Dumb Idea of the Day

This morning’s Dumb Idea: a multi-touch Internet tablet that detects when I lay eight fingers in a row.  This will cause a virtual keyboard to pop up right where my fingers lay.  Instead of moving my fingers to adjust to the virtual keyboard, the keyboard will be warped to fit where I put my fingers.  Then, I just start typing… each key pressed has some deviation from the middle of the key I’m trying to hit, and the keyboard is adjusted and warped as I type to take into account the drift of my hands.  Even the size of the virtual keyboard will depend on my finger placements – wider spacing implies a bigger keyboard.  There will be no need to keep my hands placed together, or even lined up horizontally.  Ideally, this is integrated with some predictive software that heuristically guesses which key I meant to hit when I miss badly.  The virtual keyboard should be translucent so I can see what’s under it.

Next year we’ll start seeing “Internet tablets” that have low-power touch screens that are very readable in bright sunlight.  Running Ubuntu Netbook Remix, they will be powered by Nvidia Integra ARM-based processors with rocking graphics that extend battery life several fold while dropping cost.  The killer application for these devices will be e-book readers, like the Kindle application that I have installed on my Ubuntu machine right now.  The main drawback of these tablets will be the lack of a keyboard.  A warpy multi-touch keyboard is just what’s need!  I suspect I could type just as fast, maybe faster, than on a real keyboard.

You might want to have a protective cover for your Internet tablet, the way people have covers for their iPhones.  I want mine to be clear and thin enough to let me interact with the multi-touch screen.  I also want it to have ridges in the shape of the keys of the keyboard, so I can feel them.

I’m a slow reader… sort of.  Actually, I’m only average, at about 250 words per minute.  However, for someone who studies computers and reads all day, my reading speed is terrible, and painful.

Lately, I’ve found a partial solution.  My eyes never translated printed characters to sentences very efficiently, and that is my bottleneck.  However, using the old IBM Eloquence text-to-speech synthesizer, I’m able to create audio-books that run at about 530 words per minute.  The voice sounds so smooth and slow, running at close to the speed I think at.  It’s a wonderful feeling after all these years to find a way to absorb books at the speed I want them.

This is how blind people feel.  A decent Braille reader only reads at about 60 words per minute.  Blind children are encouraged to learn Braille, but they naturally find their listening speed zooming to over 800 words per minute, without any extraordinary effort.  As a result, most blind children are not learning how to spell words, only how they sound.  They are in fact illiterate.

I can imagine a day when computers learn to talk and listen, just like a person.  On that day, it is possible that children will lose any significant incentive to learn the crazy way we spell words in English, and instead prefer natural speech interfaces.  Our children could literally forget how to read.

I’ve published the very first version of ShareALot.org, which is a site to help the unemployed work for each other.  It basically is a tool to try to implement my dumb idea where we use the internet to help the unemployed.  I could use good feedback, so if you get a chance, check it out.  Thanks!

I’m interested in doing my part to help the ailing economy by creating a barter-like web site.  You may have seen recent interest in sites like U-exchange.com. I find these sites very interesting, but hugely limited.  For example, one young girl is trying to find a way to pay for her braces to be maintained, since her father can no longer afford the dentist.

She says her braces are falling apart.  She’s willing to trade cleaning services, and that’s great, so long as there’s a dentist near her who needs their house cleaned.  In general, simple two-party barter trades are too difficult to set up, which is why we invented money.  But what happens when you have no money?  What this girl needs is a way to accept an IOU from any of the millions of working mothers who are desperate for some day-care and cleaning help, and to be able to get her teeth fixed by passing along those IOUs in trade, or to sell them for enough cash to pay for the work.

I think this is doable, and can be done legally.  The rest of this post is a long-winded brain-dump of my first thoughts on the boring technical details… They are really rough.  I could use help improving them.  Anyone interested in helping, just e-mail me at bill@billrocks.org.

Read the rest of this entry »

The Stimulus Plan ambitiously aims to double the installed base of alternative energy in a short three years, so these should be exciting times.  However, just throwing money at the problem is a bit like saying “Anyone who wants to invade Iraq will get a 50% tax credit on their expenses.”  It wont get the job done, at least not done right.

For example, First Solar just broke the $1/watt barrier.  At this manufacturing cost, we could dramatically reduce our dependence on foreign oil while reducing our greenhouse emissions.  However, production at this price depends on cheap availability of a metal called tellurium.  First Solar bought about 4% of the world’s supply in 2007, probably enough for a gigawatt or two of new solar panels.  To make a huge dent, we want to scale up to a good fraction of a terawatt, and the math doesn’t work out.  Also, First Solar has little interest in selling it’s panels for under $2.50/watt, and Obama’s plan gives them reason to believe they can maintain that price.  Dell sells dozens of user-customized models of PCs at under 10% margin.  Why does First Solar need > 100%?  We also have limited production of Indium, a metal used by competitor NanoSolar, who also produce super-cheap solar panels.

Scaling up these technologies rapidly should involve government-scale planning, the way we scaled up military production in WWII.  Take the best technologies (meaning cheapest for the most part) in wind, solar cells, solar thermal, nuclear, geothermal, and cookie-cutter replicate them across the land.  Instead, we’re just going to give First Solar and others a reason to continue charging too much for their products, while expanding slowly, and at tax-payer expense.