4 ms·
Edgewrite and some of my own experiments
- TBF-RnD 7y agotrying to create a resource that might turn into a collaborative book I find that there are a lot of places where keyboards doesnt cut it - VR - Disabilities - international settings - productivity for tablets My hope is to create a framework in code where experiments can be made and benchmarked. Also to create a loose umbrella organisation that can coordinate the effort.
- ColanR 7y agoEdgewrite looks like the same kind of thing the old Palm Pilots used.
- extra88 7y agoGraffiti, yes. Edgewrite should perhaps be called "Cornerwrite;" in its optimized alphabet, the every stroke begins and ends in a corner and every change in direction is at a corner. https://en.wikipedia.org/wiki/Graffiti_(Palm_OS) https://en.wikipedia.org/wiki/Graffiti_(Palm_OS)
- TBF-RnD 7y agoYes that would indeed by an accurate description!
- TBF-RnD 7y agoYes the author of edgewrite talks extensivly about grafiti but claims some advantages over that system. I'd like to do a web demoable implementation of edgewrite but it's kind of hard for me to get a hand on a Palm Pilot. Have you used that system?
- bobbiechen 7y agoI saw a guest lecture last semester from Rob Haitani, the UI designer on the original Palm Pilot [1]. He talks about their Graffiti text entry from about 00:07:00 to 00:12:00: One thing we learned from our testing was, the absolute speed and accuracy weren’t actually the critical point. It was really the overall experience... So with the [Apple] Newton [handwriting recognition] and other technologies, you would write something, and you would go one Mississippi two Mississippi, and boom it would show up. And if it’s wrong, oh crap. You erase it, and then you try again like one Mississippi two Mississippi. But with Graffiti you’re just going a-b-c-d-e, and if you make a mistake, like your keyboard on your computer, you just have this backspace stroke, and so it’s very fast to recover from errors, so you just keep going. I think there's a parallel here with swiping gestures for words on smartphone keyboards, where I find that when the gesture is misinterpreted, it's frustrating because I'll have to delete the entire word and redraw it, or tap out individual letters. Which just feels bad. As a side note related to that, in the same class there was also a guest lecture from Shumin Zhai, who worked on SHARK and subsequent systems which would eventually become ShapeWriter/Swype [2]. Very cool overview to how the system works conceptually. [1] https://scs.hosted.panopto.com/Panopto/Pages/Viewer.aspx?id=0beb423b-d1a0-4a99-bc52-aa30010e3b2a https://scs.hosted.panopto.com/Panopto/Pages/Viewer.aspx?id=... [2] https://scs.hosted.panopto.com/Panopto/Pages/Viewer.aspx?id=01f1a779-b24a-4686-bea1-aa01015a6627 https://scs.hosted.panopto.com/Panopto/Pages/Viewer.aspx?id=...
- TBF-RnD 7y agoThank you so much for your comment and the links. Looks like a lot of tasty material for me to put my teeth in. Yes it seems that swype's spiritual ancestor somehow is edgewrite / graffiti text entry systems. please check out my rss http://tbf-rnd.life/feed http://tbf-rnd.life/feed if you want to get more info. I have a long backlog and I hope some interesting views as well. Also there'll be code!
- k__ 7y agoI looked at some videos that showed Edgewrite. It seems like a rather quick method, but wouldn't it be quicker if all letters started and ended in the same corner?
- curtis 7y agoI use MessagEase[1] on my phone. I don't think I'm any faster than I would be with the conventional keyboard, but it seems like my accuracy is a lot better. It wasn't that hard to learn the basics, but I still have trouble with less frequent letters and symbols. Part of that is just that I don't text a whole lot so I don't get enough frequent practice outside of the most common characters. Edgewrite and Graffiti both have the advantage of leveraging off of your existing writing knowledge, but I have to say after using Graffiti that it sure seemed like it would be a lot faster to just abandon writing as the metaphor altogether. Actually, I guess that's what MessagEase originally was, since Wikipedia tells me that it was originally released for in 2002 for the Palm. [1] https://en.wikipedia.org/wiki/MessagEase https://en.wikipedia.org/wiki/MessagEase
- TBF-RnD 7y agoI did see and try messagease and it mirrors some of my ideas quite well I'd say. So from a mathematical point of view I really believe in it.
- 333c 7y agoMessageEase seems interesting. I've just downloaded it. The swipe gestures seem like they are slow, but maybe with practice the larger sized targets offset that. This comment was typed using the keyboard. I'm going to keep this installed and see if I can get faster. I also notice that there's no autocorrect, which I miss especially when it comes to typing the pronoun "I."
- curtis 7y agoIt will auto-uppercase stand alone "i", but, yeah, as far as I know it doesn't do general auto-correct. I've never been a fan of auto-correct, so I regard that as a feature. I recognize that that's a minority position, though. Actually, the uppercase support is one thing that I dislike about MessagEase. For swipes you just swipe one way and then back to where you started, which is easy and natural. For taps (your most frequent characters) you have to draw a circle. I've found this to be error prone, and the fact that it's different for the two sets of characters tends to trip me up. If you text as much as a teenager, you could probably get this into muscle memory, but I can never quite get there.
- andrewshadura 7y agoA friend has developed an input method called Aronetis, best suited for 12-button mobile phones. Unfortunately, with the move to smartphones it's become much less relevant.
- TBF-RnD 7y agoInteresting I'll add it to my back log. Maybe there are some interesting conclusions that can be drawn from it. Do you think It'd be possible for me to get in touch with him?
- andrewshadura 7y agoI believe there are some demos still online, but in any case I think he still uses his Google account aronetis ;)
- bobbiechen 7y agoJacob Wobbrock, who created EdgeWrite, also published on gesture recognition with the $1 Unistroke Recognizer [1], which started the $ family of recognizers. If you think there's some sort of gesture-based writing scheme that would do better, you can use the gesture recognizer and focus directly on how to design your gestures instead of having to implement recognition yourself. There's a pretty cool demo with a Javascript implementation in a couple hundred lines here: http://depts.washington.edu/madlab/proj/dollar/index.html http://depts.washington.edu/madlab/proj/dollar/index.html [1] http://faculty.washington.edu/wobbrock/pubs/uist-07.01.pdf http://faculty.washington.edu/wobbrock/pubs/uist-07.01.pdf
- TBF-RnD 7y agoThat's right I saw that. Didn't have time to add it to text yet though. I did consider it for an implementation. Seems like Jacob Wobbrock went on to become quite influential after his work on this.