Hey Linkbane,
I think you’re really dramatically misinterpreting my intended tone here. I’m not trying to insult you, or even suggest that you’re wrong. I certainly don’t want to waste anyone’s time. I just didn’t/don’t understand the logical argument, because as far as I can tell both the premises and the reasoning have been insufficiently elaborated.
* * *
Thanks for throwing in this part:
If you want to talk about why you use the thumb on the less used hand, the reason is simple. When you're actually in need of typing fast, it restricts the moving hand greatly by forcing it to keep level on the keyboard.
That’s just the kind of detail I was looking for. I'm not sure that I agree though, especially when it comes to a custom physical layout. Personally, I don’t find that my hands really need to move that much, and in particular [when typing on a regular keyboard with QWERTY layout] my left hand doesn’t move much at all. My right hand moves somewhat more to hit the return key, the further-away shift key, the really badly placed backwards delete key (moving this to a thumb is SO nice), the 'Y' key (it’s a bit more common a letter, and requires slightly more reaching than 'B' because it’s diagonal up, instead of straight across, given the typical position my hands are in), and all the random punctuation and symbols essential for programming ... but it’s still not really all that much movement. I think by just the “how much hand movement is required” metric, I’d tend to recommend using the left thumb for the spacebar regardless of the specific character layout, because of how much further over the right hand part of the keyboard extends, and especially how many more keys the right hand ring finger and pinky are responsible for.
But to be honest, I doubt it makes tremendously much difference. I’m curious what differences you’d see if you took two batches of typists and trained some to use only their right thumb for spacebar, and others to use only their left thumb. My suspicion is that the differences would be negligible.
* * *
I’m decidedly not an expert here, and in particular I haven’t spent any substantial amount of time examining carpalx. As far as I can tell, carpalx is a computer program, which, based on the particular set of assumptions baked into it, gives scores to key layouts. It’s a tool which is only useful insofar as (a) its premises are correct, and I’m not sure I buy that; and (b) the program does what you want, but for instance I don’t know if carpalx is designed to analyze new three-dimensional layouts – it seems like to do so would require a detailed model of hand anatomy. But again I haven’t studied it in any detail. It’s definitely not a “study”.
For example, one of the premises behind carpalx is that particular fingers will type particular keys... but they don’t seem to have lead off with an analysis of which finger it’s most comfortable to use to type each letter (for even an average shaped hand, not to mention especially large or small hands). Or maybe they have, and I just didn’t see it. To make this concrete.. here’s approximately the way I use my fingers on the keys of a standard layout keyboard, which is based on tilting my hands slightly inward to reduce wrist twisting, and is based on which motions are then most comfortable for me:
You’ll notice that this differs from the standard recommendation of, say, high school typing teachers, or the model-defining assumptions of Carpalx and similar software. Using my own assignment of fingers to letters, I would arrive at different conclusions about which layouts are more or less optimized than someone else would get who only used the “standard” finger to type each letter. But if you observe various typists in practice (i.e. if you just watch people in coffee shops, or at the office, or whatever) you’ll notice that most people have their own idiosyncratic typing styles which have similar kinds of deviations from “standard”. Indeed, I don’t think the standard way was very carefully thought out, and many of the standard motions are quite uncomfortable. For example, I find typing the 'x' and 'c' keys with ring and middle finger, respectively, is quite uncomfortable, even though this is the method typically considered standard.
I am genuinely curious to see some studies about which fingers get used for which keys by real people, in practice, and also genuinely curious to see any study related to ergonomic keyboard design with solid experimental design; almost all of the keyboard-related ergonomics studies I’ve seen suffer from all kinds of problems: small sample sizes, biased user selection, bad controls, bad data analysis, etc.
I’m not trying to flame you, or bait you into flaming me. If anything I’ve said seems dismissive or insulting, I sincerely apologize; that was certainly not my intent. I admit I’m sometimes not very good at conveying the tone of friendly and respectful inquiry that I’m aiming for. [This is the internet: if there’s one thing we’ve learned in the last 30 years it’s that discussing anything on the internet tends to lead to misunderstandings.]
If you’re ever in San Francisco, drop me a line and I’ll buy you a beer.
Your original post was so flawed that it was ridiculous.
Which post, and what’s the flaw? I wasn’t aiming for ridiculous.
Also, I really like the little bit about my spelling and how you made that your first point.
I thought you were talking about some kind of thing called “fiver”, which I have never heard of. If you actually meant “five usage maps across layouts”, I still don’t know what you mean [which is why I assumed it was the name of something specific]. Which usage maps are we talking about? (What’s a “usage map”?) Do you have a link?
You then try to ridicule me, by bringing up random suggestions about anatomy and how I could possibly walk you through it
I’m not joking here. If there’s a good argument here based on hand anatomy, I want to understand.
* * *
Maybe to clarify my motivation here a bit: I want to build my own ergonomic keyboard, write my own firmware for it, and come up with a new character layout for it, which matches whatever shape hardware I end up with [I’m not sure how much carpalx will help me here: I might have to also write some software for optimizing the character layout given my own scoring of key combination ease/efficiency... but I should investigate]. So I want to understand everything I can about efficiency, accuracy, and comfort on existing keyboards, so that whatever I finally make ends up being effective.
I’ve personally found that if, in my custom ergo keyboard shape, I position finger and thumb keys in such a way that requires no substantial reaching to hit any key, then the use of thumbs and the use of fingers on the same hand seems relatively independent. That is, holding down a thumb key for shift and then using keys on the same hand to type capital letters doesn’t seem to be especially slower than just typing those letters without the thumb held down. Likewise, pressing a thumb-key spacebar with the thumb on the same hand whose fingers are used for letters right before/after it doesn’t seem to be any slower than pressing a thumb-key spacebar on the other hand. This seems to be more true when the thumb keys are placed very near to the neutral resting position of the thumb, and angled to align with the natural directions of thumb joint motion. A standard keyboard definitely requires some constant thumb flexion away from its neutral position, and the standard keyboard spacebar is not ideally oriented to take advantage of thumb motion. So I can imagine that on a standard keyboard, a hand whose thumb is completely unused will have an easier time pressing finger keys.
Have you ever used any non-standard-shaped keyboards, especially ones with extra thumb keys such as the Maltron, Kinesis, Ergodox, or μTRON? If so, what did you think of them?
* * *
Inre the other thread, about MX blue switches, I think you’re maybe confusing me with another poster, or interpreting my words in a different way than I intended them? I wasn’t trying to insult/dismiss/antagonize you over there either. I was just asking which switches you’d tried, and suggesting some other types of clicky switches that you might enjoy. Thanks for explaining what you liked and didn’t like about MX blue and Model F switches: I thought your explanation was entirely reasonable; these choices are, of course, a matter of personal preference and taste.
* * *
wolfv, hopefully I’m not derailing your thread. :-)