= 27d567bc2d48d9f40e9ccc20ea370b15)
Open Thread June 6 - June 12, 2016
If it's worth saying, but not worth its own post (even in Discussion), then it goes here.
Notes for future OT posters:
1. Please add the 'open_thread' tag.
2. Check if there is an active Open Thread before posting a new one. (Immediately before; refresh the list-of-threads page before posting.)
3. Open Threads should be posted in Discussion, and not Main.
4. Open Threads should start on Monday, and end on Sunday.
= 783df68a0f980790206b9ea87794c5b6)
Loading…= b715d02e85f7b3b4ae65ca3d3e450868)
Subscribe to RSS Feed
= f037147d6e6c911a85753b9abdedda8d){kind=logo}
Comments (126)
Spammer, apparently around since 2015 but not booted then.
Done.
Thanks!
An AMA by Federico Pistono, author of "Robots Will Steal Your Job, But That's OK" and "How to Create a Malevolent Artificial Intelligence."
I'm no expert (paging gwern?) but could an AI have it's code put inside a DNA? Idea from https://en.wikipedia.org/wiki/DNA_digital_data_storage
DNA encodes proteins, basically. So by manipulating the DNA code you could construct the proteins you want. Let's push the idea further and postulate that you can construct a biological creature to your specifications. But then what? That creature will be limited by biological constraints. Wetware doesn't get very fast or very strong.
You mean like humans do?
Ingres updated the Lesswrong survey results to include the write-in responses people gave. I think it's worth posting this here because most people probably missed it. And it was one of the most interesting parts of the survey. Here are some of the files, there are many more on the post:
Community Issues Now Part One
Philosophy Issues Now Part One
Community Issues at Lesswrong's Peak Part One
Philosophy Issues at Lesswrong's Peak Part One
Rejoin Conditions Part One
These were just the writeins. It isn't representative of the average survey taker. There was also a section of multiple choice responses for issues. I think the multiple choice section will be more complete, since most people filled it out with yes or no answers, and it included the main grievances people have. But it hasn't been analyzed yet.
[Link] Nudge Theory
I'm somewhat surprised that Nudge Theory hasn't been discussed or linked here. It seems to be a recap of many well-known biases and such and a set of well-structured counter-measures or exploits. These latter parts being the interesting thing.
Example:
Other are "Sympathy", "Accessibility", "Likeability", "Relevance", "Mood-change", "Fear", "Facilitation", "Sensory". All of these have dark-art potential.
more positive examples are
Counter
By
Counter
By
And so on for many more biases.
If you are a programmer, you might be interested in IPSC 2016, an online competition of individuals and teams of at most three members. Participation is free, you can register online anytime you want (even during the competition). You can use any programming language you want; the programs themselves are not rated, only whether you transform the inputs to outputs correctly. Anyone can participate, but there will be separate ranking for teams of high-school students.
It is on Saturday, June 18th, 11:00-16:00 UTC.
To whom it may concern:
Apparently, if you need antitoxins to botulinum, rabies or tetanus in Ukraine, right now you've got to BYOB.
Does anyone have any good studies on the benefit of jobs, net of pay? I have a vague memory of some, but am not finding them when I look.
Edit: Might help to clarify purpose. I'm inclined to believe the answer is "not much," and that employment is generally a bad thing, apart from the not dying of want bit. I want to see the strongest arguments that I'm wrong, and y'all steelman better than pretty much anyone I know.
Compared to what?
Compared to not having a job presumably. But you raise a good point. It might be easier to find relevant studies by looking for research on the effects of unemployment or retirement.
I can easily come up with very very different subgroups who "do not have a job", e.g.:
The benefits of having a job are likely to be very different for them.
Oh? I was thinking of a study I saw and lost, but differences in benefits to those groups sound fascinating to me also. I would not have guessed the answer to be all that different, again net of pay. I won't ask you to run me a free study (but if you want to... ;) ) but do you have any basic ideas on the matter philosophically?
A study requires data which I neither have nor can easily get :-/
Handwaving my guesses about job benefits...
I see similarities, but the differences are useful too. Thanks for the reply.
I've self-identified as three of those things as the same person (retired, housewife, and independently wealthy ("trust fund kid" feels like I'd imagine the 'n' word feels to a black man or "faggot" to a gay one. Pretty unoffendable myself, but just fyi) )as full disclosure.
If I find the study I want, I'll let you know. Thanks for the help!
So, them's fighting words in your neck of the woods? Does uttering them dramatically raise the probability of someone being punched in the face in the immediate future? or gasp! not being invited to the next bbq?
It's somewhere a little below the 'n' word or the 'r' word, but above "douchebag" or "liberal." As one might imagine, it doesn't come up much. And again, I was commenting on how it feels from the inside, not on how it looks to the audience.
The rankings of insults in subcultures is a fascinating topic :-) What's the "r" word, redneck?
Yes, compared to not having a job. Working for a certain salary vs. getting that salary and not working.
Just some anecdotal evidence:
For some people, jobs provide a scaffolding of their daily time. It's what makes them wake up in the morning. Take the job away from these people and here is what happens: wake up at 10AM, slowly make a breakfast, eat it, watch the TV a bit, make a lunch and eat it, now it's 3PM, feels like too late to do anything meaningful, so just watch the TV, make a dinner and eat it, watch some more TV, go to sleep, repeat over and over again. It seems like people without jobs should logically have more free time, but some of them actually manage to achieve less while unemployed.
For some old people, jobs provide social opportunity that is hard to replace when they retire. Suddenly, instead of spending their day interacting with a dozen people they know, they spend their day alone at home, complaining that their children don't visit them more often.
Without routine, some people's lives "fall apart". Humans are not automatically strategic. In theory, a life without job (with basic income, or retirement money, or just savings that allow you to take a really long vacation) should be better, but in practice sometimes it isn't. Think about all the superstimuli around us: for some people, their job may be the only thing that gets them offline most of the days.
(Disclaimer: This all said, I would still prefer to have a basic income and get extra 8 hours of freedom every day. There is a risk my life would become less amazing than I imagine it, but I would gladly take the risk.)
There is a possible counter-argument that the effects of losing the job might be only temporary. Like, if people have been conditioned for years to organize their lives around their job (and school), of course it gets them out of balance when the job suddenly disappears, because they never had the opportunity to learn how to organize their lives for themselves. But given enough time, they might develop the skill.
Parkinson's Law ("work expands to fill time available") can lead to some pretty ridiculous results.
Thanks for that Vil. I accept these benefits, and agree that there might be a better way, but it's always good to know what issues are likely to occur to one.
I'm a pretty well off person. Recently I went jobless for a while. I missed, above all, the structure. Like, this may sound cliche, but because I didn't have to get up I didn't regulate when I went to bed. So I started to go to bed later and later, and thus, get up later and later.
It isn't that my sleep schedule got off sync with the world's (although, of course, it did that), its that it got off sync with itself. Week by week I'd stay up a little later, get up a little later. It became hard for me to commit to things ahead of time, because I might well sleep through it.
Ultimately, before I started interviewing again, I had to spend a week pretending I had a job. That is, setting my night clock to go to sleep, my wake clock to get up, etc.
Looking back on the months I spent between jobs, I accomplished less towards my day trading and writing goals than I did in an equivalent period in my last gig. For me, at least, jobs are good for more than money. I doubt I could make myself do them if they didn't pay, mind, but that seems to be the reasonable truth.
Thanks for that Walter. I feel like I have a sense of this phenomenon (I retired at 33 and was expecting to be WAY more bored and less satisfied than I am), and am very interested in a full study (I haven't found it; any of y'all run psychological studies for money?), but I hear more stories like yours than the alternative.
Were you looking for employment, or were you on the fence between retirement and a break? Knowing you have to go back eventually might be a factor.
Why are you interested in a study? Studies typically tell you about the averages and in many cases the averages are not what you need. In some cases, they are, actually, what no one needs.
Some people fall apart without externally imposed structure, but some people thrive in the absence of constraints. The latter are often called "self-directed" or "self-motivated" or some other term like that. Both types exist, not to mention the intermediate cases, of course.
Because I plan on doing some more serious campaigning for a more aggressive GBI (among other things) than what a lot of people advocate. I plan on making the case that there is absolutely nothing wrong with someone deciding to just live off the dole and not work, and that people who choose that are often more in the way of people making progress than helping them when they show up to clock hours. I also plan to assert that people who don't have to work, effectively on penalty of death if I want to sound dramatic, will have a better wheat/chaff ratio for what they do do. Of course I want to make sure it's true first :P
If I get to design the study it'll be a little different from the study I seek because I don't think I can expect anyone to have done that study. And I'll be more interested in the whole study than the executive summary for exactly the reasons you describe.
Within which framework? From an individual point of view, sure. From the point of view of the society, not so much -- someone has to produce value which this person will consume. Arguing that it's psychologically healthy to not work isn't a relevant argument here.
You know that the primary function of the markets is provide incentives for wheat and disincentives for chaff, right? They perform this function quite well. You will argue that without the guiding prod of the market people will produce more of better stuff all by themselves?
Well I have a much longer argument for this in the book, but I propose that the amount of work people will do because they want to is more than enough to run society. 40h per person per week (ish) is, in my view, largely makework.
Of course.
The markets have a major confound, imo, in the form of pro-job policy. I believe, and I have some support for this but not enough to prove the point yet, that if "jobs creation" did not occur as a political activity, the market would normalize below the level people would produce without, again using the provocative language descriptively not manipulatively, the lingering threat of dying of want.
That's classic Communist utopia straight out of Karl Marx. "From each according to his ability, to each according to his needs". Why do you believe this to be true?
Another question, related to the role of the markets as conduits of information, is why do you think the work people will do is the work that other people need? As a first-order approximation I would expect that you won't have any problems having your portrait painted, but your clogged toilet will stay clogged for a long time.
First, that's not self-evident. Job creation policies mostly reallocate labour (from productive use to less productive). Getting rid of make-work jobs, in the absence of other regulations, will just free up these people to be employed in other areas where their talents can be utilized better. The net effect would be higher productivity but not necessarily a lower level of employment.
Besides, do you want this "below the level"? You interpret this a lots of leisure. I interpret this as a poor society.
Initially I thought I'd take a few months, veg out, etc. After I lived that way for a while though I became interested in getting a new job again. Took a couple weeks from deciding to work again to getting the job.
So, sort of both? Started out with no strong opinions on work again vs. not, migrated to looking for job.
Right on. Thanks!
Apologies for the digression and please feel free not to answer, but I'm curious: How were you able to retire at 33? Successful startup? Family money? Working in finance plus reasonable frugality?
Just genetic lottery. My family owns a chain of convenience stores in upstate NY, and after some time in banking I decided I'd prefer not to work any more. I am writing a book, but I don't feel comfortable calling myself an author until I publish it.
I'm comfortable talking about it as long as I don't feel like I'm being perceived as bragging about something over which I have no control (which is stupid, and which I see in other people all the time)
I guess the important question for most of us is: "Is that something I could try to replicate?" Obviously no, unless your family is planning to adopt some LessWrongers. :D
Still, we can hope that your experiences will be useful for our children one day.
Yup. And being at K&C for most of my life, any windfalls get passed forward, so I'm not competing with anyone for those going forward. But not especially helpful in replication.
Oops. Hit a one way button. Will just use edit to rewrite next time But now I know what "retract" does :)
Wow. Look at the "Top Contributors, 30 Days" on the sidebar. Overall activity perhaps has found a new low here on LW.
What was the previous low?
There was a low point just before the LW Survey came out and everyone got the mutual karma bump for posting "I did the survey" temporarily inflating contributor scores. But now all those +1s have fallen outside the 30 day window, and not even the open thread is attracting comments any longer.
I don't know. Not even sure this is the lowest. It looks pretty bleak as far as activity though. I would love to see some data visualization detailing the downfall...
Connotational disclaimer: There is nothing necessarily wrong with this.
I mean, if all people in the "Top Contributors, 30 Days" list would now write thousand new one-karma comments, the numbers would increase, but the quality of the website probably wouldn't improve.
Oh, dear. Psychology, can you please get your act together and stop being an embarrassing mess?
Is this some kind of meta-psychological experiment? At least let's hope it is...
Maybe it's a clever way to find out political bias in academy. Write two papers claiming that one or the other side of the political spectrum is scientifically linked to insanity, submit both papers to random journals, and see which version becomes the official academic "truth".
Just kidding, but it would be an awesome experiment.
If it is, it's about gullibility, isn't it? X-/
VC capital and startups in AI and machine learning
http://renewableelectron.com/venture-capital/
and an interesting article on a Mars constitution, based on the Antarctic an Outer Space Treaties.
http://qz.com/702624/as-silicon-valley-lays-plans-to-colonize-mars-researchers-offer-a-blueprint-for-governing-it/
Which words?
I'm currently working with an uncommon non-English language, and in the near future, may have the opportunity to ask a native speaker for translations of terms not available in any existing dictionary. Which words, terms, phrases, and ideas do you think it's most important to be able to use?
Eg, to start with, I'm going to make sure any gaps in the basics of math and physics are covered: negative numbers, chemical elements, planet names; the sorts of things you'd find in a typical SETI primer. And I'll be including some present-day technologies that were invented since the most recent dictionary available - cellphones, the Internet, etc.
But if I can convince this individual to supply translations for concepts such as "existential risk", "intelligence explosion", "cryonics", or "decision theory", which ones should I actually ask about? (Especially as they have their own life, and I may only be able to get so many translations.)
Put another way, for people who aren't using English, what post-1970ish ideas are the most important ones to have specific words for to be able to talk about?
Put yet another way, as a thought experiment, if you were involved in helping put together a new conlang like Klingon, which concepts would you want to be sure the language included?
Put still another way, for a language that simply adopted the English word 'cheese' for the new concept, which post-1970ish-concept words do you think should be created from the language's own building-blocks, and which should use English's trick of simply incorporating a foreign term?
This actually happens all the time -- some new word is invented, often in English, and now the other languages have to find a way to deal with it.
One possibility is to take the English word, and do the minimum necessary modification to make it feel natural, such as add or modify a suffix... more or less what a neural network trained on similar kinds of words would do. Another possibility, if the new word is composed from existing roots, use the translations of those roots and try to connect them in similar way.
Sometimes the rules do not allow an analogical operation, for example, in some hypothetical language it could be very unusual to derive the adjective "existential" from the noun "existence", so you might end up with a more clumsy phrase for "existential risk" such as "dangers threatening the existence" or similar. Maybe not this specific example, but sometimes other languages are less flexible in some places than English.
Ultimately, the rules are not 100% exact; even if you translate the term breaking some of the rules, as soon as the translation gains momentum, people will use it, even if some language purists would complain.
Looks like they have narrowed down the telomere argument to show that stem cell exhaustion is the key to, at least blood and immune system support. Looks like birth cord/placental blood storage from your birth is the best life extension insurance....
https://www.newscientist.com/article/dn25458-blood-of-worlds-oldest-woman-hints-at-limits-of-life