I found http://www.senseboard.com/ a few years ago. It's a keyboard-substitute that clips over the back of your hand, and interprets the tendon movement as you "type" into text.
The device is elegant and compact, but better yet it's an elegant solution to the problem of mobile typing/data entry. It's the missing link for a new era of mobile computing - it completes the trio: Cellphone, Pocket Projector, Senseboard.
Use just the phone, or the phone + projector, the phone + senseboard, or all three. That trio would be smaller, more portable, more comfortable and more flexible than any netbook.
Now is the perfect time; screen size and data entry are the major limitations for mobile devices. Powerful PDAs and cellphones are here. Phones with 'proper' operating systems. Pocket Projectors are just appearing to get around the limitations of small, low resolution displays and they wont stop at being 640x480 and dim. Soon they will hit netbook screen quality. But data entry is still limited to those foldaway keyboards that are both big and unwieldy compared to a phone, full of delicate moving parts, and still small and cramped to type on. Senseboard is a much better alternative.
The big problem is - it's vapourware. Best new product of 2001. Nothing but a couple of minor press releases since then.
"So what?" you ask. Well, I've mentioned it here a couple of times and had no discussion. I decided to submit it directly as a link to publicise it a bit - it plummeted from the new submissions page with no comments and no votes. That's life, eh? I would probably have left it there except I said this:
http://news.ycombinator.com/item?id=570994
and got 27 upvotes. 27! For a pernickety grammar correction one-liner! Right now, that contrast is really getting under my skin. I know there are technical people here, people who've used the Twiddler, successful business founders, hardware and software developers, VCs, people who might see what I can't, might also care, and might have influence, so I Ask HN:
Do you think the Senseboard is as cool and potentially game changing as I think it is? Do you think it could be a successful product?
Can we outsiders do anything to encourage it into existance, and - should we?
(I have no affiliation with the company, I'm just tired of seeing what seems to be first class ideas like this sidelined while streams of clone digital cameras, media players, GPS receivers, cellphones and so on pour into the market as if the world wont be satisfied until there's ten models for every human - and that needs to happen yesterday!)
I would put relative probabilities of those two as somewhere around 1:9.
I look at this thing as an engineer and I see a signal processing problem that means it's probably pretty easy to build something that's right about 70% of the time, and with great effort you could get to 85%... but that's terrible for an input device. Maybe I'm wrong. But I also know for a fact I've encountered tons of people who have faced other problems like this, and in their boundless optimism are sure that the last few percent are just a matter of trying harder, and some of them go on to blow millions on "trying harder" on what is fundamentally impossible. The truth is, electronics really don't care how optimistic you are.
Most of the 'first class' ideas you think are being overlooked are actually the second case. 3D interfaces, a whole whackload of input devices (including the fun case of exotic video game console interfaces), energy sources, new circuit types that are going to blow silicon out of the water, and the list just goes on and on. Most of them just plain don't work like the advocates said they would.
Some of the advocates are honestly wrong. Some... are not.
Just about the only thing that factors in this thing's favor is that exotic input interfaces have historically faced a very steep uphill battle. Dvorak and other alternate keyboard layouts have gotten very little traction, and that involves no extra hardware at all. I've used a couple of exotic input devices or methods that work perfectly fine but stand no chance of general acceptance because people have no interest in learning how to use them.
(I would point out that I've left open the possibility that it does work. But I'd want to see a lot of evidence not coming directly from the company. Everybody always claims awesomeness.)