* As currently structured, it's obviously a problem in game theory, behavioral economics, and group collaboration, not technology. Contestants will register as individuals, not organizations -- of course, people will collude and collaborate, but they'll have to determine not only mechanisms for intaking, validating, filtering, and crediting distributed reports (your basic crowdsourcing problems), but they'll also have to figure out the basis for distributing the prize. (And $40,000 is in the realm of "real money.")
Presumably, a large enough group to win will be in it for the bragging rights (I'm sure everyone is thinking "Anonymous"), and simply donate the money to charity, to avoid all those messy distribution issues.
However, since the registrant has to be an individual, the winner of record will still get stuck with the tax on the $40K prize, even if they donate it all to charity. Something to consider.
* The rules are only two pages right now. I'd be really surprised if the rules (which "may be modified at any time without notice") don't iterate and grow dramatically in between now and December.
Of particular concern is the potential griefing issue -- anyone with any modicum of deviousness has already said, "Hey, we should just fly a bunch of fake weather balloons to fuck up everyone else."
(The rules state that official balloons will be accompanied by DARPA representatives -- we're going to need some authentication method beyond "Some asshole wearing a navy blue polo shirt holding a clipboard." It'll have to be communicated far enough ahead of time to propagate, but not so far in advance to be easily spoofed.)
Griefing goes to the idea of motivation: You might not be able to mobilize and organize an credible nationwide team to win, but I can easily see groups of griefers -- call them "cells" -- popping up on game day:
Looking at costs, an 8-foot diameter balloon holds about 270 cubic feet of helium; with cost of balloon and tank rental, that's a few hundred bucks, which is easily spread across a few friends.
* Hopefully, no one will try shooting down balloons. That would be a bit much (as well as a violation of the contest rules). The rules cover changed locations, but I don't yet see anything about accidental or manmade balloon failure post-launch.
As to other potential hurdles (balloons in really out-of-the-way -- but still road accessible -- locations, I don't find that as interesting as the challenge of dealing with griefers, especially highly-motivated ones bent on undermining the game.
Of course, since no one ever accused DARPA of being dumb, it really depends on how they implement the Web site reporting mechanism (will it provide feedback when you submit a location, etc.)
Watch the skies.
Tags:
Posts
12 comments:
Baloon validity is relatively easy to do with public key cryptography. Just post a public key for the contest and at each baloon sign the lat/lon coordinates with the key. This increases the barrier to entry for participants because they'll have to be savvy enough to know how to verify the signature. But putting that validation system in place might be enough to dissuade most potential griefers.
Griefing? No.
Spoofing to slow down the competition? Absolutely. It was the first thought that popped into my mind when I heard about the contest.
No public key will prevent a good counterfeiter from reporting dozens of fake balloon sites, to include geo-tagged photos of the fake balloons.
This is definitely going to be an interesting contest.
Anon: We'll see -- the instructions to date indicate that it's the lat/longs that count, not any validator.
Dirty Name: Definitely the possiblity of EVE-style infiltration shenanigans. Each team will need to come up with its own regimen for verification, and those less cautious of who they allow to participate might get bitten.
Red40k.com will be setup to take balloon location submissions and paypal $3000 to the first email address associated with a correct balloon location, if we win the $40k.
Winning this contest depends on going as meta as possible. The most open, public, many-tentacled group is going to come out on top.
Looking forward to seeing how the public group fares in this...
There are a variety of different ways to try to win this competition but I'm sure most of them leverage, in one way or another, having as many people as possible know about the competition. Everyone should help spread the word!
We're trying to do that on Facebook and on our website:
http://www.redballoonrace.com
Check us out and shoot me an email! Tell me what you think!
-Tom
Everyone is trying to get the word out. My team, DeciNena, will win because we have the coolest name. We're also offering to share a smaller prize amount with team members who don't find a balloon, as long as they demonstrate they're out working for us on B-Day.
http://decinena.com/
Hey man, I met you at a Fark party a while back, cool to come across you while looking for info on the DARPA challenge.
Rob -- heya, yeah, I just saw the Fark Network Challenge thread went green earlier today. I don't envy the person/people who will try to coordinate it.
There is an iPhone app and everything for this:
www.armyofeyes.com
We've got a strong team and plan on giving all the prize money to charity (Red Cross). If you would like to help, report your sitings to
http://www.ispyaredballoon.com/
or facebook
http://www.facebook.com/group.php?gid=201028633372
Instead of trusting that a team will actually pull through for you, why not win the prize for yourself: http://www.openredballoon.com - the site is completely transparent and shows all balloon locations. The catch? You have to pick the winning combination of balloons!
Post a Comment