SJ Klein brought one of the first factory-made One Laptop machines to the Berkman Center for show and tell yesterday. SJ is the director of content for One Laptop, tasked with figuring out what sorts of texts and information get preinstalled on these machines and on the base stations/servers/ethernet-connected nodes that will be distributed with the One Laptop machines.
(One of the problems OLPC needs to address is the naming issue. Some people are calling it the “OLPC Laptop”, others the “X0″, others the “hundred dollar laptop”. SJ referred to it at least once as the “One Laptop”, which has an ominous Tolkien ring to it, to my ears.)
OLPC in Cambridge just took shipment of “a thousand pounds” of laptops, roughly three hundred machines. Unlike the last version of the green machine – which were hand assembled – these machines have been built in a Taiwanese factory. They’re part of a set of early version which are heading, in lots of a thousand each, to the five countries that have signed on for the first phase of the laptop project (Argentina, Brazil, Nigeria, Thailand and Libya – those were the countries SJ listed – commenters have pointed out that Thailand has pulled out of the initial phase of the project.) I consider the factory-built laptop a pretty major milestone, a useful retort to people who’ve been referring to the machine as vaporware.
The machine is – speaking in purely technical terms – adorable. It strikes a nice balance between having the approachability of a children’s toy (it looks remarkably like a green speak and spell, especially when the screen is laid flat on top of the machine) and feeling rock-solid. It’s surprisingly hefty for the small size – about three pounds, SJ says, one pound of which is the thick plastic shell.
Like Weinberger, I found the machine bafflingly hard to open. I don’t know whether it’s a design feature that you need to flip up one of the bunny ears to open the case, or whether I’m simply too accustomed (read: “old”) to opening laptops from a central latch, but I noted that every single person around the table seemed to be thwarted by the mechanism.
Once open, the main impressions I had were just how impressive the screen was and just how tiny the keyboard was. I noted in an earlier post on the laptop that the machine is not a “cheap” machine, but a kid’s machine. You’ll get that impression immediately once you try typing on the keyboard. It’s really, really small if you’re used to a full-sized laptop keyboard. I can imagine growing used to the keyspace distance with time, though, and the feel of the keys was surprisingly satisfying under the green membrane (which will likely do a good job of protecting the machine from liquid spells, or from the sunflower seed shells that always end up in my PowerBook keyboard…)
It’s been difficult to get information on the exact specifications of the screen OLPC is using in this version – part of this is because the OLPC project has been doing some extremely innovative work on LCD design and is patenting aspects of the screen design. SJ showed off the two modes of the screen – a bright, clear color mode created with (I believe) multiple, colored backlights, and a low-power reflective mode. In the reflective mode, the CPU shuts down and stores the image in a video buffer – this allows the machine to run for many, many hours on a full battery, as it draws only .3 watts in this mode. Going full-tilt, with the video system in full color, the draw can get as high as 5 watts… which is still far lower than power draw on a conventional laptop.
Because we all wanted a chance to play, I didn’t get much time to experiment with the Sugar operating environment. It wasn’t an immediately intuitive environment to me – it took a while to realize that there’s a need to “focus” on the center of the window to have an application pay attention to you – otherwise it thinks you’re interacting with the Sugar “border” around the edge. I played a bit with the word processor, managed to surf a page or two in the web browser, launch the Squeak environment (complete with flash video) and get utterly baffled by TamTam, the sound synthesis program. (Wayan’s got a much more thorough overview of Sugar on his excellent OLPC blog.)
The “eureka” moment for me came when I hit one of the icons and suddenly found myself staring into my own face. The machine has a pinhole video camera for videoconferencing – SJ reports that video has been so popular with early users that the team is looking for ways to let students make and share short videos. It’s hard for me to explain just how impressed I was by the ability of the machine to do videocapture – this is a machine with no rotating storage, and only 512MB of non-volatile memory to store the OS, the applications and any data generated. Where the %@#$^$%! do you even put the video you capture in that environment? It’s really remarkable.
My main complaint about the current version of the machine is the sluggishness of the Sugar interface. I’m guessing this is less a function of processor power and more a result of the environment being actively under development – I strongly suspect there will be major speed tweaks later in the process. The fact that Christopher Blizzard and his team loaded up DOOM on the box and were able to play it, using the Playstation-style keys and the machine in flat-panel game mode, suggests to me that the machine has some horsepower and that the problem is in the alpha-stage software. I hope that early testers of the machine aren’t turned off my this… or that my experience of sluggishness had more to do with inadvertently starting half a dozen applications, slowing the machine down.
Needless to say, I want one. I think I’ll settle, in the meantime, for putting the Sugar development environment on a Ubuntu box so I can play with it and get a bit more out of my next experience with the prototype the next time SJ decides to come by for a visit…