I don’t think it is a good idea to invoke any sort of controversy without some specific novel point to make. I would not object were it just a thought experiment in an open thread, but good cause is necessary for a top-level post. As I said to Jack, I would hold top-level posts to a higher standard than “don’t see the harm”. By the novel-insight standard, “The Cameron Todd Willingham test” fails on two grounds:
The general idea is an obvious generalization of something already explored in some detail previously: examining the evidence associated with a criminal case and drawing an independent conclusion about the guilt or innocence of the defendant.
The post neither promises nor delivers substantial analysis on any subject.
It is my contention that what Kevin wrote does not show enough effort to justify a top-level post.
It is derivative of the Knox posts. But it asks us to do new things with our skills. In the Knox test the right answer was just that AK and RS have a very low P(Guilty) and RG has a high P(Guilty). This post asks us to compare the relative innocence of two people for two different crimes, with different circumstances. These exercises are important because right now all we can do is basically say “low” or “high” of “fifty-fifty”. I mean, yeah, people gave numbers but I’m skeptical they mean that much. We need to get a better sense for how probabilities vary at the extremes and this will help. [Aside: Frankly I think we should be doing this kind of thing way more often. Someone tried to do this for the probability of a nuclear attack in this months open thread and no one would even try to put a number on their current beliefs (so there couldn’t even be an extended discussion of the issue).]
Kevin also asks the interesting question about a Bayesian justice system which I suspect will lead to a lot of discussion.
And I’m not saying it deserves to be promoted. It doesn’t. But that doesn’t mean it deserves negative karma. A +5 or +6 would be fine.
Is the post a reasonable open thread comment? Yes. But throwing out the bare questions, questions unaccompanied by any serious attempt to answer them in detail or elaborate on their relevance to our purposes, is not a substantial contribution to the community.
Edit: Why should I care about assigning numbers in cases like these? Why do I need a better sense of how probabilities will vary at the extremes? Why should I expect to gain such a sense from this task?
And what would a Bayesian justice system even look like? It’s not obvious, and the post provides no guidance.
It’s not obvious, and the post provides no guidance.
The comments fleshed it out, and that’s what I wanted. I reserve the right to ask questions via top-level posts, if I care enough about the question to risk karma on the asking. If it’s a sufficiently good question, the question and answers together represent the “substantial contribution to the community”.
It’s not obvious, and the post provides no guidance.
But the comments did, and that’s what I wanted. I reserve the right to ask questions via top-level posts, if I care enough about the question to risk karma on the asking. If it’s a sufficiently good question, the question and answers together represent the “substantial contribution to the community”.
Transferring from “The Cameron Todd Willingham Test” a meta thread I began:
I don’t think it is a good idea to invoke any sort of controversy without some specific novel point to make. I would not object were it just a thought experiment in an open thread, but good cause is necessary for a top-level post. As I said to Jack, I would hold top-level posts to a higher standard than “don’t see the harm”. By the novel-insight standard, “The Cameron Todd Willingham test” fails on two grounds:
The general idea is an obvious generalization of something already explored in some detail previously: examining the evidence associated with a criminal case and drawing an independent conclusion about the guilt or innocence of the defendant.
The post neither promises nor delivers substantial analysis on any subject.
It is my contention that what Kevin wrote does not show enough effort to justify a top-level post.
It is derivative of the Knox posts. But it asks us to do new things with our skills. In the Knox test the right answer was just that AK and RS have a very low P(Guilty) and RG has a high P(Guilty). This post asks us to compare the relative innocence of two people for two different crimes, with different circumstances. These exercises are important because right now all we can do is basically say “low” or “high” of “fifty-fifty”. I mean, yeah, people gave numbers but I’m skeptical they mean that much. We need to get a better sense for how probabilities vary at the extremes and this will help. [Aside: Frankly I think we should be doing this kind of thing way more often. Someone tried to do this for the probability of a nuclear attack in this months open thread and no one would even try to put a number on their current beliefs (so there couldn’t even be an extended discussion of the issue).]
Kevin also asks the interesting question about a Bayesian justice system which I suspect will lead to a lot of discussion.
And I’m not saying it deserves to be promoted. It doesn’t. But that doesn’t mean it deserves negative karma. A +5 or +6 would be fine.
Is the post a reasonable open thread comment? Yes. But throwing out the bare questions, questions unaccompanied by any serious attempt to answer them in detail or elaborate on their relevance to our purposes, is not a substantial contribution to the community.
Edit: Why should I care about assigning numbers in cases like these? Why do I need a better sense of how probabilities will vary at the extremes? Why should I expect to gain such a sense from this task?
And what would a Bayesian justice system even look like? It’s not obvious, and the post provides no guidance.
The comments fleshed it out, and that’s what I wanted. I reserve the right to ask questions via top-level posts, if I care enough about the question to risk karma on the asking. If it’s a sufficiently good question, the question and answers together represent the “substantial contribution to the community”.
But the comments did, and that’s what I wanted. I reserve the right to ask questions via top-level posts, if I care enough about the question to risk karma on the asking. If it’s a sufficiently good question, the question and answers together represent the “substantial contribution to the community”.