Thinking about the sub-$100 laptop

I’ve been meaning to weigh in on recent developments on the sub-$100 laptop project Nicholas Negroponte is spearheading. While I join nearly everyone else in the hope that millions of net-accessible sub-$100 laptops find themselves in the hands of students around the world, I share the skepticism some of my ICT4Dev colleagues appear to have about the project. As he so often does, Russell Southwood at Balancing Act beat me to the punch with an article in this week’s issue titled: “Bottom-of-the pyramid markets – the sub-USD100 laptop and the search for the holy grail” – very much worth a read.

The article makes the helpful point that, despite somewhat breathless coverage, the $100 laptop being designed by the MIT Media Lab is still pre-prototype. While sketches of the device are exciting and thought-provoking (the Media Lab has always offered great demos…), there are real doubts about whether some of the technology exists to make the device possible as something other than a prototype.

Most ICT4D geeks – myself included – think that the most difficult part of producing a sub-$100 laptop will be the display. The specifications for the laptop call for a full-color screen capable of changing mode into a very bright black and white display for use in full sunlight. Such a display needs to draw very little power and be highly resistant to breakage and shock. The recent News.com article on the project suggests that Negroponte is planning in using a 12″ flexible e-ink display which has a target price of $12.

e-Ink is a technology that’s been the next big thing for almost a decade – those of us who routinely hang around the Media Lab have seen it demo’d and discussed since the mid-1990s. Unfortunately, bringing it to market has been a challenge. After several statements promising handheld devices using e-Ink technology in the near future, there appears to be an actual device available, the Hanlin eBook, using a 6 inch 800×600 pixel black and white e-Ink display. (The Sony Librié also used an e-Ink display, but was extremely unpopular due to absurd DRM restrictions on its use.. interested in buying a technology that destroys any content you purchase within 60 days? Didn’t think so.)

While I’ve heard great things about the display quality of e-Ink technologies, nothing I’ve heard about the tech suggests that it’s available in anything other than black and white. My understanding is that e-Ink is built around molecules that are white when no charge is applied to them, and turn black when a charge is applied – the molecules live in a film which is overlaid on a backplane – either an active matrix backplane, like in a cellphone or laptop display, or a flexible backplane that behaves more like paper, but has lower display quality.

(Andy Carvin, blogging a recent talk by Negroponte on the laptop describes the screen as “a dual mode display – a 7″ screen, as well as a reflective display that can be read in bright daylight”, currently costing $35 – unclear whether or not this is a color display or black and white.)

With absolutely no data to back up this thinking, it strikes me that Negroponte and crew might be proposing a color active matrix display with a layer of e-Ink over the top. In indoor situations where the laptop had access to AC power, it would function in a color mode, but when running on battery or outside, it could use the active matrix to drive the e-Ink layer, using less power. Whether this would actually possible (wouldn’t this require e-Ink to have a third state – translucency?) and how it could be possible at $0.10 per square inch, I don’t know…

Southwood’s critique of the sub-$100 laptop focuses less on the display and more on the vast graveyard of low-cost technology for the developing world projects he, I and others have visited far too often. A few years back, the project everyone was excited about was Simputer, the linux-based handheld computer designed by Indian IT experts and licensed for manufacturing by firms around the world. After a great deal of enthusiasm and hype, the product that came to market was more expensive than promised, less powerful and has never been produced in large volumnes. Southwood points out that the firm licensed to sell the machine in Sub-Saharan Africa, African Digital Bridges, dropped the product because selling it “was not worth sustaining operations.” (Jamais Cascio has a typically thorough and excellent overview of other low-cost computing projects at WorldChanging – it’s skeptical about the HDL, but a subsequent piece is somewhat more positive.)

Southwood is skeptical that any of these devices is going to be sufficiently appealing to African consumers that they’ll be willing to actually purchase them: “Try going to the informal settlements of most major African cities and explain to potential customers why they might want a cheap PDA or indeed a cheap laptop. Negroponte is giving them away because he knows what an unbearably difficult task it would be to have to actually sell the things to people who might actually want them and have the money to buy them.”

Instead, Southwood’s more excited by Africa’s rapid embrace of the cellphone – no government has subsidized cellphone purchases by their citizens, yet many millions of Africans have been able to purchase new or used GSM phones and operate them on a basis that’s been massively profitable for some cellular carrier – in other words, Africans are able to pay for communications technology when devices are made available that meet their needs and when governments are willing to open markets sufficiently to make these networks possible.

While I share Russell’s excitement about mobile telephony in Africa and some skepticism about the market appeal of low-cost computers, I think it’s clear that Negroponte is proposing these computers not as a consumer product, but as something analagous to school textbooks, purchased by governments, given or sold to students on a subsidized basis. While tens of millions of Africans have been willing to invest their money in mobile phones, even more make the substantial investment in school fees for their children.

But the real challenge for Negroponte and crew is making these computers cheap enough that they’re part of a student’s school fees without making those fees unaffordably high. Even at a hundred dollar price point, with hundreds of millions of school-aged children in Africa, this is a project that’s going to require tens of billions of dollars. Total US aid to Africa last year was in the neighborhood of $4 billion dollars. Is counting on governments to fund these purchases – either through taxes or donated funds – really the right way to bridge the educational digital divide? There I share Russell’s doubts.

The next “due date” for the sub-$100 laptop is a prototype due in November – like everyone else in the ICT4D world, I’ll be waiting to see it with hopeful and skeptical eyes.