Are you seeing red? Balloons, that is...

[turil.livejournal.com]

Here's an interesting contest by DARPA, celebrating the internet. Network with folks you know around the US, and see how many of the red DARPA weather balloons your network can find. The MIT team is sharing the prize, while doing an experiment, too. It's just started today, and probably won't take more than a day or two to finish, so check it out asap, if you're curious.

It might be fun...

MIT Red Balloon Challenge Team

Comments

Well, DARPA has a history of fixing "competitions".

[nvidia99999.livejournal.com]

Remember when, a few years ago, they had the initial competition for novel autonomous vehicles? Originally, they had designed a difficult path that spanned several miles of a rocky desert. Several tens of teams participated (including MIT). The first time, all vehicles crashed within a hundred feet, or something equally pathetic. A PR disaster. So, what did they do the next year? They simply made the path incredibly easy, without revealing the cheats. This time, some teams finished (and not because they had really improved their prototypes). The event was on all medial outlets: DARPA was funding supercool research again...

Re: Well, DARPA has a history of fixing "competitions".

[leko.livejournal.com]

That's not really accurate. The first time the CMU vehicle went several miles before getting stuck on a rock or something and then spinning it's tires until they caught on fire. They spent the next year improving that vehicle, and developing another. I don't know the specifics of the course changes, but many friends of mine worked on CMU vehicles, and I can tell you they definitely improved things (like activating a bunch of sensor systems that they didn't use in 2004 because they didn't have time to test them before the race) for the 2005 race.

Re: Well, DARPA has a history of fixing "competitions".

[nvidia99999.livejournal.com]

What can I tell you? Just because they added sensors does not mean performance on the original (realistic) course would have improved. If DARPA officers admit they cheated for PR purposes, I believe them.

Re: Well, DARPA has a history of fixing "competitions".

[leko.livejournal.com]

I'm not arguing that they didn't make the course easier. My problem with your comment is your incorrect assumption that the teams didn't improve things after the first race.

Did you even watch the races? The first year the vast majority of the vehicles crashed well before the course became 'difficult.'