Hm. And I thought there'd be more interest here in the Android and ChromeOS angles. I guess not.

--scott (Director, New Technology, OLPC)

Yeah, OLPC (and the XO) have been making steady progress for the past few years. There are "grown up" versions of the XO with more standard keyboards, and the XO-1.5 have more memory and processor than before, etc.

That's the XO-2 design, which was discontinued. The XO-3 design is a single screen -- look for more details on this soonish. (The renderings on the olpc website are not up to date.)

Yes, the design of the laptop is intended to make it clear that it is a children's machine.

The hand-crank was always a creation of the confused media coverage. You *can* power the XO-1/1.5/1.75 with a hand crank -- you just wouldn't want to. Your arms are not the strongest part of your body, and the cranking motion is inefficient. OLPC invested in multiple different power technologies for different parts of the world -- step-powered generators, cow-powered generators, small and large solar panels, car battery multi-chargers, etc. Large-scale solar power is the only thing which deployment countries have been eager to adopt in large numbers. We respond to our deployments.

Incidentally, the lower power consumption of the XO-1.75 unsurprisingly makes it work better on solar power as well.

The original G1G1 program also had large support and marketing costs.

Take a look at javascript, especially ES5.

And this language, in particular, combines all the worst features of C, Perl, and Python. I can't find a single example in the tutorial that doesn't make my eyes bleed.

Interesting link -- but CS does provide mechanisms for creating "trust worthy" bundles securities, in the form of one-way functions. If the seller says, "I distributed the asset types among these securities using a random number generator built on a cryptographically secure one-way function with the following seed", it is possible to have a high degree of confidence that the distribution really is random. The seller can rejigger the seed but the one-way function (statistically) prevents more than a certain amount of tampering. (Of course, you can still try to tamper with the ordering or identity of the input securities -- discuss!)

RC4? Really?! Dude, that's totally broken -- see http://en.wikipedia.org/wiki/RC4#Security and especially http://en.wikipedia.org/wiki/Fluhrer,_Mantin,_and_Shamir_attack . Ron Rivest makes lots of good stuff, but all of "Ron's Codes" have been broken. (Except maybe RC6, but the AES committee determined that Rijndael was better than it.) Classic example of amateur cryptography actually resulting in a system that is *weaker* than the alternative. Le sigh.

The litl uses low-travel scissor-action keys, like the Apple keyboards, not rubber keys like the old 'chiclet' keyboards and the OLPC XO-1. The keyboard is pretty pleasant to type on. See http://en.wikipedia.org/wiki/Chiclet_keyboard#Recent_developments

The support videos at http://litl.com/support/ give a good idea of how the UI works -- and they also feature Annie, litl's friendly customer care rep!

It's not running GNOME. It's running a custom shell based on X11/Clutter, with a few GTK widgets to avoid having to reimplement a textbox for the 1337th time.

The WSJ article was not linked, but it's at http://blogs.wsj.com/digits/2009/11/04/litl-introduces-its-web-based-netbook/

Actually, I think you've got cause and effect backwards. Other GNOME guys have been hearing about big-shell's use of JavaScript informally from Havoc, and decided to adopt it for gnome-shell since it's worked so well for litl. If you look at the commit history of gjs you'll see it's about evenly divided between litl people and gnome-shell people now.