Comparison of text editing methods | Hacker News Reader

Comparison of text editing methods

devonzuegel.com

53 pointsluu16 comments

I spent a couple years developing some software for doing text editing in a new way, not quite in any of the categories mentioned. The specific piece of hardware I was aiming to use at the time was the Microsoft Kinect—but the main goal was to develop software that would work for a variety of low precision and/or resolution input devices, where resolution refers to the number of input symbols made available by the device.

I partly just ended up reinventing the idea of a structured text editor, but did also build a new variation which I'd bet hints at a still worthwhile and largely unexplored direction (as a text editing UI concept). You can see a video here: http://symbolflux.com/projects/tiledtext (the most interesting part is at about 1:15 imo)

I'm unlikely to continue with that project exactly, but I still think we would gain a lot by figuring out better methods of making text editing efficient with low precision/resolution input devices. It would give us a lot more options for interacting with computers, from VR to mobile devices, to sitting on your couch with a projector, or maybe even a DDR dance pad kind of scheme—and I have reason to believe it could be made as fast or faster than contemporary methods (at least for writing computer programs—not natural language).

"One strategy I've developed to make voice-to-text seem less odd to bystanders is to talk into my phone up to my ear as if I were on a call (rather than talking into it like a microphone as most people do with voice-to-text)."

Wow this is brilliant, and I can't believe this wasn't obvious to me when I've considered voice typing but felt weird about it.

I just got a Note 9 and I love that I can pull out the pen while the screen's off, jot a note and put the pen away without having to unlock my phone. And in the app it can be OCR'd for searching, editing etc. I'm sure once I get into the habit of using this I'll be forgetting ideas less.

If you're worried about bystanders, talking to your phone is always going to be weird...

The actual solution is to not give a shit what others think about what makes you productive.

I told my coworkers that they gonna hear me talk to myself time to time. Im using Jota text editor on android + handsfree earphone to add todo tasks, notes, ideas and so on. It was odd in the beginning.

People will now advocate constantly holding up a device irradiating heat as close to your skull and brain as possible for long periods of intense thinking (which already stress and heat up the brain, naturally). I get headaches within minutes of long or intense phone calls. Voice control looks more like a joke and less like cool future as it becomes concrete. Said this before, but I just like vim better.

It works quite well - though I disagree with his reluctance to hold it up to the mouth like a mic (speaking only) rather than up to your cheek like a phone (speaking and listening). It's usually More comfortable to me. Maybe we are just doing it in different contexts.

Either way, making it immediately obvious you're talking into a device gives the implicit signal "I'm talking, but not to you" so you don't have to do any further signaling, which is usually less clear anyway (head shake, etc.)

I liked this comparison quite a bit.

Has anyone also written a comparison of different environments as they affect our ability to think/write/type?

For example, I love the general experience of working in coffeeshops: the background noise, the occasional chitchat, the generally nice lighting... But the ergonomics are awful, usually, if you are there any length of time.

On the other end of the spectrum, it's really good for me physically to work at a standing desk. But while standing up, it's easy to accomplish familiar tasks, but I find it really hard to concentrate on anything that demands really deep thinking or reflection.

Anyway, I wish there were a roundup of different possible work environments...

And then for really deep thinking and architectural reflection and meandering there's nothing like lying in bed a bit for a pseudo-break.

For note taking in meetings, I've sometimes wished that wireless computer keyboards had a note taking mode.

In note taking mode instead of sending your keystrokes to your computer, the keyboard would store them in an internal buffer. Later, you could tell the keyboard to play back the saved keystrokes to the computer, which would look to the computer like they are being typed.

This is a cool idea! I can see it being totally feasible and definitely useful as a piece of software (not part of your keyboard) toggled by a key combination.

I can imagine that if you aren't a great touch typer, you would run into spelling and formatting errors that would make it fairly unreadable. As you would not be able to address them as you type.

Next time you are in a meeting, try typing with the screen switched off.

Yes, there would be more errors than typing with a screen. But keep in mind that you can look at the keyboard, and that can help a lot. The hardest pat when typing without looking at the screen is correcting errors. I am typing this message looking at the keyboardd, not the screen, and occasionally I miss, but am not usre if I missed one or two keys, and so do not know how to backspace and correct it. Also, a coule times I gor half way into a sentence and wanted to go back a few words, and had no idea how to do it. Still, it would be good enough to provide usable notes.

NOTE: I am now looking at the screen. The part above was typed with my head leaning over the keyboard and looking almost straight down, and has not been edited. I could see some glow from the keyboard at the top of my vision from the bottom of the screen, but all of the text was in the top half, completely out of my vision.

If you don't mind connecting it to your computer via USB, an AlphaSmart is exactly what you're looking for, and it's from the 90s :)

I was hoping this would be a vim / emacs / qwerty / dvorak showdown.

are speaking and keyboard really comparible at all? certainly when coding if I was to speak "open parentheses float width comma float height comma totalArea close parentheses cursor up beginning of line start function comment...". No idea how I would speak camelcase "totalArea" and have the system distinguish between on word and two or how I'd distinguish "cursor up"(move the cursor) from "cursorUp" a constant name etc...

I can imagine context and auto formatting could help but even then I'd expect spoken code to be 10x slower than typed.

I made a similar comparison in 2007

http://lumma.org/microwave/#2007.07.12.2

Reply on news.ycombinator.com