I've found Givewell to be incredibly analytical and thoughtful in their analysis of how to get the most bang for the nonprofit buck, and I always consult them before giving. I'd encourage everyone to visit GiveWell next time they have some reason they're going to give money (or some area they want to give to) but they're not sure to whom to give.
(Fwiw: before I entered tech I was a consultant helping big foundations (like Gates, Packard, etc.) assess their performance (as part of http://www.effectivephilanthropy.org). So for a tech person I consider myself reasonably well-versed in nonprofit evaluation--and how difficult it is.)
Seems awful trivial five years later. Givewell does excellent nonprofit analysis, and they don't seem malicious or unethical in any way that I've seen.
Yup. It's unfortunate that this still dogs them in internet forums, but it's really only on internet forums where someone always insists on bringing it up.
One thing I love about GiveWell is that they publicize and learn from their mistakes, this one included. http://www.givewell.org/about/shortcomings They're really, really good people. Like everyone, they did something dumb. Unlike most, they learned from it and spend their lives trying to improve the world.
For an example of their work, an analysis of the value of giving cash directly to those in need (instead of, say, a cow): http://blog.givewell.org/2012/12/26/the-case-for-cash-2/
Their top recommended charities: http://www.givewell.org/charities/top-charities
(Fwiw: before I entered tech I was a consultant helping big foundations (like Gates, Packard, etc.) assess their performance (as part of http://www.effectivephilanthropy.org). So for a tech person I consider myself reasonably well-versed in nonprofit evaluation--and how difficult it is.)