Less Wrong is a community blog devoted to refining the art of human rationality. Please visit our About page for more information.

Comment author: pepe_prime 13 September 2017 01:20:21PM 10 points [-]

[Survey Taken Thread]

By ancient tradition, if you take the survey you may comment saying you have done so here, and people will upvote you and you will get karma.

Let's make these comments a reply to this post. That way we continue the tradition, but keep the discussion a bit cleaner.

Comment author: SolveIt 15 September 2017 11:18:52PM 17 points [-]

I have taken the survey!

Comment author: Lumifer 07 July 2017 04:21:21PM *  1 point [-]

"naturals" at their chosen activity

It's not a function of activity, it's a learning style (to use an overused expression) which applies to all kinds of activities.

All my attempts to learn stuff in a supportive setting worked very well, and all my attempts to learn stuff in isolation failed.

Right, but the typical mind fallacy is a thing. I'm your opposite -- I learn best by myself and a class/group just gets in the way for me -- but I know that people unlike me exist :-)

Comment author: SolveIt 10 July 2017 09:36:25AM 0 points [-]

There's also a strong selection effect. Guess what kind of people you'll meet in classes!

Comment author: username2 09 July 2017 04:46:36PM 2 points [-]

Also the first few dozen chapters of HPMoR are terribly written. It is rather horrid, strained, constipated writing. Particularly if you view the early releases of the text, not the revised text that is currently available. The writing got decently good towards the middle, and was top notch by the end. But that was after thousands of pages written and lots of feedback on every chapter. No surprise, lots of writing practice and (critically, to the point of this thread:) feedback leads to becoming a better writer.

Comment author: SolveIt 09 July 2017 09:32:28PM 1 point [-]

I feel uncomfortable criticizing HPMoR for its writing when it clearly succeeded at its job beyond all expectation.

Comment author: wizard 09 July 2017 01:24:36PM *  0 points [-]

Title is misleading, I thought you were going to talk about self-improvement in general for which this was an interesting statement but it's about academic learning so why not "against lone-wolf learning"? What you said is both, (1) already consensus pretty much everywhere, everyone thinks taking classes is best, it's improbable that even someone from LW isn't aware of this viewpoint (2) clearly wrong, but not something I feel discussing and I'm only replying to complain that I was click-baited into reading this.

Comment author: SolveIt 09 July 2017 09:27:56PM 0 points [-]

I don't see how this is limited to academic learning.

Comment author: blankcanvas 30 June 2017 07:14:12PM *  0 points [-]

I'm a high school dropout with my IQ in the low 120's to 130. I want to do my part and build a safe AGI, but it will take 7 years to finish high school and a bachelor and master's. I have no math or programming skills. What would you do in my situation? Should I forget about AGI and do what exactly?

If I work on a high school curriculum it doesn't feel like I am getting closer to building an AGI, neither do I think working on a bachelor would either. I'm questioning if I really want to do AGI work or am capable of it, compared let's say if my IQ was in the 140-160's.

Comment author: SolveIt 01 July 2017 11:28:21AM 1 point [-]

For almost all people, their comparative advantage won't be in AI research, and they'd do more good doing whatever they're best placed to do, and donating a portion of their income.

You don't give enough detail for us to give specific suggestions, but unless you have extraordinarily compelling reasons to think that you were born to be an AI researcher, I wouldn't recommend making major life changes for the sole purpose of maybe becoming a fairly average AI researcher in ~10 years.

Comment author: Duncan_Sabien 01 June 2017 03:10:42AM *  1 point [-]

I ask, on a meta level: was this question rhetorical?

Because I suspect there's literally no answer I could give that would satisfy you (but I hold that suspicion lightly, and will believe you if you tell me I'm wrong).

The thing is, any true and honest answer to "how will you defuse that situation?" is something like 20% principles and 80% context/subtlety/reactions in the moment.

The answer to your question is yes, I can defuse the situation, and the confidence in the yes comes from the fact that I have defused such situations before—when other people caused them. I've also defused such situations when caused by me, but outside of the context of fights within a romantic relationship (which I claim is a special case, and where I also played the role of defuser more than the role of exploder when I couldn't just head things off at the pass in the first place), the last time I caused such a situation was about sixteen years ago. I learned how to not cause them.

And no, I can't give you a blow-by-blow that will sound convincing, because again, it's all context. And in other places in this post, where I listed general principles and heuristics, people who were already predisposed to be hostile pointed out that, from their perspective, it sounded a lot like empty platitudes.

So I'm curious what the point of the question was, and if it was to honestly ask, I'm curious what sort of answer would actually satisfy you.

I'll take silence to mean "it was a rhetorical question."

Comment author: SolveIt 01 June 2017 11:40:48AM 1 point [-]

Yes, that particular question was rhetorical.

But my more general point is that I think you're wildly overconfident in your ability to manage difficult social situations because I think very few people could successfully navigate the issues that will arise if this goes wrong, and you haven't given me enough reason to think that you're extraordinarily good. What little I know of you (this comments section) points towards you being a fairly regular person that gets upset when people pummel you with unfair criticism and reacts in fairly regular ways. I am not convinced that is good enough to undertake a dangerous and BINDING venture.

Since I think it would take an extraordinary person to pull off a soft landing if this goes catastrophically wrong, it would take rather extraordinary evidence to convince me that you are such a person. The sort of answer that would satisfy me is of the sort that involves a good number of other people testifying that they know based on experience that you would be able to handle the worst-case scenarios.

Comment author: Duncan_Sabien 01 June 2017 02:19:09AM *  1 point [-]

You can connect those dots, but I do not. In particular, I'm less flipping out at handoflixue in particular, and more loudly signaling strong rejection of what they're doing. In other words, it's much, much, much more about "everyone else" at this point than it is about handoflixue, who I made a policy level decision not to cooperate with many comments back. I reject the implicit assumption in your post that "always be quiet/calm/nice/polite back" is actually a good rule—in real life, Gandhi only wins against an enemy who's willing to update, and however much handoflixue has indeed rolled back their tone, they haven't even tried to stop strawmanning and jumping to conclusions.

You can certainly disagree with me about whether these policies I'm following are net good or optimized in ways you'd endorse, and that's entirely cool—the point is not to please everyone, but to be 1) principled, 2) consistent, and 3) transparent. Nobody who would enter this experiment (i.e. be intrinsically interested AND make it past all the filters) will end up behaving as poorly as handoflixue. That's kind of the whole point of filters—to prevent people who embrace and endorse unacceptable-according-to-the-subgroup behavior.

Comment author: SolveIt 01 June 2017 03:01:17AM 1 point [-]

I think you're being wildly optimistic about your vetting procedures. I don't think you can reliably predict how people will react in high-stress situations with your filters.

in real life, Gandhi only wins against an enemy who's willing to update

Well too bad, because your hypothetical screaming roommate isn't willing to update, and they're screaming in your face at 2am anyway. Can you defuse the situation? Or will you end up with, at best, a messy eviction that's traumatizing for all parties involved?

Comment author: Duncan_Sabien 01 June 2017 12:26:01AM *  3 points [-]

Point of fact/order: I have recruited ZERO people as a result of this post; that was never its intention, I already had a set of ~20 people plausibly interested and THIS IS WHY I CONTINUE TO ENGAGE WITH EVERYONE OTHER THAN YOU, STOP SLIPPING IN STRAWMANNED NEGATIVE STEREOTYPES INTO THE VAST MAJORITY OF YOUR COMMUNICATION HOLY CRAAAAAAAAP.

Only one new person has expressed interest, and has greater than thirty percent odds of getting in; by this point, I feel justified in saying you're a jerk; get somebody else to post your reservations if you want them addressed. You have BY FAR earned the right to be ignored.

(I'm curious what sort of mental process leads you to be overconfident in a false/straw conclusion ten times in a row, and yet still not pause and do any sort of meta-check the eleventh time, but alas, I shall not find out.)

Comment author: SolveIt 01 June 2017 01:22:37AM 5 points [-]

You flipping out in response to text comments, despite having the luxury of time and privacy to compose your responses doesn't bode well for how you'd react to a member screaming in your face about how you hoodwinked them into an abusive arrangement.

You may feel that handoflixue is strawmanning you, assuming bad faith, etc, but the person screaming in your face could do much much worse, even if you did everything right! If you can't handle this level of criticism gracefully, you're not fit to lead anything like your proposal.

Comment author: Alicorn 17 March 2017 01:46:56AM 21 points [-]

If you like this idea but have nothing much to say please comment under this comment so there can be a record of interested parties.

Comment author: SolveIt 19 March 2017 01:23:41AM 2 points [-]

I am interested!

Comment author: chaosmage 10 February 2017 05:01:29PM 3 points [-]

The simple inroad would be intellectual disability.

Right now you're disabled if your IQ is below 70 and you have trouble functioning in your everyday life. These are 2 to 3 % of the population and there's a societal framework already in place for them.

If you could gradually raise that IQ threshold, you'd achieve much of what you want to achieve here.

I don't know who determines that threshold, but whoever it is is probably more approachable, and more likely to listen to reason, than the public at large.

Comment author: SolveIt 12 February 2017 01:43:22PM 1 point [-]

Indeed. I'd also like to point out that even though already having this framework in place, we're pretty much clueless om what to do about it. This is despite the fact that these cases should be the most treatable!

View more: Next