Thursday, February 07, 2008

XO US Pilots... Rollouts?

OLPC News tips us off to the OLPC in NYC blog, documenting this:

Teaching Matters will run a collaborative pilot with the Department of Education to test One Laptop Per Child mobile computing devices in connection with our Writing Matters content in a middle school ELA classroom. The purpose of the pilot is two-fold. First, we want to determine if the OLPC device can significantly lower the cost of technology access for schools by lowering the total cost of ownership (hardware and ongoing maintenance.) Second, we will test this environment in conjunction with a curriculum designed to improve teacher practice in the teaching of writing. The curriculum has been designed to take best advantage of one to one computing environements.

They're doing a great job of updating their performance on a daily basis, so one gets a very clear sense of what they're doing and how it feels. So, yay! I'm glad their doing this. I would note, however, that it is highly unlikely that they'll end up with any kind of meaningful data about TCO or the efficacy of the XO's by April. I mean, I expect them to find that the hardware works well, and the software is more dubious and the source of most of the support issues. But the software can be reasonably expected to change a lot over the next year or so. It is too much of a moving target to really evaluate yet.

Time... that's the key element now... how long can they keep the production lines running until the software catches up with the hardware? I have no clue. When I think about real OLPC deployments in the US, here's what I hope can work out. Let's hope they don't try to do a major (i.e., citywide) rollout in the middle of the year (just because, that's always insane). Then let's also hope that nobody tries to do a major rollout by this September. I hope nobody thinks there is time for that. So, more pilots '08-'09, real deployment fall '09 makes sense to me. If the software isn't in much better shape by then, it isn't going to be ever. I just hope they can keep the ball rolling that long.


Wayan said...

While I agree with your premise - OLPC needs to improve the software and there needs to be conclusive pilots - I do not think OLPC has the time to wait until Fall 08, much less Fall 09.

With all of his "10 million laptops" bluster of 2006, Negroponte painted himself into a corner with his hardware suppliers. He's already been pilloried for his 07 sales, even though for any other technology project, it would've been respectable.

If he doesn't deliver serious numbers in 08, I doubt there will be any XO's produced in 09.

Tom Hoffman said...

Well, it is just hard to know what is going on in the minds of suppliers, what their spreadsheets really look like, what it might mean if some of the OLPC technologies are spun off to commercial products (thus keeping production lines running that are also used for XO components).

I mean, I can see someone meeting with Negroponte, listening him say "We're going to sell 10,000,000 next year!" Going home, saying to each other, "What do you think he'll actually sell? 250,000? 2,000,000 in three years? Can we make any money off this?"

If they decided to go ahead, it isn't like they are going to put out a press release that says "X Corp. thinks Negroponte's sales estimates are off by an order of magnitude, but we're going to participate anyhow."

I would imagine the electronics suppliers in question are a pretty unromantic bunch. I can't imagine that they didn't make contingencies for the kind of slower launch that we've seen.

OTOH, I do kind of feel like NN had to throw out some big numbers as a kind of initial bargaining position to stake out the scope of the ambition of the project. Better to claim 10,000,000 and ship 750,000 than to predict 750,000 and only ship 50,000.

Regardless, we'll know how this all unwinds soon enough.