geekhack
geekhack Community => Off Topic => Topic started by: gosinger on Sat, 16 May 2015, 03:53:20
-
Hi,
I'm rather new to the mechanical keyboard scene, and have just started to think about 60%/75% keyboards, ISO vs ANSI, DE vs US vs UK layout, etc etc. While I have grown up using ISO-DE, at work my subjective feeling is that I mainly write java-code or mails in English.
Out of curiosity, I'd like to know which characters I type at which frequency in general and within which specific applications, so I could "optimize" my layout for my personal commonly used characters. The first thing that pops into my head was simply writing a key-logger which also includes functionality to determine the active focused window that input is sent to. Based on the collected data it should provide a statistical representation (i.e. a heat-map for different layouts / activities / etc...). The result should be an objective statement on how frequently I really use certain sets of keys (i.e. F-keys, German umlauts, different types of braces, etc) and which layout would grant access to those with the minimal amount of key-strokes.
What do you guys think of this approach, am I thinking too "mechanical" (from an engineering perspective) about this or could such a piece of software actually be useful?
Obviously there are certain security concerns that would have to be handled when writing such an application even for personal use (don't want my passwords to show up in there :)), no doubt about it....
-
My guess is that it would be very difficult to interpret the data you get. When you program, you spend most of the time thinking, not writing. I guess you could, to borrow a programming term, "trim" the time interval around every word you type. A second possibility is to track specific characters and compare time between the character before it (and weed out the anomalies). Brackets would word fine, because they're typically at the end of a word.
A second possibility, is that you could make a typing test for (similar to the ones online), but for code. The program could check which layout is active and store your results in a database. The results could be compared between layouts, especially with focus on keys that differ between the two layouts. Keep in mind that it takes a while to get used to a layout, even longer to become proficient, like you are with DE.
I switched to UK layout a few months ago, and plan to further switch to US when I get my ANSI keyboard. The latter move is simply to have more alternatives with both keycaps and keyboards. I make less typos for special characters like brackets. The key placements are just so much better suited to programming. I've made a custom layout for זרו that I access with ALT GR. Alternatively, you can just switch between the layouts when you need to write something in German. I personally found that to be confusing, and I often forgot which layout was active.
-
You're late to the layout party.
Essentially, it's not worth it..
Typing SPEED is not a limiting factor for HUMAN output, be it programming or erotic-novels.
So optimizing the layout, WHILE possible, will ultimately be fruitless..
It's very hard to produce Material @ even 33wpm..
Many GHers are over 130wpm in transcription speed ON qwerty, but ask any of us to write you a novel.. That speed is worthless..
And in terms of "Maximum Speed possible".. Sean Wrona (fastest guy publicly known) can only do 160wpm tops on unfamiliar text.. (using qwerty)
So in that sense the limitation is our "mental" capacity to parse text, where again, layout play only a minor role.
-
You're late to the layout party.
Essentially, it's not worth it..
Typing SPEED is not a limiting factor for HUMAN output, be it programming or erotic-novels.
So optimizing the layout, WHILE possible, will ultimately be fruitless..
It's very hard to produce Material @ even 33wpm..
Many GHers are over 130wpm in transcription speed ON qwerty, but ask any of us to write you a novel.. That speed is worthless..
And in terms of "Maximum Speed possible".. Sean Wrona (fastest guy publicly known) can only do 160wpm tops on unfamiliar text.. (using qwerty)
So in that sense the limitation is our "mental" capacity to parse text, where again, layout play only a minor role.
What about finger fatigue, isn't distributing the workload the purpose of Dvorak ?
-
You're late to the layout party.
Essentially, it's not worth it..
Typing SPEED is not a limiting factor for HUMAN output, be it programming or erotic-novels.
So optimizing the layout, WHILE possible, will ultimately be fruitless..
It's very hard to produce Material @ even 33wpm..
Many GHers are over 130wpm in transcription speed ON qwerty, but ask any of us to write you a novel.. That speed is worthless..
And in terms of "Maximum Speed possible".. Sean Wrona (fastest guy publicly known) can only do 160wpm tops on unfamiliar text.. (using qwerty)
So in that sense the limitation is our "mental" capacity to parse text, where again, layout play only a minor role.
What about finger fatigue, isn't distributing the workload the purpose of Dvorak ?
That's true. That layout aims to place keys in a more sensible way than qwerty.
The problem is that you have to learn the muscle memory of that layout and it is fairly difficult to do. People spent months for wpm significantly slow than their old layout.
To me, I think people would be better off improving their keyboard technique than switching to a new layout.
-
There shouldn't be much "finger fatigue" if you're on cherry or topre. they're both standardized well below the forces necessary to fatigue fingers.
MX-Blacks maybe.
But layout is NOT the cause of finger fatigue. the travel distance between keys is mostly accomplished by wrist and arm. The finger is not working against a "force" to do this.
When we DO SEE fatigued fingers, it's due to improper technique cultivated on standard-bottom-out style keys, where the people press really hard, BEYOND the force necessary to ensure activation.