--different species have used different parts for the thumb. A Jurassic pterodactyl ("wing-finger") had the earliest known opposable thumb. Pandas have their choice. Primates are late to the party, but might have a claim to most use. Some cats have joined.
The most important key-chord is space, followed by backspace.
One of those pointless little distinctions that English language makes but many other European languages don't.
Another example is "velocity" vs "speed"—what other vector quantity has two distinct names, one for the vector, another for the vector's magnitude? The acceleration doesn't, for example; and distance and displacement are two slightly different concepts that deserve two distinct names.
It's also one of the places where US units really break down. Working with pound-force, pound-mass, and slugs is really easy to mess up if you have to work that way. You're often better converting to SI, doing whatever calculations you have to do, and then converting back at the end.
> Are you telling me that I have four fingers in
> English but five Finger in German?
In some languages, such as Arabic and Hebrew, you have twenty. In these languages the colloquial word used for digit does not distinguish between digits on the hands and digits on the feet.
Though here is the definition from Merriam-Webster, which I'd consider the standard reference for US English.
any of the five terminating members of the hand : a digit of the forelimb especially : one other than the thumb
That seems more correct to me because I don't think in normal usage saying that "most people have 10 fingers and 10 toes" would be considered an inaccurate statement. And, if someone said they have four fingers on their right hand, you'd assume they meant they were missing a digit.
And even the Collins dictionary has a countable noun meaning with the usage example: "The fingers of a glove are the parts that a person's fingers fit into."
> Though here is the definition from Merriam-Webster, which
> I'd consider the standard reference for US English.
M-W considers itself a "descriptive" dictionary, not a "prescriptive" dictionary. That means that it will tell you what people who use words mean to say, but it will not tell you the actual meaning of the words. Therefore it is absolutely NOT a standard reference for English.
The nuance is sometimes important, M-W was the first dictionary to give "figuratively" as a definition for the word "literally" due to the way some people use the word "literally" online.
Much as I dislike literally==figuratively I'm not sure what descriptive vs. prescriptive has to do with standard reference. I'm somewhere between the poles. On the one hand see figuratively comment. On the other hand, languages are living things. And, to my example, if you tell me you have four fingers on one hand, I'll ask you how the accident happened.
I also think you'll find a lot more organizations that use MW as their reference standard than other US dictionaries.
Descriptive: Describes how people use the language.
Prescriptive: Describes how language should be used.
Descriptive dictionaries are highly favoured by those who like to effect change (I'm deliberately avoiding the word Liberal because I'm not in the US and I understand that word to be politically charged for many HN readers). Prescriptive dictionaries are more favoured by those who keep traditional values. Descriptive dictionaries can be used to say "I'm using the word correctly" and prescriptive dictionaries can be used to say "you're using the word incorrectly".
But organizations mostly want to talk to people in the language that they use. They have zero interest in being the language police. They may decide that their writers shouldn't use literally to mean figuratively. But, at least where I work, we're mostly interested in using the language that best connects to readers.
I believe that the real intention behind the statement is that not all users of their product are going to have 5 working fingers if they've had some sort of accident or similar.
That line of thinking doesn't get you far with anything. Nothing is universal. You can always find someone with some disability that prevents them from using a product.
Accommidating as many people as possible is good, but you can never accommidate everyone. Same goes for all of these "things programmers believe about X", be it names or whatever. You absolutely need to provide working product for majority of people first and majority of people have at least two names or in this case 5 fingers.
As a mechanical and ergonomic keyboard aficionado with an immobile right thumb, I agree 100%. Many of the cleverest "ergonomic" keyboards heavily overload the thumbs. This is great for moving stress away from the weak, fragile pinkies but only works for people with two useful thumbs!
Difference is one is generic for typing anything, and the link there is purpose-built on a per-application basis, but the idea of having a small amount of keys for a precise job is very interesting.
It existed and was too fucking expensive for what it did and how it worked even in the terms of mechanical keyboard connoisseurs. (It was mechanical, but to cut some costs they used the worst Cherry switches they could lay their hands on.)
Still, it could be praised as a proof of concept that such things were doable. I hear there's a kickstarter on a keyboard with Eink screens, which is reportedly not as hideous.
The Lebedev studio also produced an Optimus Mini Six macro pad which was more useful and lived a bit longer after the Maximus bit the dust.
Read my comment history. Also, please try to be respectful in future on HN. If you make an unproven unsubstantiated claim, and someone says you're mistaken, ask yourself for your own references first.
Alternative chording scheme - take sequence into consideration.
Chord starts with first key depressed and ends when last is released.
One-key chords: 5
One of 5 keys to press, one way to release key.
Two-key chords: 40
1 of 5 possible starting keys,
1 of 4 possible second keys,
5x4 20 openings.
With 2 keys held in, two ways to release: 1 of 2 keys to release first, 1 remaining key to release last.
A total of 40 (20x2) 2-key chords, 45 total chords so far.
3-key chords: 360(*)
5x4x3 for 60 openings.
1 of 5 first keys,
1 of 4 second keys,
1 of 3 third keys.
3x2x1 for 6 releases.
With 3 pressed, 1 of 3 to release first, 1 of 2 to release second, last key released.
Total of 360 simple(*) 3-key chords.
405 (5 + 40 + 360) chords so far.
Yeah, biggest effort is not the hardware, but wiring the muscle memory, but I figure out motor system should be great at this if you're motivated.
(*) Simple chords - where we first just press and then just release, more options are available if we are ready to keep the sequence going by pressing and releasing and pressing again.
NOTE:
Not original, this is my take on some one else's scheme, I'll look for the reference.
I used twiddler 2 (I think). It was cheaply made chorded keyboard. I admit I did hold down the keys rather hard while learning to use it. Soon after I'd learned the patterns the keys started breaking. I took it apart to take a look, all the keys were connected by tiny bridges of plastic. When the plastic bridges broke the keys became loose and non-functional.
I have the bluetooth version. I like it. I have learnt enough to use it reasonably well, but not well enough to use it as a daily driver. There is a significant hump to get over, and it's not easy to force yourself to practice with it. My latest problem with it has been that one of the keys has got reluctant to register presses, which makes it a lot more frustrating to use.
I have one I bought maybe 8 years ago. It's the wired version. There was some pretty good tutorial software for learning to type but I could never really get the Ctrl key combos to work which made emacs hard to use. But for just letters and spaces and tabs it worked surprisingly well. You can change the layout also.
The use case I bought it for was note-taking in class. My thought was to learn how to use it on my left hand, keeping my right hand free for drawing diagrams in my notebook or anything else that was hard to type out via just text (like math formulas).
I also choose to learn it on my left hand. The way you use it your hand is kinda strapped in so if you want to do anything thing else (like open a door for instance) it's better to have your dominant hand free.
I've had one since the early 2000s. I never learned how to type on it. I tried it for an hour and it was really hard for me to learn (although I was trying to do it in my non-dominant hand). I really should try to find that thing and learn it again.
Wouldn’t a T9 keyboard work better for people who texted before the smartphones? I imagine it’s like riding a bike: I’d probably be able to type blindly with one thumb again after a bit.
I'm sure you could even make T5 though it probably works less well with only 5 keys.
For those who don't remember: T9 would have one key for "abc", own for "def" and so on. You didn't have to tap the first two key twice to type "be".
You would simple tap the first and the second key, then it would look for words matching "(a|b|c)(d|e|f)" and conclude the most common such word is "be".
I think you could further argument it with grammar awareness, so really you tapped the length of the word. For ambiguous words you would have a key to switch between words.
Everything was deterministic, so you'd quickly be able to type without looking.
I think the only limit is the time it takes to train a human to write using such keyboard.
And perhaps more problematic how bored said human would get, and hence, give up :)
If you can effectively type 70,000 characters with 26 keys, certainly you can type 26 characters with 5 keys. It would be even easier if you made it two handed.
(In other words, I think this could someday be a practical text input method)
I'll strongly suggest changing the approach from character input to phonetic input like the stenotype machines.
Creating a nice set of rules prioritising writing speed is of course no easy task, however I think it is the only viable way in my opinion to make it usable.
These were around as far back as the 80s IIRC - and when I was at Intel in the late 90s - there were stories of this one engineer, who would a recumbent bike down San Thomas Expressway to the SC campus, and he had chorded keyboard he made attached to the handles on his bike - and the legend was that he had a sat phone on his bike and he would code on his way into Intel while just keeping track of everything he typed in his head. He didnt have a display...
There were also some pants in the early 90s that had some of these built into the thighs of the pants.. so you could type with your arms basically at rest on your lap or at your side.
Would be curious if it'd be worth it to add either a sixth button on the side to be pressed by the thumb, or to add something like a 5-way switch [1] to the top. You wouldn't have to use all 5 ways, and it'd greatly increase the amount of characters possible, potentially allowing whole words (and, the, patient etc) to be hit by buttons.
I bet a lot of what the paramedic friend types is repetitive and industry specific. Could speed things up with some macros/text expander type of functionality.
The multi presses reminds me a lot text messaging back before smartphones. Most kids could type quickly without really looking and using just a thumb.
Hi, paramedic friend here. The whole "paramedic notes" thing was just a very dubious pretext. I was looking at a small notebook page of notes after a call (no patient identifiers, not that it would have mattered), and was curious if anyone would be able to make out a single word. Stavros is Greek, so he was the obvious choice. That's where the story in the post picks up (with a little creative license).
In reality, laptops are quite common on ambulances, and get used during many (even most) "routine" calls. The issue with note taking is with more critical patients, and there are a variety of reasons why a notepad and (cheap, disposable) pen are the best solution. The biggest reason is because I can use them with a gloved (potentially contaminated) hand and not worry about how to decontaminate them, since I'll just throw them out. This device has many of the same failure modes as the laptop (namely, it would be impossible to clean).
A lot of medics will skip the pad and just write on the non-dominant gloved hand. Personally, my favorite technique during especially busy calls is to put a strip of 2in/5cm wide cloth medical tape on my thigh and just write on my leg.
But "practicality" wasn't really the point here...
There are a few challenges I can think of there. The most obvious is that there are things that are accurate and relevant, but I might not want to point out to a patient in the moment (a combative drunk patient might take offense to my documenting the presence of multiple empty liquor bottles, for instance). Obviously this information is part of the patient's medical record, and available to them in the future for review if they want, but there are things that I don't necessarily want to speak out loud in the moment.
The other, more subtle, reason is that free form handwriting gives me a lot of flexibility in on-the-fly formatting to emphasize important information, etc.
The esp32 never ceases to amaze me. I bought a set of five of them from Banggood awhile ago, and I was amazed how quickly I was able to build some pretty interesting stuff, like a wireless arcade controller (using the same bluetooth keyboard library that the author mentioned, actually).
It's fantastic, I have tens of them because I constantly use them around the house, in things like alarm clocks (https://www.stavros.io/posts/do-not-be-alarmed-clock/), cat toys (no writeup there), scales (maybe later) and other random things.
I remember in the late 70s seeing keyboards like this for sale (via ads in the PC magazines of the day, like Creative Computing). What was most notable was they were built on a hemisphere (dome) which feels like a more comfortable shape for extended use.
I don't know how useful this might be for your friend (thankfully I have no idea what the inside of an ambulance looks like) but this configuration could be mounted on the armrest of a chair or even better on the side of the chair so you could use it with your hand hanging straight down. (Or on your belt if you wanted to go full nerd, I suppose.)
That is very cool!
I think the double-presses and three-button presses are pushing it for usability. I wonder if there is a version where you can keep the general form and idea and increase the ease-of-use.
I'm thinking add a 5th side-key and a second top-key and use the top ones solely as toggle, not as letters. This way you have 3*5 keys total and fill in the rest with a predictive model, maybe?
Or how about a layout with a second side-key column on the side that faces outward. You'd get 2*10 keys with just those columns and the toggle (or 3*10 with double toggles).
Using a predictive model would be awful for the inspiration use case (must keep eye contact with the patient), but probably more convenient for most others.
Personally, I'd rather have the chords than have toggles—a keyboard should really be stateless. Even the existing rarely used states (num lock, caps lock) do more to cause me frustration than help.
Sorry I was unclear, by toggle I meant more like shift than capslock, you have to keep pressing don't know what's the correct terminology for those kind of keys.
I would propose four standard keys on side and a thumbstick on the top. Thumbstick is easy to operate and has at least 8 discernible directions, plus 8 half-pushes. That's more combinations than you realistically need so maybe you could relieve the pinky finger and use just 3 buttons.
There are keyboards where each finger goes into four (five) switches, one on each side, but I imagine some combinations will be impossible to type without the keyboard being stationary relative to the hand.
It seems like every hardware geek makes one of these at some point. I think it's really cool. Designing a chording scheme for individual keys is a fun problem, kind of a nerd snipe. But in the end everyone concludes, "It's too slow to be useful."
I think where the hand held keyboard might be really interesting is in combination with steno (http://www.openstenoproject.org/plover/), but I haven't seen that done.
It seems that handwriting is 24 WPM. It would be neat to get that input speed on a computer using one hand without much more exertion, and choosing letters from a menu seems like more exertion.
I have a SpiffChorder [1] built by Greg Priest-Dorman. It has 7 keys (three for thumb). I can write at about 40-50 wps and code in emacs reasonably, but it is slower, so I only use it when I am eating something or have one hand busy. Greg mentioned having two keyboards in each hand should be faster than a single traditional keyboard, but I never tried that.
I have that one starred. ASETNIOP looks quite interesting to me. I just wish it was open source. Without that, I don't see how I could possibly adapt it to the various operating systems I use day-to-day.
Hey, I love the innovative designs from g Heavy Industries!
Unfortunately, I've had unsolved problems with 2 out of 4 orders over a couple years. First issue was the Georgi I ordered just never arrived. Only the lowest cost mail option was offered at the time, and it had no tracking or insurance. No refund. The second issue was the GergoPlex Heavy I ordered. I was able to work out tracked shipping at an extra cost, but after receiving it, I saw it had obviously been dropped by some delivery person. The GergoPlex Heavy was made of steel, and it was shipped in a thinly padded envelope instead of a box, so it must have hit the ground with a lot of force and insufficient impact protection. I bend back the MiniUSB port to make it functional, but the steel plate was also bent at a corner and was much less feasible to fix. My living in the US certainly makes shipping and handling from Canada more complicated, but I don't think I'm being unreasonable in expecting tracked/insured shipping and protective packaging for a $250 item. I've since given up on ordering any more, but this may just be my bad luck.
Since there's a plenty of chorded keyboards already, someone should come up with some sort of a scat keyboard, or rather input method.
Each keystroke is represented by a sequence of three notes that you sing into the microphone. Three notes produce two intervals which, measured in semitones, are converted into keys pressed (being, say, duodecimal digits). The relative duration of notes sang can be interpreted as modifier keys.
Buy a premium version that plays a fine selection of fusion jazz backing tracks with drums and synths, while you solo.
I often have ideas while walking and I always thought this kind of device would be good for jotting down thoughts while taking a walk... I could always use a phone with some dictation note-taking software, but I think a small chorded keyboard device that records everything, has a great battery life, and can sync via bluetooth or wifi would be very nice.
I have done this with the twiddler3 paired with my phone in my pocket (you can also use it in recording mode, but I didn't). I don't do it a lot, but I enjoyed it for what I did with it.
I'd quite like some sort of audio feedback while doing it, but I was accurate enough that it could still work as notes fairly well.
Neat. Embedding a multiple axis gyro or accelerometer so you can recognize wrist gestures could be the next step to get creative with additional keys. Ex. left flick could be delete.
Great work! I've been wanting to do this but add springs to it to make a finger strength builder that I can use as keyboard controller input for playing pc games.
Curious if there are any chorded keyboards that use tilt sensors to make character input depend on button plus position, rather than button combinations alone.
wow this is really awesome. Actually was look at creating a novelty device for myself. Steam controllers touchpads, logitech MX's scroll wheel. ps5 haptics. Buttons with a great tactile feeling.
If anyone knows community that like building things like this, I couldn't find anything. (mainly because i have no idea if it has a name)
Definitely check out the mechanical keyboards part of Reddit. There are a lot of ordinary keyboards with fancy switches but I’ve enjoyed getting into the ergo layouts like [Soufle](https://josef-adamcik.cz/electronics/let-me-introduce-you-so...) which I’m just about to build for myself.
The Romac is a numpad version. Same techniques all around. Use a Pro Micro chipset and install the QMK firmware and you’re off.
Seconded, I just finished my first full-size keyboard using QMK (writeup on that coming later) and I love it! I'm typing on it right now, it fits my hands exactly and was LOTS of fun to make.
A few years ago in university I tried taking notes using an Apple wireless keyboard with my iPad in my backpack. It was surprisingly liberating to just type and not have to look at a screen. I'm sure I looked like an idiot but it was an advanced software engineering course so it wasn't to far leftfield :)
That was my first reaction too but it's growing on me! With IntelliJ on an external Mac keyboard (because, 2017 era MBP, yeesh) for some reason they didn't group the F keys. I mean at least there are F keys. So it's always stressful to try to step over, because the continue button is right next to it (F8 and F9) and it's easy to lose minutes of context buildup with one wrong button press. And then, HTF do you force-step-out? Is it Alt-Shift-F9? Cmd-Ctrl-F9? Alt-F9? Oh it's Shift-F8. But unfortunately by now I just relaunched the entire project.
It would be so cool to mount a Soviet nuclear control room type panel under my monitor with those keys spelled out. Oh and a mute button.
Happily IntelliJ keys are configurable. Settings->Keymap. Debugger actions are under Main Menu→Run→Debugging Actions.
Sadly it's difficult to see all the mapped keys at once, and the search can't search for the key (only for the action a key might be bound to) so making new maps is a bit tedious. It's possible to do so via making a plugin, but that's probably even more effort.
There's a trade off between speed, ergonomics, and moving your fingers around the keyboard to find keys not accessible with your hands on the home row vs using all the keys on a keyboard.
I'm a big fan of the model M. Its one of the loudest keyboards produced "Their buckling-spring keys are noisy enough to be inappropriate in quiet locations such as libraries and medical facilities." [1]
That was actually one of suggestions I (the Josh in question) brought up. I'm also a ham radio operator, and my morse isn't _great_, but it would definitely be faster than trying to learn this chording.
In general though, nothing is going to beat the speed (and disposability) of a cheap pen and paper.
Englebart had a iOS app that allowed you to chord type as well.
1. https://en.wikipedia.org/wiki/Chorded_keyboard