« October 2006 | Main | December 2006 »

November 18, 2006

Library Grants, Laptop Gambles

James Surowiecki, author of The Wisdom of Crowds asks the simple question of the OLPC: "Will this laptop save the world?" in MIT's Technology Review.

olpc-saves-world.jpg
I've gotta go with.... "No."

It compares the OLPC project to Carnegie's construction of public libraries across the States:

The first library he built in the United States, eight years later, opened in ­Braddock, PA, where Carnegie Steel had one of its biggest mills. A year later came the Carnegie Free Library of Allegheny, PA. The Allegheny library was important because it was the first funded according to the model that Carnegie would follow thereafter: instead of simply paying for and endowing the library, he offered the town a large initial grant on the condition that it agree to pay for the library's operations thereafter. (In what came to be known as the "Carnegie formula," towns generally committed to an annual budget--for maintenance, new books, and so on--that equaled 10 percent of Carnegie's original gift.)

Take note of that parenthetical - Carnegie donated funds to build libraries. The towns had to demonstrate need and provide land, as well as pay 10% of Carnegie;s original grant to support the ongoing functioning of the library:

By the time he died in 1919 ... Carnegie had given away $350 million of his fortune; he spent more than $60 million of it to build more than 2,800 libraries, including almost 2,000 in the United States and almost 700 in Great Britain. His donations had so effectively revolu­tionized public opinion that by the middle of the 20th century, it was the rare American town that dared go without a public library.

Part III of the story addresses critics of the project, first dismissing the cell phone objection and then tackling the question of whether an expensive laptop project is the best use of limited budgets:

It may be difficult for poorer governments to justify spending a good chunk of their education budgets on laptops. But the ­reality of both philanthropic and government spending is that money often goes to projects that do not help as many people, or people who are as needy, as other projects might. These projects may not be perfect, but they can still do tremendous good. In the post-Reconstruction United States, after all, there were lots of worthwhile things Carnegie could have done with his money; in fact, in many of the towns where he built libraries, citizens grumbled that their tax dollars should be going to something that really mattered. Yet in the long run, one would be hard pressed to say that either Carnegie or the taxpayers wasted that money, because the social benefits of disseminating knowledge are so immense.

There are three major problems I have with this comparison. First, we're talking laptops, not libraries. At the same time that Carnegie was putting his money towards building laptops, it came at the same time of the peak of library expansion in the U.S., and books were not exactly a new phenomenon. The OLPC is a dramatic change for the communities which will receive it. Second is these communities. In the Carnegie Formula, the communities had to fulfill certain requirements; in the OLPC formula, the national government is the only actor, glossing over communities which may not want it or may not be able to take advantage of it for various reasons.

Third and most importantly, Carnegie was a philanthropist. He was giving his fortune via grants away to build libraries. The OLPC is a non-profit foundation, the recipient of grants to build laptops, but the governments which are signing up for them will pay, and pay dearly. These are not grants which set up new technologies, new sources of information which become readily self-apparent, this is an untested new technology that requires an upfront, massive, debt-financed payment. For developing nations considering signing on, they must ask not only will it succeed at all, but will it flourish and cause enough economic growth by its education of the youth to enable the government to pay back the billions of dollars in loans that it had to take out, and will continue to take out, to provide its students with OLPC laptops. With the OLPC estimate of 30B/year globally ($7B/year more than the Bank's entire 2005 lending), this is gambling a nation's economic future on untested technology.

digg this!digg this del.icio.usdel.icio.us redditreddit slashdot bookmarkSlashdot StumbleUpon Stumble It!

November 14, 2006

OLPC Implementation Costs

I'm now writing some OLPC posts for OLPCNews.com; my first post is uncovering some of the hidden costs in OLPC implementation - it turns out that the $100 laptop costs almost 10 times that over the first five years of its implementation once you look at training costs, maintenance, and Internet costs!

digg this!digg this del.icio.usdel.icio.us redditreddit slashdot bookmarkSlashdot StumbleUpon Stumble It!

November 09, 2006

A Million Laptop Hangover

[olpc]Over the next few years, ministers of education worldwide will be waking up from their celebratory launch parties announcing the agreements for the OLPC laptop -- with a hangover of a million laptops sitting on a wharf. What can you do with a million laptops? Hype aside, is there any plan for distribution of the physical boxes, or, more importantly, diffusion of their usage?

I can't help but recall two twin projects that I helped with as an IT volunteer with the Peace Corps in the Caribbean. Two separate programs, one well-funded, one less so, were both trying to spread a low-cost computing alternative (in this case, a keyboard with a small LCD monitor with USB and IR connections - not only do these do less than the OLPC computer is designed to do, they cost two to three times as much). The under-funded project had one Ministry official designated to manage all the distribution, do teacher training and followup visits. Despite valiant efforts, this project failed to get much traction with the overloaded teachers. The better-funded project was able to first do more preparatory work to find interested teachers and administrators, as well as fund teacher training workshops and do many more followup site visits, and was able to get some limited traction in schools with a concentration of interested parties.

I'm not arguing that funding matters, I'm arguing that networks matter -- networks of teachers and administrators, community leaders, and implementation personnel helping with technical issues. This is greatly lubricated by increased funding when transport and telecom communication is cost- or time- prohibitive. This is all an attempt to remind us that you can't rely upon the "obvious" advantages of a new technology to lead to instant adoption by a community. The history of development projects is full of failed attempts, or worse - successful ones that failed later on. This has happened time and time again with "green revolution" projects, where a development project finally convinces a community to switch to a "new, better" crop, only to see it fail due to local conditions or strong preferences overlooked by the development agency when they ignored millennia of farming practices for a few
decades of experimentation.

I am only being critical of the OLPC project here because I honestly want it to succeed. The design of the system addresses such a wide variety of the exact challenges I faced doing IT work. Power supplies are bad or non-existent. Rural telephony/connectivity is unavailable, dust, heat, outdoor classrooms (which mean bright
sunlight, not very useful with most laptop screens), or lock-in to proprietary and expensive to maintain systems and software. The OLPC project has designed around so many of these problems, no doubt from close observance of common, local physical obstacles to ICT usage that it's astounding that they seem to have completely ignored the non-physical obstacles.

digg this!digg this del.icio.usdel.icio.us redditreddit slashdot bookmarkSlashdot StumbleUpon Stumble It!