I agree that a good portion of the "hiring crunch" right now is an issue of mis-allocation rather than raw under-supply. Yahoo! has tons of great engineers wor

I agree that a good portion of the "hiring crunch" right now is an issue of mis-allocation rather than raw under-supply. Yahoo! has tons of great engineers working on really hard technical problems. What qualified engineer there wouldn't be better off at even, say, Facebook? The engineers who work on Sherpa, for example, are highly qualified. I have a hard time seeing Coderwall as the solution, though. (Not that you're arguing that it is.) Why? The core design decisions in the product are going to attract the kind of people who like to show off. Badges? Achievements? Public profiles styled a la the Facebook timeline? Those are the engineers who, by and large, are already showing off. They have public personas, whether on GitHub, Twitter, major projects they contribute to, or just being "personalities" in their community. The product that gives exposure to engineers hiding inside companies like Yahoo! needs to be something less, well, game-like. I'm not sure what that looks like, but that's my first impression of Coderwall. It feels like a cargo cult. All these little gamey doodads don't add up to a coherent thesis about how to get engineers more exposure. It feels more something that came out of this kind of conversation: Person A: Hey, we need to increase engagement. How can we do that? Person B: Achievements! That'll do it! This is why GitHub is so marvelous. It didn't start out with the idea of getting engineers more exposure. The exposure is a side effect of making it easier to contribute in public than in private. That'd be my starting point for doing something really radical: what can we make public that used to be private, in a way that excites developers? Making GitHub activity "even more" public, or aggregating it in a way that's easier for non-technical people to consume, is probably not the first thing I'd try. It is the easiest, though. (For me, that's a reason not to do it, rather than a reason to do it.)