14700622
story
Comments: 125 + - Hardware: BlindType — the Amazing Keyboard of the Future on Friday July 30, @12:13PM
background: url(//a.fsdn.com/sd/topics/topicinputdevices.gif); width:70px; height:68px;
inputdev
background: url(//a.fsdn.com/sd/topics/topicportables.gif); width:53px; height:54px;
portables
kkleiner writes "BlindType has created a new touchscreen keyboard program of the same name that changes size, orientation, and position to match your wandering fingers as they type. BlindType also features some of the most impressive typing correction software I've ever seen. The result is a practical touchscreen interface that knows what you meant to type, even if you make mistakes. Lots of them. In fact, you can type without looking at the screen at all."
Related Stories
Everyone is entitled to my opinion.
All trademarks and copyrights on this page are owned by their respective owners. Comments are owned by the Poster. The Rest © 1997-2010 Geeknet, Inc.
How is that novel? (Score:5, Funny)
I've been typing without looking at the screen for my entire life!
Re:How is that novel? (Score:5, Funny)
I wonder what happens when you want to type garbage or inane abbreviations: e.g. lol omgwtfbbq brb afk QQ.
A virtual hand probably materializes and slaps you.
Parent
Re: (Score:3, Insightful)
Just add 'lol' 'omgwtfbbq' 'brb' 'afk' and 'QQ' to the dictionary. I highly doubt that they didn't realize that people would want to add words, names, and places that they don't know about. Same system to add those can be used to add all the annoying text speak you want.
Re: (Score:2)
Same with any typing correction features, it only works on proper English (or whatever dictionary used).
It will be useful for typing more formal things like documentations, reports, etc. Not for chatting.
Re: (Score:2)
Why? You'd shortly have added all your shorthand and abbreviations to the custom dictionary, and then it'd be perfectly happy with informal use.
Re: (Score:2)
Same with any typing correction features, it only works on proper English (or whatever dictionary used).
There was the famous CNET blooper where an iPhone wouldn't let Brian Tong type "kewl", thinking he intended to write "kewpie".
Re:How is that novel? (Score:5, Interesting)
I wonder what it looks like if you try to code with it.
for (i=0; i4; ) {lease(r3,i); go( &i);}
becomes:
For I pop Ike. O Pleasure I'll goo You.\n
maybe?
Parent
Re: (Score:2)
Re: (Score:2)
how about a laptop keyboard with a 2mm stroke for the last 7 years? it's pretty damn close.
Re: (Score:2)
Maybe he is blind?
Re: (Score:2, Insightful)
Summaries published by kdawson ought to have some kind of warning label, perhaps something like "Warning: This summary may not reflect the contents of the related article!"
Re:How is that novel? (Score:5, Insightful)
They do, where it says "Posted by kdawson". Assuming you aren't new here.
Parent
Re: (Score:2)
Can't believe it hasn't been done (Score:3, Interesting)
Re: (Score:2)
What happens when someone else uses the keyboard though?
I often have other people read text messages aloud to me while driving and then dictate the response. I don't think this "amazing keyboard of the future" is for me. Next thing you know its either trying to account for someone else's waggling on your device, or its entirely unusable to someone else because it accounts for your waggling too much
Re:Can't believe it hasn't been done (Score:4, Interesting)
Parent
Re: (Score:3, Interesting)
Same here. I actually prefered T9 over my smartphone keyboard, but I definately prefer the physical thing over any touch interface I've encountered.
Re: (Score:3, Interesting)
Re: (Score:2)
From all information that has been posted so far, there isn't any learning going on, except for custom words. The demonstrations they've given in person don't seem to have involved any prep either. So it would seem that it would work fine for your use case.
Re:Can't believe it hasn't been done (Score:4, Informative)
SwiftKey comes close, and has really good predictive text guesses that are usually right and will let me pick the exact thing I typed from the displayed list if I am typing a formerly unknown proper noun, for example, but it has some minor problems on my Nexus One that are really frustrating.
First, it tends to recognize keypresses near the lower right corner as backspace or enter and on the bottom row as space keys. It's very smart about character keys, but doesn't apply the same analysis to punctuation keys - it should be obvious that I wouldn't type three characters then space then a garbled half of the word - it's like it stops processing when it thinks I meant to hit a space bar. Duh. Blindtype seems to be smarter about that.
And second, it occasionally seems to fail to register or filters out the first 1-2 characters of a word I am typing if I type too fast. It's almost like it thinks they were so quick that they were mistakes. This is very frustrating and needs to be a configurable option for fast thumb typists like myself.
These are the reasons I'm looking forward to BlindType, or at least a fixed version of SwiftKey. It seems from the BlindType demo videos to at least address the first problem because it processes a set of keypresses as a batch. Whether I'm faster in practice with BlindType, SwiftKey or something else will remain to be seen.
And yes, I've tried Swype too and have it installed but don't use it everyday. It's nice and quite fast if I'm sitting there and looking at my screen, but I can't use it at all when I'm walking around or otherwise multitasking, which is why I'm so eager for BlindType.
Parent
Re: (Score:2)
After a few minutes of typing on an iPad I could see the need for this - I agree that it seems like a very natural and intuitive step to take when you have a potentially free-form keyboard layout.
I hope they will be able to get it onto the system in such a way as to replace the OS keyboard rather than merely have it run in its own app (so you have to copy and paste text out of it).
The iPhone kind of does... (Score:2, Interesting)
I figured that the Android/iPhone keyboard would look at finger movements on each key to try to see if you pressed in the center like you wanted that letter or far to the side like you didn't and adjust accordingly much like this
The iPhone keyboard does actually take a lot of slop into account - if you type and shift to accidentally press other keys, the final word will correct based on keys that were almost where you hit, so that you don't have to be totally precise - the correction is pretty good on the i
Re: (Score:2)
Re: (Score:3)
It didn’t take long to make. It just took a while for somebody to conceive of it.
It’s not just checking finger position relative to the keys, either. It’s creating a “pattern” for the word that you type, then comparing that pattern to the patterns of known words and picking the best match. It is insensitive to scale or angle, so I’m guessing they’re using distances and angles between each tap. But how...
I imagine that, just like matching noisy data to a best-fit str
Feh (Score:5, Insightful)
I still find a touchscreen keyboard to be a bit wonkey...for me, it isn't an accuracy problem, but a tactile problem.
Re: (Score:3, Insightful)
With the auto correction on most touch phones these days typing out complete words and sentences isn't really a problem once you get used to it. Especially hard to get past for me was wanting to hit backspace to correct it when you see mistakes, if you just plow on through it will usually fix the errors as you go. The problem is trying to type things that aren't complete words; email addresses, URLs, abbreviations, uncommon names and punctuation are all still very difficult to type and I don't see how thi
Re: (Score:2, Insightful)
It might provide an acceptable solution to the person who never learned to type or spell.
Failure in naming the device (Score:2, Insightful)
Re:Failure in naming the device (Score:5, Informative)
If you had watched the video you would have noticed that this is mainly for mobile devices with no physical keyboard (read: zero tactile feedback). If you can take a mobile device with an on screen keyboard and type perfectly fine without looking at it you've got great motor skills. But what happens if it is slightly askew since your fingers are drifting slightly due to no tactile feedback? This fixes that and allows you to literally have no keyboard displayed on the screen and still type coherent words and sentences.
This isn't about learning to type on a real keyboard, it's about not needing one all together. Just know the basic layout and type away on the screen with no keyboard visual.
I'm seriously hoping for one to come out with a small camera you mount on the screen that watches your fingers and allows you to type without the need to project a keyboard. Spacing wouldn't matter anymore and you could just type away with your eyes closed and the software is smart enough to guess what word you meant.
So BlindType is pretty spot on. You can't see a keyboard (blind) and you can't feel one yet you're typing just fine.
Parent
Failure in reading the article summary (Score:2)
Gee, that's SURELY new... (Score:3, Informative)
In fact, you can type without looking at the screen at all.
Gee, that's new! Typists from decades ago were able to do just that. It was called "training" and "expertise".
Seriously, though, I expect two distinct problems with this:
1) How well will it handle "non-US slopiness"? Sloppy typing in Spanish (etc.) is quite different from typical english-language slopiness.
2) IT'S NOT A MODEL M KEYBOARD!!! There, I said it. I don't care for "the keyboard of the future" if the "keyboard from the past" is still alive and well and functioning nicely. Actually, make that "the keyboard from the past and present". :-P
Re: (Score:3, Informative)
Re:Gee, that's SURELY new... (Score:5, Funny)
If you can plug a model M into your smartphone and use it conveniently I will be impressed sir.
I'd rather crush my work smartphone that keeps ringing all day with my Model M keyboard. :-)
Parent
Re: (Score:2)
Agreeing with the Model M comment.
Also: obligatory one-handed typing joke, taken as read.
Also, less facetiously, I wonder how well it accommodates switching scripts/Unicode ranges or special typing for codepoints. (Haven't yet rtfa.)
Re: (Score:2, Insightful)
Re: (Score:3, Interesting)
From what they've said, it appears to be language-independent. It's more to do about interpreting why you touched the screen in a certain place, so what language you're trying to type... it's just a different dictionary to match against.
Re: (Score:2)
Gee, that's new! Typists from decades ago were able to do just that. It was called "training" and "expertise".
Not looking at the "screen" (i.e. platen) while typing was generally due to the fact that you were reading off copy and typing it in to create another (typically more legible) copy. Now that computers can do copy and paste without a man-in-the-middle, we do what manual typists did when coming up with original work: we look at what we're typing so we catch our own mistakes. Reading off paper copy still occurs, but is less common. (Though sometimes you're reading off electronic copy that prevents copy & p
hrmm (Score:5, Funny)
Not for me yet (Score:2)
Call me when it can accurately decide when I meant when typing commands using vi or emacs.
Re:Not for me yet (Score:4, Funny)
I've tried this. It constantly replaces VI commands with "I'm an erudite prick". I didn't have the courage to try emacs...
Parent
Re:Not for me yet (Score:5, Funny)
Oh, in that case, probably "I'm an erudite GNU/prick".
Parent
Re: (Score:2)
Can we call you when someone figures out how to parse that comment? ;)
Swype is better (Score:3, Interesting)
Re: (Score:2)
I can believe that Swype might be better (I haven't tried it), but this might be easier to learn. Will be interesting to see which attracts more users.
Re: (Score:3, Interesting)
Swype is great, and comes pre-loaded on many phones [Droid X, Galaxy S series of phones]. Only problem is that it adds to your dictionary anything that you tap out. That includes garbage from URLs like the end of a bit.ly link. Without manually deleting garbage like that, it starts to become unusable.
They should allow you to see and edit your dictionary like you can with T9.
BlindMarketing (Score:2)
Also, the blindness part... in normal keyboards you should see the monitor, you get some input both from the physical keys and of course, from what is being displayed in the monitor, and that helps writing without watching the keyboard. But in a tablet (as shown in the article) you only have the monitor, you see the keyboard becaus
Spelling Nazis Rejoice? (Score:2)
As this program appears designed to interpret what you want to say based on actual, English-language words, it would be interesting to see how it would handle poor spellers.
Personally, I am probably somewhat of a spelling nazi, as I cannot stand how inept some persons (seemingly the younger generation--get off my lawn!) appears to be at spelling. If this is released, I would imagine that poor spellers would either (a) be forced to finally learn how to spell (again, get off my lawn!), or regrettably more
Re: (Score:2, Funny)
Victory for iPhone users (Score:2)
Now you can type actual messages on an iPhone [1] without wanting to pull out clumps of your hair!
[1] or any physical keyboard-free smartphone
How I would code it (Score:3, Interesting)
Let's think of how they might have designed the algorithm for this. In the videos, it looks like it is treating one word at a time, so let's consider that scenario here as well. I would define the problem as assigning to every possible word the probability that it is the word the user intended. I would use Bayes' theorem [wikipedia.org] to achieve this.
First, assume a prior probability distribution over all words. Words not in our dictionary, and words of the wrong length, we give probability zero. The remaining words can be assigned equal probability or, better, a probability proportional to their frequency in the language. If you want to be fancy, you could have more sophisticated models that knows which words are likely to come after others and such things.
Second, for each candidate word, what is the probability that the user would tap the screen as they just did? A model for this could be that the location of each tap is drawn from a Gaussian probability distribution centred at the intended letter with a known standard deviation and that each tap's deviation from its target is independent of the others'.
Finally, Bayes' theorem states that the posterior probability (the one we want to calculate) of each word is the word's prior probability (from step one) times its likelihood (the probability of step two).
To implement the arbitrary position, orientation and size of the keyboard, we redefine the problem from finding the probability of each candidate word to finding the probability of each tuple (intended word, keyboard position, keyboard orientation, keyboard size). Make it simple; have each element of this tuple to be independent of the other and use flat distributions for all keyboard parameters. To choose the most likely word, you could either pick the word of the most likely tuple or, more correctly, for each candidate word, integrate over all possible keyboard parameters (weighted with their prior probabilities) to get the probability of that word. Likewise, you could introduce the standard deviation of the taps as another element of the tuple, with its own prior distribution.
I suspect this method is a bit to heavy on computation cost and power consumption, so if you cannot find a clever way to do it fast, you might have to cut corners in the rigor (or do something completely different).
(Can I come work for them now?) :-)