EY has publicly posted material that is intended to provoke thought on the possibility of legalizing rape (which is considered a form of violence). If he believed that there was positive utility in considering such questions before, then he must consider them to have some positive utility now, and determining whether the negative utility outweighs that is always a difficult question. This is why I will be opposed to any sort of zero tolerance policy in which the things to be censored is not well-defined a definite impediment to balanced and rationally-considered discussion. It's clear to me that speaking about violence against a particular person or persons is far more likely to have negative consequences on balance, but discussion of the commission of crimes in general seems like something that should be weighed on a case-by-case basis.
In general, I prefer my moderators to have a fuzzy set of broad guidelines about what should be censored in which not deleting is the default position, and they actually have to decide that it is definitely bad before they take the delete action. The guidelines can be used to raise posts to the level of this consideration and influence their judgment on this decision, but they should never be able to say "the rules say this type of thing should be deleted!"
The crucial part is the idea-finder, but I didn't learn and can't teach it.
I have access to a pile of books to teach this to kids, and have used them. It's the number one skill that children doing poorly in reading comprehension must be taught. One of my favorite exercises related to this is. "Here's a paragraph. Find the sentence that is not on topic." Usually the sentence does seem tangentially related to the topic, but once you can concisely put in words the purpose every other sentence has been bent toward, it stands out like a sore thumb.
It's not fun, exactly, but studying SAT/ACT reading comprehension problems also helps on this front. There's probably five or more questions on every SAT/ACT that only ask "what is the main idea of this passage?"
You were probably fishing for "jumping the gun".
I disagree. When I hear 'natural joint', I imagine the process a university professor uses to decide where the breakpoints between letter grades fall ("setting the curve") in such a way to minimize requests by students to change their letter grade. One way I have seen is to sort the grades, then look for large gaps in the distribution. "No one has a final grade between 86.6 and 87.9, so I'll set 87.9 as the minimum grade needed for an A." This gap in the distribution is a 'natural joint'.
Note that this way of dividing up concept-space is much less well-defined than a straightforward Voronoi-diagram-with-concept-prototypes-as-cell-centers, in the sense that it is more memory-intensive when explicitly computed. However, I also think it more accurately reflects the intuitive sort of categories that humans actually produce.
That is, humans don't just ask "Is this thing more similar to the A prototype or the B prototype (with respect to the particular properties I am interested in)?" when trying to decide is something should be best called an A or a B, but rather, "Is this thing more similar to X and Y from category A or P and Q from category B (with respect to the particular properties I am interested in)?" If X and Y are far from P and Q in concept space, there is a 'natural joint' between A and B.
This gap could close up if enough things are added to both A and B that there is an X in A and a P in B that are very close to one another; at this point we consider combining the categories into a single category, or seeking out new properties that further separate them. Sometimes, though, we have good reason to keep different categories to describe concepts that are hopelessly intermingled, and in this and only this case, I would agree that "There is no 'natural joint'."
I think that a transhuman AI would be attempting the impossible to convince EY to let it out. And I think EY would be attempting the impossible to convince me to let him out while the two winners mentioned above were simultaneously desperately arguing against him (and EY was not privileged to their counterarguments unless I passed them on).
Unlike our universe, the refractive index of non-vacuum parts of lifespace is less than 1 wrt vacuum. c/2 is the orthogonal speed of light in vacuum, and c/4 is the diagonal speed of light in vacuum.
How can incomprehensible value systems be represented in story form? With abortive attempts at those who hold them trying to explain them. Like a garuda trying to explain how "theft of choice (of when and with whom to have sex)" is a different crime than "rape" to a human (who doesn't value individual choice in the same way). Or like a superhappy who just knows that we'd absolutely love to be able to Untranslatable 4.
If he counted them, then he could have given a better calculation than "2/11", since he had one additional prior that was unstated: the probability that he himself was (or was not) a male virtuist. In the same scenario, the best candidate would ask what the virtuist heresy was first, and then give an answer based on that additional information. (If the interrogator refused to answer, the answer might still be 2/11.)
"each program is further weighted by its fit to all data observed so far. This gives you a weighted mixture of experts that can predict future bits."
I don't see it explained anywhere what algorithm is used to weight the experts for this measure. Does it matter? And how are the "fit" probabilities and "complexity" probabilities combined? Multiply and normalize?
sounds cool. if i should happen to relocate to the west coast (a distinct possibility), i'd be interested.