Eliezer

49 ms·

481.
by Eliezer 14y ago
What makes you think it works on small family groups? All we can conclude from ev-psych is that making stuff up didn't cost the imaginers much reproductive success personally - we have no idea what it did to the actual victims having stori
482.
by Eliezer 14y ago
Not criticism of Musk, just a general observation: "I think" humility is super cheap, it takes 0.3 seconds or 8 keypresses to add it like table salt. Stating concrete exceptions to your generalizations is a much stronger indicator of non-
483.
by Eliezer 14y ago
Taking money from everyone to fund science... could be argued, I suppose. Taking money from successful Silicon Valley entrepreneurs to fund government science would be merely stupid. In other words, what I object to about the parent comme
484.
by Eliezer 14y ago
Allow me to join with PG in congratulating you on a great snappy comeback. Ignore the haterz, PG gets it worse and you wouldn't believe the ones I get.
485.
by Eliezer 14y ago
Sometimes people appear to me unskilled, sometimes they appear unknowledgeable, sometimes they're thinking in twisted ways that they picked up from their environment, sometimes they're falling prey to cognitive biases, sometimes they're mak
486.
by Eliezer 14y ago
You know, you're right (about the dangers of morality clauses). Consider my original content 70% retracted or at least flagged for requiring modification. Maybe you could have a list up-front of who you don't deal with (Scientology, patent
487.
by Eliezer 14y ago
Scientology certainly operates in violation of the letter of the law - framing people for murder is illegal, I'm pretty sure - they just avoid conviction by bribing local judges and so on. I don't see why you can't just tell the patent trol
488.
by Eliezer 14y ago
Graham's number is VASTLY less than TREE(3), so you lose, actually. Graham's number is around A(64)(4), i.e., the result of iterating the number of Knuth arrows 64 times starting with 4 Knuth arrows, while A(A(187196))(1) is an extremely w
489.
by Eliezer 14y ago
Got a ref on the 5 percent part? (Ignore this if it's only for gas; I'd be interested in 5% all purchases.)
490.
by Eliezer 14y ago
Did you ever have a word with your history teacher? Did anyone in your class ever stand up and contradict the history teacher, about anything, and not get slapped down for it? Did the history teacher ever once say, "Huh, I guess you're rig
491.
by Eliezer 14y ago
Belief is subjective. Reality is objective. Since "truth" compares belief and reality, the output is not subjective. The belief that I can fly can make me expect to fly away when I jump off a cliff; only the truth of that belief can prev
492.
by Eliezer 15y ago
I tried to schedule a sleep study, was told it would be a month, tried to price a sleep study somewhere else, was told it would be over $2000... ...gave up and bought an automatic CPAP machine off Craigslist for $650 from someone who knew h
493.
by Eliezer 15y ago
The next space race will be between Google and Paypal.
494.
by Eliezer 15y ago
Gwern, the main surprise coming out of the post-Summit workshops was that literally everyone wished that uploads could come before AGIs. The problem isn't that high-fidelity uploads are less trustworthy. The problem is that the neurosci
495.
by Eliezer 15y ago
That doesn't quite sound like what the content of the article says - it's more, "We used the two algorithms from the first Progress Prize that gave us most of the benefit, and the 107 blended algorithms required to get the next 1.7% improve
496.
by Eliezer 15y ago
I don't consider myself a big-L libertarian, but I can't help but notice the "Regulation failed, we need more regulation!" template. If the incredibly burdensome Sarbanes-Oxley was nonetheless "inadequate", then maybe you just can't protec
497.
by Eliezer 15y ago
Your statement is false as a matter of fact. We have a one-year life-outcomes followup study planned, but the preliminary indications don't fit with 'status quo', and there are plenty of participant comments on the LW thread which explicit
498.
by Eliezer 15y ago
Can we have a list of startups that are doing this? Quixey would be on it, and I'm pretty sure there's one or two others I've heard of, but I don't remember offhand.
499.
by Eliezer 15y ago
You shouldn't trust yourself, because the outside view is that most people don't get around to it. Give 1% at start; you can always give more later.
500.
by Eliezer 15y ago
I overwhelmingly disagree. The number two limiting factor on philanthropy is that, despite good intentions, people just don't actually get around to it . Giving away 1% at startup-start is a supremely good suggestion just because it mea
501.
by Eliezer 15y ago
No, we took surveys and everyone except one person rated in the 9-10 range, I believe.
502.
by Eliezer 15y ago
It's possible that somebody is confusing this with an LW meetup, or possibly the Megacamp (that didn't work nearly as well as the one-week Minicamp which is why we aren't repeating it), or something else. Attendees of the previous Minicamp
503.
From 2,000 words per day to 10,000 with self-experimentation(thisblogisaploy.blogspot.com)
5 points by Eliezer 15y ago|0 comments
504.
by Eliezer 15y ago
Ahhhhhh. They really ought to state the customer use-case on-blog, I use Dropbox on only my own computers and didn't understand the point at all.
505.
Be Specific (especially during PG's office hours)(lesswrong.com)
113 points by Eliezer 15y ago|30 comments
506.
by Eliezer 15y ago
If you think most comments are bad, and you have an explicit prior that reflects this, then your ranking will update to "Hey, the comment was actually good!" at the optimal speed - no more, no less - given the rate of incoming ratings. Why
507.
by Eliezer 15y ago
It's neither precise, nor elegant, and certainly not rigorous. On this approach, an item with 1 upvote and 2 downvotes will be ranked below an item with 1000 upvotes and 2000 downvotes, as one commenter pointed out; and in general, bright
508.
by Eliezer 15y ago
I read the article, went "The ^&#$?" and came here to post this, more or less. It's amazing the $&#! people will get up to when they don't know Bayesian statistics.
509.
by Eliezer 15y ago
I think the off-the-shelf solution to this is to purchase fuzzies and utilons separately - be Batman if it keeps you going, but maybe also give $10,000/year to efficient charities on the side? http://lesswrong.com/lw/6z/purchase_fuzzies_an
510.
by Eliezer 15y ago
Awesome! Even $10 would have satisfied the intent of the question. It's people who seem to understand the concept but give literally zero that make me want to hit something with a stick.

More ›