SilasBarta comments on Help Fund Lukeprog at SIAI - Less Wrong
You are viewing a comment permalink. View the original post to see all comments and the full post content.
You are viewing a comment permalink. View the original post to see all comments and the full post content.
Comments (276)
Good grief, people. There are conspiracies that need ferreting out, but they do not revolve around generating fake data about the effectiveness of an alpha version of a rationality training camp that was offered for free to a grateful public.
I went to the minicamp, I had a great time, I learned a lot, and I saw shedloads of anecdotal evidence that the teachers are striving to become as effective as possible. I'm sure they will publish their data if and when they have something to say.
Meanwhile, consider re-directing your laudable passion for transparency toward a publicly traded company or a medium-sized city or a research university. Fighting conspiracies is an inherently high-risk activity, both because you might be wrong about the conspiracies' existence, and because even if the conspiracy exists, you might be defeated by its shadowy and awful powers. Try to make sure the risks you run are justified by an even bigger payoff at the end of the tunnel.
For me, this isn't about making SIAI transparent; it does quite enough in that regard. It's about stopping an information cascade genie that's already out of the bottle.
Let me put it this way: right now the ratio of "relying on the assumption of mini-camp's success for decision making" to "available evidence for its success" is about 20-to-1. As I warned before, it's quickly becoming something "everyone knows" despite the lack of evidence (and major suspicions of many people that it wouldn't succeed going in). And that believe will keep feeding on itself unless someone traces it back to its original evidence.
It doesn't reassure me that I'm told I have to keep waiting before anything's conclusive, yet they can declare it a success now.
I just want the reliable evidence they claim to have, rather than just dime-a-dozen self-help testimonials. They collected hard data, and I gave them a list of things they could provide that are easy to gather and don't compromise privacy, and are much more likely to be present if the success were real than if it were not. Even after AnnaSalamon's circling of the wagons I don't see that.
I think this is largely a case of people reading different things into 'success'.