In Defense Of The Stylus
techcrunch.com
techcrunch.com
We learned that most people who were happy with pen-based computers had gotten them at a steep discount. They felt that the products had shortcomings but that was ok because they were so cheap.
The only two exceptions--pen-based computing products that had found significant success at their MSRP--were Wacom and Palm/Handspring.
The key to the mystery was handwriting recognition. 100% recognition is impossible, even for humans. But while we accept human misses, there is pretty much no tolerance for misses in a computing interface. With a miss, the value of the pen input (speed and natural feel) is lost in the need to correct.
Wacom succeeds because it does not even try to do handwriting recognition.
Palm/Handspring succeeded by inventing new motions for per-letter entry that people had to learn (called Graffiti). This reversed the user expectation: if the recognition failed, many users would assume it was their own fault for not doing the Graffiti entry properly. They wouldn't blame the machine, so their opinion of its value did not suffer.
Styluses are still hugely popular for animators, artists, and some graphic designers. Wacom is still going strong. But any computer that relies on a stylus for text entry is probably doomed to fail.
Going from a stylus back to a touchpad feels like going from pens to finger paint. Mice feel clunky as well. Try it out!
By no means should a cell phone require one, but there are vistas opened by styli that we may want to explore after the novelty of "holy cow, tablets! they finally work!" has worn off.
For artists and designers, the stylus will be used for many years to come, but for casual, everyday use, it simply isn't user friendly. While it may be natural to draw or write on a screen with a stylus, jabbing at buttons on a screen with a stylus is no more intuitive or natural than jabbing at a screen with a finger, and in many use cases (scrolling, zooming, panning) is much less natural. The case for the stylus becomes even weaker in the face of technology like Siri, which elegantly solves the problem of text input.
The Bret Victor "Rant" mentioned in the article makes the case for exploring more dynamic, tactile, interactive experiences, decidedly not limited to one mode of input like the stylus. I, for one, hope for a future that is much more ambitious than the stylus, with new forms of interaction that combine hands, voice, and beyond. Things like this http://www.chrisharrison.net/index.php/Research/PneumaticDis... and this http://www.youtube.com/watch?feature=player_embedded&v=x... that truly push the boundaries of human computer interaction.
On the other hand, it's ridiculous that the only way to digitize these notes is by scanning, or typing them in, or using linux-unfriendly Livescribe pens that require special paper. For centuries we've interacted with documents using styluses. For doing math, for marking up papers, for scribbling notes in the margins, and for countless other activities it is simply _the_ right tool for the job. People like Steve Jobs declared unilateral war on styluses because they were being used to design suboptimal interfaces. Jobs had a point, but in rejecting the stylus outright, he threw the baby out with the bathwater.
For consumption tasks, browsing the web, etc, a touch interface is far preferable. For those tasks where touch isn't sufficient, a stylus isn't at all better. A stylus just doesn't work without the tactile response of paper (which the textured Wacom tablets can offer but a portable device can't).
Direct touch gave the biofeedback needed to make operating the device personal. Your touch makes things happen, rather than you touching it with an 8 inch pole.
It seems an irrelevant detail to a logical mind but this was key element to iPhone's success.
There is this good PBS documentary on psychology of marketing. If you don't have the time for the whole thing, fast forward to 0:48 and watch for one minute—it is very representative of the marketing mindset I am talking about.
I'm a grad student in math. If I'm taking notes for a class, I prefer to have my laptop and LaTeX the notes on the fly.
If I'm working on my research, I find working with pen and paper is a psychological help: it's a good interface for working on math. The only problem is that pen and paper is a bad system generally, for a couple of reasons:
1) most of the notes ends up being crap/scratch work, it'd be really nice to be able to copy and paste the good stuff out easily.
2) Paper is heavy, I always have to decide which of my notes to carry around.
3) It's easy to loose your notes.
I'd love to have a tablet with a stylus good enough to write mathematics with.
http://www.pcworld.com/article/188223/the_long_fail_a_brief_...
Maybe the lack of success has been due to poor implementation, but I wonder whether styluses are just inherently problematic:
1. Truly reliable handwriting recognition always seems to be just over the horizon, and I wonder whether it always will be. Honestly, I often can't read my own handwriting. Sure, you can store handwriting as "ink," aka a bitmap, but that's like choosing a fax as your file format.
2. Switching from stylus to finger to (possibly) keyboard is awkward. You need someplace to put the stylus, and it makes one-handed operation pretty difficult.
That said, I actually think I'd enjoy having a stylus for input sometimes. I hope that someone eventually figures out how to put all the pieces together. (Too bad that Microsoft Courier didn't make it.)
But yeah switching is a pain. and the interaction vocabulary has to be rewritten for stylus input.
As to iPad's the same issues shows up, using a Stylus on top of a screen is less accurate than using a separate tablet, not to mention it's resolution is terrible for creating real art. Sure, it's better than a cellphone, but fingerprinting is plenty for most people's level of artistic talent.
PS: My sister is a graphics artist and she get's sub-pixel accuracy with a cheap qualcom tablet. Covering up part of the immage your working on is reasonable, but when using pen on paper she often rotates the paper to minimize the effect which slows things down with clunky computer controls.
Does the reporter honestly think that he hated the stylus?. Of course not, he just realized that it was not practical at the time, it was not practical to carry pounds of extra weight for using stylus as you need a backplane active matrix that consumes battery all the time.
In the video you see Atmel people drawing and the enormous lag it has, this was unacceptable for Jobs of course.
I made my own homebrew software for the DS lite that used the stylus, for instant(no perceptible lag)sketching, with antialiasing and 3d. It worked very well, but after a month of using it, the screen wore out because the screen is resistive and plastic. A little expensive note pad!!
That said, imagine if styluses (styli?) were mass produced and available as ubiquitously and cheaply as pencils. Lose one? No problem, there's another one on the table over there.
In that case, I can see styluses catching on big time, and as noted in the article, their precision would give them huge inherent advantages over fingers.
Stylus slots are not new. All the iPaq-like PDAs had them as well, and people still complained about losing them. I never lost mine but it was a relatively common complaint. (I still have it in fact, tucked away in a drawer in my "smartphone" from 2004[1])
When I first got my Wacom drawing tablet, it took me about an hour of trying out handwriting in GIMP to get used to this. As long as you hover the stylus only a few millimeters above the tablet, you can even see the mousecursor move around to where you're pointing (takes a little bit of practice, that).
So, while nice, it's definitely not strictly necessary that seeing what's "under" the styles happens on the same plane as you're drawing on.
I found editing maddening on an iPad because of the inaccuracy. You can roll your finger to adjust the cursor position, and the magnifying window helps, but it's not as good as simple cursor keys (disclaimer I was just trying it out in an apple store, so there may be ways around this problem that I don't know of).
(Disclaimer: I haven't tried it yet. Would be interesting to hear others' feedback.)
No luck so far!