Skip to content

OLPC / XO

2008 April 28
by Hélène Martin

We’ve had our OLPC through the G1G1 program for long enough that I find myself entitled to an opinion. It’s not positive at all. We got ours mostly as a fun coffee table ornament, but also out of a sincere hope that it could have value for the developing world and that we could contribute developer hours to the project. Having wasted yet another long series of hours on OS woes, I’m about ready to call the whole damn thing a completely futile endeavor. I love the hardware and hate the software. Maybe Windows will make it easier to use. Harr.

Still, I think it has value, and I’d like to get it to a point where I can reliably use it as a travel computer. It’s small, light, cheap, not enticing to steal and has reasonable battery life, though nowhere near what promised. The default browser is mostly useless, but Opera is a very nice alternative. Skype on there works reasonably well. And I love, love, love the screen.

Some thoughts on my experience so far:

Intuitive?
I’m spatially challenge, but I’m vaguely intelligent. It took me about seven minutes to get the XO opened, and that was with Yaw’s suggestions and both of us puzzling over it. No one I know has opened the darn thing in under two minutes. Some people simply walk away and others come pretty close to snapping the thing in half. I know kids will be in classrooms when they receive them and get instruction and all, but still. I think it’s great that it holds shut really well and has an amazing form factor but… arrows? Hints? Anything? It’s just a frustrating way to get started.

The thing takes A Long Time to boot. Very long. Minutes. The visual feedback doesn’t start right away, so I hard reset the thing in its boot sequence three times thinking it had hung before just letting it do its thing and realizing it just took a while. Now that it hibernates, that might not be such an issue. But still, another source of frustration.

I really like the different zoom levels Sugar has — network level, group level, machine level and app level. But I don’t think it’s very intuitive. Most applications toss you into an environment that’s pretty difficult to discover. My younger brother is 18 and brilliant but I had to spend some significant amount of time explaining Etoys to him (which is not some magical thing. It’s just as clunky as Squeak). Again, this will probably come up in a classroom context where explanation can be given, but I’m disappointed. I thought Sugar was supposed to revolutionize the GUI design world. Far from. Many of the metaphors are the same, just somewhat stylized.

Keyboard mappings are really inconsistent across applications, in my experience. Sometimes the X works to quit, sometimes it doesn’t. Sometimes copying works, sometimes it doesn’t. Again, irritating. It makes me want to play with balls and hoops because those always do EXACTLY WHAT I EXPECT. Which brings me to my next complaint…

Unpredictable behavior
I stopped using Windows long ago because I felt like I was mostly at the mercy of completely random patterns of behavior. I’d try the same thing five times and get five wildly different results and that was driving me crazy. I’ve never seen anything as bad as the OLPC. We got a machine with build 656. Yaw upgraded to candidate-691. It somehow reverted back to 656. I updated to candidate-691. This morning it was back to 656. I upgraded to candidate-703. Rebooted. Failed at installing activities. Rebooted. Failed again. Rebooted. Back at 656.

At first we could su and get a root shell. At some point su stopped working and we had to click a button. This morning, su worked fine. And now, back at 656, for the first time, I get /bin/su: permission denied and no root shell from the terminal activity. But I can log in as root from the overall terminal.

So I’m trying the 703 upgrade again, but I feel crazy. Maybe this isn’t such a big problem because children will not be expected to perform upgrades on their own machines. Fine. I can buy that. But if there’s some magical sequence of things I keep doing that revert to a previous version, I don’t know what those poor kids will think.

Education?
There’s a fair amount of opportunity for fun on here, and that’s great. A lot of it is somewhat educational or at least provides access to more information and varied experiences. Fine. I’ll buy all that. But when teachers get asked what kids in pilot programs do with their XOs, answers range from “oh, they measure the distance between each other” to “they really like the camera.” Really? That’s the best you can say?

There’s value to learning through osmosis. It’s how I learned almost everything I know about computers — just poking around. But I worry that a lot of the knowledge gained this way won’t really be all that transferable. The applications are… well, toys.

Impact?
Again, the hardware is beautiful, and I’m extremely glad we have it. But I really worry about the impact of this project. The politics of it all have been heartbreaking to watch and the software is far from usable, in my opinion. I wonder whether the overall effect will be positive — what if some governments find themselves short millions of dollars with plenty of paperweights and as just as many hungry mouths to feed? I worry that the reaction will be to move away from funding educational technology programs, which would be a very unfortunate result.

No comments yet

Leave a Reply

Note: You can use basic XHTML in your comments. Your email address will never be published.

Subscribe to this comment feed via RSS