The Innovator's Opportunity with OLPC


Way back in ancient history, in 1997, Clayton Christenson of Harvard wrote The Innovator's Dilemma, explaining a seeming paradox in high-tech.

Why is it, he asked, that in many product areas, innovation does not come from established companies, but from startups? Why is it that companies doing exactly what the experts tell them to do, that is, to listen to their customers, are constantly going under? The short answer is, because the next big market isn't with your current customers. It's with a vastly larger group of would-be users who couldn't afford your previous products, or couldn't carry around the huge devices of previous generations.

It would be easy to go too far with Christensen's analysis. As a counter-example, Intel has been doing fine for many generations of product, because it is able to make effectively bigger but cheaper products in a roughly constant physical size range, and its existing customers, along with ever-more-numerous startups, then seek out the profusion of new markets for new processors, memory, and all the rest.

People like to talk about autos vs. the buggy-whip manufacturers, or dinosaurs vs. mammals, as analogies for tech competition. Those are OK as far as they go, but remember that mammals didn't kill off or even out-compete the dinosaurs. Some of the dinosaurs turned into birds, and are still here with us in profusion.

Why do I bring all of this up? Because Mary Lou Jepsen brought it up in a video about the new Pixel Qi screen technology:

She says, "The cool thing about the Pixel Qi technology is, you know, poor kids in Africa got it first. You know, it's the classic Innovator's Dilemma." Mary Lou proved the market for her innovations in an area that most people didn't believe could exist, and now her company is in hundreds of market segments.

In our case, the new market is hundreds of millions of children who couldn't possibly afford conventional computers out of current income. Nevertheless, with education and the resulting economic growth their countries should be able to afford them as an investment in their future. Some people are still arguing whether that is so. I take a different tack. It isn't a question of existing facts on the ground, but of ethics, the obligation to change those facts. Ending poverty is worth so much that we should all simply decide to make it so.

She went on to describe the new 3Qi technology, which is even better, and has nothing left of the original OLPC XO screen. Laptop Magazine reviewed one of the 3Qi prototypes hacked into an Acer machine side-by-side with a conventional screen.


People ask Mary Lou, do we really need a contrast ratio of a million to one, or 120% color saturation? Isn't 100% enough? Well, in fact, those are great for watching movies, but they aren't the specs for a book reader. Her focus is on being able to do both well, including reading in sunlight. (People always ask whether we need hardware improvements, since they obviously aren't required for current applications. I was on a panel in 1981 where we were asked what anybody would do with more than 640K of memory.)

Another point is to encourage designers to use architectural ideas from the XO in order to save power and extend battery life. This means having an integrated screen controller and integrated wireless that can run while the main processor turns off. So if you think that the XO inspiring the netbook market was a big deal, you haven't seen anything.

Mary Lou has more technical tricks up her sleeve for better displays at even lower power, which I don't want to get into here. If you want the tech, go watch the video. And even if you don't want the tech details, go watch Pixel Qi vs Kindle vs Toshiba R600 vs regular LCD tablet, where she demonstrates the 3Qi screens in comparison with other technologies.

CGI rendering of the XOXO.

So where does that take us? Aha! It takes us to the XO-2 screens. According to the designers, that means multi-touch haptic screens usable as display or keyboard.

In English, 'haptic' means that when you push on the screen, it can push back or vibrate, so you can tell what you are typing, and more importantly where the centers of the keys are. Essential for touch typing. Multi-touch means that you can type key combinations, or play multiple notes on a music keyboard, or activate several triggers in a game at the same time.

Mary Lou says that she doesn't yet know what screens Pixel Qi will supply for the XO-2. It's up to the designers at OLPC to decide what they want, and then the two teams together can discuss how to get that at the best price.

Us haz teh bestest toyz.

Keep current on XO innovations - subscribe to OLPC News via RSS Feed, Email or Twitter


Related Entries


So what's with this story?...
Pixel Qi is quite almost 4 months now, XO-2 is not even in the drawing stage and "inivator's dilemma" is hardly the OLPC case.
Is there something more to it?...

I think "Innovator's dilemma" still holds true. The trick is in not looking at a large company like Intel (or IBM, or Oracle, or Cisco) as a monolithic enterprise or to look at innovations such as "Ethernet" as a one time deal, but to look at smaller, interdependent pieces. After all, Ethernet has had a long run from 1972 to 2009 and its gone places. Yes Ethernet is commodity, but Ethernet is also quite innovative!

Nicholas Carr wrote "IT Doesn't matter" ( back in May 2003 followed by his book "Does IT Matter?" (talk about priming the bookseller pump!). The interesting thing about the article or the book isn't what Nicholas Carr wrote, but what others wrote back about. See the responses at Particularly interesting is Bob Metcalfe's response (hence my example of Ethernet).

On a related note, I was walking around the expo floor with an XO at OSCON 2009, and a Intel employee stopped me and asked me a whole bunch of questions, and he listened with wide eyes. He was most amazed that such a wonderful project existed. He was also a bit disappointed that he hadn't heard of this before and wondered if Intel could help. I didn't have the heart to break it to him. So, I just gave him a bunch of URLs to look up :-)

I wish I knew when we could get some of these screens. The OLPC screen is its biggest selling feature, and the screens Pixel Qi is making are much better, but that doesn't help if I can't get one.

Very interesting and informative article. I watched that video carefully, noting that she said these screens were on their way to many OEM vendors. Prices will come down if millions are being made, versus tens or a hundred or so thousand just for OLPC. Thus, OLPC will end up paying less per unit since they will one of many buyers. Makes sense to me: good news.

To kind of sidestep the main topic just a bit, I'm sure hoping that this new netbook has one of Mary Lou's screens . . .

Note the rubberized keyboard (like the XO-1/1.5) and the ARM chip (someone else who blogs here regularly is big on these . . . same as what's in an iPod Touch and/or iPhone,) Linux OS and an estimated 10 hour battery life . . . all for around $200 when they show up in the US.

Hi Guy's,
I think The OLPC screen is its biggest selling feature, and the screens Pixel Qi is making are much better, but that doesn't help if I can't get's result is very good..

With a little imagination, that fancy OLPC tablet has existed for a few years now, and retails for about $200 in the US. It’s called the iPod touch..

XO Tablets for Sale

Buy Your XO Tablet on
OLPC is selling the new XO Tablets on for just $149. Buy yours today!



Recent Comments

Community Forum