OLPC Feels The Pinch

Chalk up another open source victim of hard days: the One Laptop Per Child project. It isn't dead, and I'm glad it's not, but I'd wondered how well these guys could weather the tough times.

A posting on the official OLPC Wiki outlines the scope of the damage: about half of the OLPC dev team is getting laid off (approximately 30 people), with the rest taking salary reductions. Most of the in-house effort is now being turned away from Sugar, the OLPC's on-board software, and towards the snazzy-looking next generation of OLPC machinery.

So what happened? An Ars Technica piece puts it this way: "The OLPC project's extreme dependence on economy of scale has proven to be a fatal error. The organization was not able to secure the large bulk orders that it had originally anticipated and fell short of meeting its target $100 per unit price."

There's also a bit of dithering about the internal conflicts that plagued the OLPC Foundation, but the fundamental problem does seem to be that you can only make something so cheap by putting it together in bulk. The economic status of the nations in question didn't even really seem to be the issue, but it sure didn't help. I'm still not of the opinion that a developing nation needs to shirk digital infrastructure entirely; it's just going to take a different form (e.g., investing in cell towers rather than fixing decrepit POTS networks).

I also wondered whether the Sugar interface is going to ultimately become second banana to more widely used, -tested and developer-supported work like GNOME. The Sugar Labs folks recently joined up with the GNOME foundation, after spinning themselves off separately from the OLPC project as a separate entity. Open source or not, the project needs developers working actively on it to keep it relevant.

Another proclamation from the OLPC folks had me raising my eyebrows even more: "OLPC will be dedicated to bringing the cost of the laptop down to Zero for the Least Developed Countries -- the $0 Laptop." How they plan to do this is not clear. If I had to hazard a guess, they would probably have to use something along the lines of the $100 Acer laptop, sold through Radio Shack and subsidized by the sales of Internet connectivity to the device. But how to manage something like that in a developing country, where there isn't as much of an infrastructure for subsidy in the first place?

The OLPC project's a good idea, but I'd rather not see it go headfirst into a wall. Especially if so much of what they're aiming for is simply unattainable without charging a realistic price for it.

Follow me and the rest of InformationWeek on Twitter.