Posts
Comments
I guess I didn't make myself at all clear on that point, I ascribe to both of the above!
Another way to avoid the paradox is to care about other people's satisfaction (more complicated than that, but that's not the point) from their point of view, which encompasses their frame of reference.
Another way perhaps is to restate implementing improvements as soon as possible as maximizing total goodness in (the future of) the universe. Particularly, if an improvement could only be implemented once, but it would be twice as effective tomorrow instead of today, do it tomorrow.
The probability distribution part is better, though I still don't see how software that uses randomness doesn't fall under that (likewise: compression, image recognition, signal processing, and decision making algorithms).
Any software that uses randomness requires you to meet a probability distribution over its inputs, namely that the random input needs to be random. I assume that you're not claiming that this breaks modularity, as you advocate the use of randomness in algorithms. Why?
(idle bemusement)
Does an optimal superintelligence regret? They know they couldn't have made a better choice given its past information about the environment. How is regret useful in that case?
So you're differentiating between properties where the probability of [0 1 2 3] is 1-ɛ while >3 is ɛ and probability distributions where the probability of 0 is 0.01, 1 is 0.003, etc? Got it. The only algorithms that I can think of that require the latter are those that require uniformly random input. I don't think those violate modularity though, as any are of the program that interfaces with that module must provide independently random input (which would be the straightforward way to meet that requirement with an arbitrary distribution).
There's a difference between requiring and being optimized for though, and there are lots of algorithms that are optimized for particular inputs. Sort algorithms are a excellent example, if most of your lists are almost already sorted, there algorithms that are cheaper on average, but might take a long time with a number of rare orderings.
Requiring that the inputs to a piece of software follow some probability distribution is the opposite of being modular.
What? There is very little software that doesn't require inputs to follow some probability distribution. When provided with input that doesn't match that (often very narrow) distribution programs will throw it away, give up, or have problems.
You seem to have put a lot more thought into your other points, could you expand upon this a little more?
The king was proposing that Orin bet 1kc, of which they only have 800c currently, in order to receive 20kc (which is twenty five times their net worth). The 200c debt was what Orin would be reduced to if they were wrong.
In such cases I'll say, "Oh! Interesting... how does that work exactly?" It seems to work out alright, and I would guess that other methods of asking for more information without implying that their statement is false are equally effective.
An addendum to [1], social security tax in the US is capped, with the cutoff being around $105k of individual income, so there may be a local dip there in percentage where the increasing income tax doesn't balance the 11% that goes to social security before that point.
Might a one half point penalty for down voting change the incentives enough to prevent mass down voting? Perhaps combined with Viliam_Bur's minimum karma suggestion. Generally I favor ideas that don't make more work for the moderators.
(I am not imagining having half karma points, rather docking one karma for every two (or n) down votes.)
Weak evidence though, easily overcome by being open to whatever things your tribe denounces.
Nothing like a good idea to get lots of names.
The "be a sheep" voting system is also known as assignment voting.
How can we share evidence effectively? More generally, how can aspiring human rationalists make group decisions?
"Arrogant pride", or at least arrogance, is listed among the traits already, that doesn't add anything to the meaning.
Imagine I'm writing a computer program that models people. People who exhibit a high number of these traits gain the label narcissist. Does this label allow me to make better or more efficient predictions about the person than the individual traits? What sorts of predictions?
A little brief reading elsewhere still doesn't explain what narcissist means here. Is it just a word that means "this group of symptoms (presumably commonly found together)"? If it doesn't mean anything else, its use carries strong negative connotations that distract from the traits it encompasses.
The argument would be stronger if I could use the "narcissistic personality disorder" test for Petunia too,
I was under the impression that you identified Petunia as narcissistic first, and was using that as a predictor for Harry's, my confusion.
While I enjoyed reading this, amused and intrigued, I am somewhat hesitant to accept it, lacking a search for traits that are anti-correlated and without exploring alternative theories explaining these traits. I'm also a little confused by
I predicted that if my theory is correct then Harry would have a narcissistic personality. To test this, I found a list of personality traits that describe a narcissist (by Googling “children of narcissistic parents” and clicking the first link),
wouldn't this be a list of traits of children of narcissistic parents? If it's not, why did you use a different list to determine if Petunia was narcissistic?
On the topic of the specific traits, we could use a lot more information about which traits are useful and which aren't. I understand there are a number of folks around here who have worked to acquire some traits, self-awareness and mindfulness (which are very clearly useful) come to mind, but they appear (very roughly) to be on a different level than, say, need for praise. With that, thank you for inspiring me to look more into them.
- Assigning probabilities of falsehood is counter-intuitive to many, using two statements would allow for the typical direct assignment of truth.
You can still assign probabilities of truth with three statements, they would merely sum to two instead of one.
My cursory understanding is that none of these proofs apply to rating systems, only ranking systems, correct?
New programmers (not jimrandomh), be wary of line counts! It's very easy for a programmer who's not yet ready for a 10k line project to turn it into a 50k lines. I agree with the progression of skills though.
You mean that most cognitive skills can be taught in multiple ways, and you don't see why those taught by programming are any different? Or do you have a specific skill taught by programming in mind, and think there's other ways to learn it?
On your last point, no. To put it colloquially, the simpler answer is more likely.
In practical terms, saying that all binary choices without evidence causes conflicts. Namely, if there's a 50/50 chance that your consciousness dissolves when you die, and a 50/50 chance that a hidden FAI captures your brain state, and a 50/50 chance that a hidden UFAI captures your brain state. That implies that in every case that a FAI captures your brain state so does a UFAI.
If the farmer is actually a subsistence farmer, they save their own seed, so they don't care about the price of seed, nor would they buy single generation seed (say of a new crop or variety) knowingly. However, if someone nearby plants single generation seed, they can end up with genetic material in their variety of that crop, which cuts their germination (or seed baring) rates the following season.
Subsistence farmers don't trade for their sustenance, they farm what they subsist upon. So perhaps the moral is do be a subsistence farmer...
Also, avoiding dying in ways that destroy brain state. I'm not sure how probable those are, or how easy they are to avoid, and if that includes dementia (and so on) it gets rather common and tricky.
Because your cars run on gasoline and would have filled the tunnels with choking fumes, without either (1) a big expensive ventilation system, or (2) expensive electrified rails that would...impose friction costs?
For 2) you can use magnetic levitation: Inductrack which now manifesting out as Skytran. (Costs are about 1/50th your transportation mass in magnets.)
I know at the moment it seems to me that the colors are far enough apart that light conditions at my PC are not the main problem. Your monitor emits light, so the light conditions matter less, mostly needing to overcome the ambient light (laptop in sun).
Most things don't produce their own color though, they reflect varying amounts of the incoming spectrum. If that incoming spectrum is different, the outgoing spectrum is different. You can take advantage of that in various ways, but it might also confound the question of what color "is" this object.
Or maybe you automatically take that into account by using the ambient light as a reference, I was wondering whether you had tested for that or not?
Under different light contexts an reflective object might be closer to either midnight blue or navy. Have you attempted using paint chips or something to test yourself under sunlight versus florescent light or anything?
Also, for sound perception: sox(1)
The relevant studies all seem to be behind paywalls, but the graphs from this one looked promising.
Jay L. Zagorsky (2007): "Do You Have to Be Smart to Be Rich? The Impact of IQ on Wealth, Income and Financial Distress." Intelligence, 35 (5), 489-501.
If someone starts a meetup in a small town, it would not be difficult for them to get a newspaper article talking about the event. Though I'm not sure Wikipedians would consider little-tiny-newspaper to be adequate coverage...
I admit to hyperbole, now, with a little more thought, I would have worded it differently. Both to clarify that it's pretty far down on our list of societal problems, and that it's more an individual level mistake rather than a systematic one (though there are systematic benefits to fewer flush toilets).
Translate it to "In x% of new non-urban houses, there are options better than flush toilets." My confidence in my confidence assignment isn't very high yet though, so I am quite open to being overconfident.
And obviously both lists are non-exhaustive.
- Flush toilets handle large numbers of people for a long time fairly easily.
- Flush toilets get clogged.
I suppose my actual belief is that flush toilets are a mistake outside of urban areas, I don't have much experience with urban living or what other poop strategies could work with it.
Advantages, flush toilets:
- Provide easy long distance transport of human waste in urban environments.
- Exchanges weekly-to-yearly chores for purchased services.
Disadvantages, flush toilets:
- Create additional dependency on water (and by extension outside water districts, electricity).
- Turn (vast amounts of) drinking water into black water.
- Create a waste product from human manure, which is a valuable resource (fertile soil) when dealt with properly.
- Adds significantly to the cost of housing (especially outside sewer districts).
Irrationality Game: (meta, I like this idea)
Flush toilets are a horrible mistake. 7b/99%
Most teaching jobs around here involve significant use of internet capable machines for grading, communication with other teachers and administration, and increasingly communication with students. Mathematics is probably more resistant to online teaching materials than most subjects though, and you may be able to find a school that eschews such things.
I wonder how much people's interactions with other aspiring rationalists in real life has any effect on this problem. Specifically, I think people who have become/are used to being significantly better at forming true beliefs than everyone around them will tend to discount other people's opinions more.
New datapoint on mass downvoting:
Sometime between this comment and my last comment, approximately all of my comments were downvoted exactly once. Seems kinda strange.
(I don't have anything that I want to post to Main prepared anyway, so karma's kinda a moot point, but I hoped this could be helpful if anyone ever does look into it, and times are included in downvote logs.)
Edit: Hehe! And within ten minutes, this one joined the rest of them...
Perhaps the reference is to "nutritional yeast", which are all dead, and won't impact your gut bacteria aside from being provided with more nutrients.
The adventurer probably does the most for me, finding new paths and places and people brings such delight. My conscious identification as a person sidestepped a bout of gender-confusion when I realised I hadn't ever identified as man or woman, merely with pieces of each.
I'm less sure about, say, my combination of someone who "gets it done" and "doesn't show imperfect work". The majority of interesting work that gets done is done coincidentally, because it needs doing, and isn't up to my standards. I've been experimenting with ways to overcome this, but with social commitments to share interesting work, not with identity changes.
I have been for a very long time a cryptic, who doesn't bare theirself to strangers, much less the public, and likes to play with words more than express clearly, I'm slowly replacing that with "an open person". I conciously push to show a more vulnerable bit of myself before I normally would these days, so far just with people that I imagine eventually sharing with anyway. I got burned pretty badly my second time trying that, but the the first time had me already completely convinced. Plus, I never could have written this if I hadn't.
I've never used Beeminder, but I find social commitment works well instead. Even teling someone who has no way to check aside from asking me helps a lot. That might be less effective if you're willing to lie though.
An alternative would be to exchange commitments with a friend, proportional to your incomes...
There has been major progress with at least one autoimmune disorder over the last 30 years. Even if there has not been an outright cure, the life expectancy of someone with HIV/AIDS, who can get treatment, is far greater now.
edit: clarifying, in case the downvoter didn't understand the combination I responded to.
Update on the collaboration:
One person has contacted me so far. We're each prototyping to our own vision with plans to share our results (with eachother at least) at some point. We'd love to exchange previews with more people, please don't let our working on it stop you from prototyping your own vision of how it might work.
And I completely agree that at this point it would be more efficient to find new audience at universities.
Less time required to reach a given number of people who are going to join in, agreed. Yet the translation can help raise the sanity waterline of a group of people that would not even consider coming to a 'rationality meetup'. I go to meetups because of the sequences, because it's worth a three to five hour journey to hang out with people who share that.
The people who gain the most from structured arguments are the people who don't need to sift through ten blog posts and hundreds of comments. The gains for the writers are more along the lines of less time reiterating arguments in different contexts.
I've been working on a tool like this. Done well it would be applicable to more than just debate... If folks want to collaborate, I'm interested.
Hence the actually useful numbers bit! Yet I do care to some extent, if for some reason I end up there in future, just less then everyone here and now. Maybe one could weight populations by inverse distance?
That's not accurate; did you mean 13/10000?
Thanks, fixed.
I have a hard time seeing people as replaceable, much less easially. Even between two people who fit some abstracted ideal, one won't replace another. Leaving that aside though, I think that the difficulty is more in finding the people who fit that ideal than their actual existence.
Of course... I thought 100 was meant to be the global mean. Lynn set Great Britian's mean, nothing like a flexible definition!
The (not very good) data doesn't bear out a 90ish global mean though, the sub-90 IQ countries are much lower population than over 90. To be pessimistic I'd take another half sigma. (92.5)
- World Population: 7 billion
- 145+ IQ (1/4200): 17 million
- Male (1/2): 8 million
- College-aged (1/10): 800 thousand
- Normal weight (3/5): 480 thousand
Actually useful numbers may be able to be obtained by using more locale specific filters.
- World Population: 7 billion
- 145+ IQ (13/10000): 93 million
- Male (1/2): 46 million
- College-aged (1/10): 4 million
- Normal weight (3/5): 2 million
Transhumanist values are probably higher than average in that group, but I have no idea of numbers there. Clicking with you and a more refined definition of attraction I can't speak to, but if you've come in contact with 5 in your time at college... There's still lots.