It's also interesting how only two ideas (crisis info and virtual corners) focus on the needs of the people, and they are at the bottom of the funding pile.

As for the winners:

DocumentCloud seems to be for source material, which sounds really cool, but I have to wonder if you can train reporters to share what is essentially their bread and butter.

Mediabugs is predicated on the idea that the media cannot be trusted or held accountable. If that's true, then what's the point? It's like shining a rotten apple.

Councilpedia is already being done in many forms, but every newspaper should be doing this.

Data visualization? Are the judges living in a cave? Plenty of charting libraries exist already in every programming language. Not to mention Web services that do this, too!

Mobile media toolkit? Again, lots of open source tools already exist. And I don't see how $100K gets them very far with software tools as complicated as these are likely to be.

The Daily Phoenix? I don't even see how this one is relevant to anything. They should have just found a local advertiser to pay for it.

Crowdsourcing Crisis Information: This is the only truly important idea of the bunch, and it's sad how little money she got for it.

Virtual Street Corners is an art project. Cool, but not sure how it relates (it's not duplicable on any kind of scale).

CMS upload utility: this one is frankly hilarious. $10K for a "utility" that you could have any programmer write for less than $500. If small newspapers really have a CMS problem, they need a better CMS, not some hacky utility.