Ever received a text message while driving, biking, in a boring class, etc., and thought to yourself, "I wish I could read this and reply without looking at my phone"?
There are 31 combinations of buttons I could press with the five fingers on one hand; in my ideal world, I would be able to use this method to input text by holding my phone in my hand. Here's the system I devised:
This encoding lends itself naturally to a Braille-like tactile alphabet that can be read by touch. It's a little simpler than Braille as well, because there are only 32 different symbols (25
) rather than 64, and it can be input with one hand rather than two (as Braille keyboards require).
Each symbol is arranged hexagonally as depicted above, with the center dot representing the thumb, and the four outer dots representing the other fingers. (The hexagonal shape maximizes the density of dots while also maximizing the distance between them.)
The general principle is that the easiest symbols represent the most common letters in English (according to the familiar ETAOINSHRLDUCMFGYPWBVKXJQZ ordering). By "easiest" I mean:
- Fewer fingers are easier than more fingers.
- Stronger fingers are easier than weaker fingers. (stronger = closer to the thumb)
- Moving the middle or little fingers without moving the ring finger is especially difficult. (If your hands are anything like mine, try it yourself and see.)
Thus, the most common letter (E) gets the easiest symbol (thumb), while the least common letter (Z) gets the most awkward symbol (index+middle+little).
Some remarks about the non-alphabetic symbols:
- The blank symbol (_) represents the space in output, but is not used in input (since you can't detect pressing zero buttons).
- The middle finger (→) and the ring finger (←) represent "space" and "backspace" in input, but are unused in output because it would be too hard to distinguish them from T and A by touch.
- The # symbol represents numbers and punctuation (e.g., #_ = "period", #A = "1").
- The full fist (↑) is the shift symbol, denoting capital letters (where it's necessary to mark them).
You may notice that one of the 32 possible symbols is entirely unused. This is the "middle+little" symbol, which is unused because it's too similar to R, and is too awkward to be used as an input-only control like → or ←.
Much like Braille, we can use single-letter abbreviations to shorten the text. For example "THE" is abbreviated as "C", which in any case is what you'd naturally do if you tried to input "T H E" really fast. (It's a pretty easy gesture as well - it's how you'd hold a fine cup of tea.)
Here's a sample text (made by poking index cards with a pencil):
Transcribed literally, this says: "ALL HUMAN BEINGS R BORN FREE D EQUAL N DIGNITY D RIGHTS# CY R ENDOWED W REASON CONSCIENCE D S ACT TOWARDS ONE ANOTHER N A SPIRIT F BROTHERHOOD#"
Here it is again, with the spaces marked for clarity:
(Note: I realized afterwards that I forgot the word "D" between "REASON" and "CONSCIENCE", in the place marked by the double red line.)
I'm not sure how easy this is to read without looking; I myself can only identify words because I already know what the text says. I'd imagine it would get easier with practice, though. Also, this text is written in the "compact style" that leaves no gaps between letters and only a single-cell gap between words, so the dots tend to run together and create a somewhat ideographic effect.
So, what do you think? Any ways you'd improve it?