Open Thread March 2019
If it’s worth saying, but not worth its own post, you can put it here.
Also, if you are new to LessWrong and want to introduce yourself, this is the place to do it. Personal stories, anecdotes, or just general comments on how you found us and what you hope to get from the site and community are welcome. If you want to explore the community more, I recommend reading the Library, checking recent Curated posts, and seeing if there are any meetups in your area.
The Open Thread sequence is here.
I was reflecting on how hard it is to get up to date on climate science recently, and I thought about Reporting likelihoods, not p-values again. So I did some searching, and while I was able to come up with plenty of discussion about absolute and relative likelihood functions, and some tutorials, and various demystify p-values posts, I didn’t see any papers which used them or reports of success using the method.
My expectation is that despite how good the idea seems, the inadequate equilibria of publishing remains, so no surprises there. So now I am wondering about how easy or hard it is to convert already published papers from p-values to likelihood functions. Part of the complaint about p-values is that papers were traditionally opaque about their statistical analysis and did not share their data sets, so overall I expect the problem to be hard. It seems to me I have seen a lot more about success with publishing data sets and sharing analysis software though, so if papers obeying those good practices are chosen that barrier would be overcome. If it is possible to publish a paper of the type “I converted these 5 other papers into likelihood functions and got an interesting result” then there is the added benefit of piling citations on to the papers which use the other best practices.
If journals-gonna-journal and so publishing that is impossible, and a conversion procedure were to be simple enough, it would be worth it as yeoman’s work to build an alternative and more accessible pile of knowledge, and as practice for anyone who wants it.
Hi, I’m a postdoc in climate science, just made an account. I’ve been reading SSC off and on for about two years, then started exploring LW more recently and wanted to join the discussion.
I’m curious what questions you have about climate science, and what resources you think are needed to make it more accessible? More blogs? More easily accessible review papers?
Welcome to LessWrong! I appreciate you popping up.
Out of the gate, I should probably say this isn’t really specific to climate science; getting up to date on any rapidly advancing field is pretty tough. The saturation of political offense and defense just makes it tougher, is all. The likelihood functions over p-values question is one that I expect would help all scientific fields more-or-less equally.
The questions I personally have are mostly about the state of the various feedback-loops or runaway-processes that have been proposed as drivers of radical climate change. For example, the clathrate gun hypothesis; the last thing I read on the subject was commentary from a scientist who had just completed sampling of methane releases in the Arctic Ocean, who thought it had already fired. Sometime later I was reading a summary which largely agrees with the Wikipedia article that the role of this mechanism in past events is not as great as we previously thought/feared, but then later still I read that clathrate mining had officially begun. The example of the American natural gas boom suggests to me that mining will probably make the problem worse. There’s plenty of stuff on the various equilibrium processes like the nitrogen cycle or the sulphur cycle; I feel like something on what are effectively dis-equilibrium processes would also be useful both for learning and for risk evaluation.
For the most part, information about this kind of thing is scattered across papers which are infrequently meta-analyzed, or buried deep in reports like the IPCC’s and limited in nuance. I think more easily accessible—and in particular findable—review papers would be very helpful to me. In particular, if papers which discuss the history and intuition of an open problem could be found, I would love that. To give you a sense of what I mean, take a look at Macroscopic Prediction by ET Jaynes. This was a habit of the author more than anything else and it runs throughout his writings; is there anyone like that in climate science?
More blogs is probably a good idea, but I haven’t delved deeply enough to find out if any of the ones which already exist are actually good—chiefly this is because of the political noise flooding my search results. What I would really like to find is someone working in climate science with a blog like Andrew Gelman or Scott Aaronson, who I could rely on to be an expositor of their personal thinking while flagging important developments. Then I could use that blog as the launchpad for my related searches, and be more productive that way.
You might very well be like “check out <person> and <person>” and completely resolve my difficulties, which would be awesome.
Thanks for the response!
It’s interesting to see what your concerns are. There’s probably less research on these kinds of feedback loops/tail risks then there should be. Part of the problem is just how uncertain they are—a combination of the difficulty of measuring things like methane releases and the problem of not being able to resolve these processes in models. Our best guide is probably paleoclimate observations, but I’m not an expert on these.
In terms of blogs, Real Climate is the best place to start. They can get combative, and they aren’t always the most rigorous researchers, but they at least give you a sense of what’s going on. Isaac Held is a giant in climate science and had a very widely-read blog for a few years, but he’s mostly gone quiet since the Trump administration took over (he’s a federal scientist). Going through his posts is a great way of getting caught up on the field. I put up notes here occasionally. And for someone with more of a “denier” bent, Judith Curry is worth checking out (though she clearly wants to push the discussion in specific directions). Finally, if you have access to it, Nature Climate Change publishes a lot of good stuff (with the caveat that it wants to publish high profile work), including summaries and perspectives which give overviews of specific subfields and questions. Any important work on tipping points and feedback loops will be in a Nature journal.
I found that so-called parapsychology research is suffering from the p-value problem badly. In one book I read some thing like that they tossed a coin 1000 times, and it came 520 heads. The probability of 520 heads from 1000 is like p = 0.01, so they concluded that their result is significant. However, it is still inside standard deviation from 500.
This helped me better understand the problem with p-values: even if they got 500 head and 500 tails, it would still have p-value around 1 per cent. But if any psi-effect were true, their result should be outside standard deviation. In other words, whet they should measure was not the probability of the result, but such probability of the result given their hypothesis.
What does “inside standard deviation from 500” mean?
Having a small p-value is exactly the same thing, at least for approximately normally distributed things like this, as being multiple standard deviations away from the norm.
The specific number here is neither “like p=0.01” nor within 1σ of the mean. Variance of a binomial distribution is npq=250, so standard deviation is just under 16. Being at least 20 away from 500 is approximately a p=0.2 event.
When I see people write things like “with unconditional basic income, people would not need to work, but without work their lives would lose meaning”, I wonder whether they considered the following:
There are meaningful things besides work, such as spending time with your friends, or with your family.
Work doesn’t have to be “full-time or nothing”. Working only for one or two days a week, or working full week but only once in a while, would probably not provide enough money to make living, but in some context it can still be useful work and provide the sense of meaning and identity.
Sometimes it is difficult to convert useful work to money. For example, helping poor people overcome poverty seems like a useful thing, but the poor people, almost by definition, would have a problem paying for such service. The obvious objection here is that there would be no poverty with UBI, so instead let’s talk about various dysfunctions that are currently correlated with poverty.
I don’t have an obvious example here, but imagine a kind of work, where it is very difficult to evaluate its impact: it could be very useful, or it could accomplish nothing; and if you offer to pay people for it, you will attract masses of scammers. A person with an independent income could volunteer to do it for free.
But I think the greatest problem is jumping from “most work is not needed” directly to “post-scarcity society with superhuman AI”. We might spend a few decades between these two; and in between, most people would have a problem to secure an income by doing economically useful work, but there will still be many useful things to do that simply wouldn’t pay sufficiently.
Friendship arises from a common stressor and withers without it. Family arises from a need to support each other and kids. If you remove all stressors and all need of support, but assert that friendships and families will continue exactly as they are, you aren’t thinking seriously.
Somewhat reducing the amount of financial stressors is pretty far from removing all stressors and need of support, though.
Yeah, fair enough. I’m just thinking that individualism has already done quite a lot in that direction. We’re much more isolated than people in past societies: many of us barely know our neighbors, can go months without talking to parents, and have few friends outside of work. So if we’re discussing further individualist proposals, like basic income, maybe it’s worth spending some time thinking about the consequences to the social fabric (never thought I’d use that phrase...)
IME, the need to make a sustainable living is a big reason for why people can’t fix those individualist problems. At least in my social circles, there are plenty of people who have all kinds of communal projects and would want to spend time doing meaningful things with people who are important to them… but none of those things bring a living, so instead they have to burn most of their time and energy earning money and have much less left that they could use to build a more communal society.
It’s no wonder that people have few friends outside work when work and family combined leave little time for anything else. NYT on why it’s hard to make friends after 30:
Also Qiaochu Yuan:
A basic income wouldn’t fix all of this, but if it would at least allow people to refuse taking on the kinds of meaningless bullshit jobs that suck your energy dry, then that would help a little. Not being forced to prioritize a job over community would be a great start.
I find the most clarifying thing in these types of discussions is to distinguish between work and employment. No one derives meaning from working at McDonald’s or at Walmart. I have a good job, and I don’t derive any meaning from it, either; it is a strictly mercenary arrangement with the side benefit of being able to learn cool things at the same time.
I would rather have the bandwidth to solve problems in my community. But I don’t and neither does anyone else for the most part, so they limp along below crisis levels.
Be very aware of https://wiki.lesswrong.com/wiki/Typical_mind_fallacy for discussions about what people do or could find meaning in. I know at least a few hourly retail employees who do get some self-worth from helping customers. It’s mixed with drudgery and annoyance, but not completely meaningless.
The good thing about employment is that it’s guaranteed that someone (your employer) thinks you’re providing value to other humans, and there is (outside of government, or government-sized behemoth organizations) a feedback look for that to be a true belief. If you weren’t providing value, you wouldn’t be paid.
That’s not true of unpaid work—it’s still the case that you can do good and provide value to others, but there’s much less feedback about whether and how much.
I predict that there will be no true post-scarcity world. We’ll reduce scarcity, we’ll make many unrewarding and low-paying jobs unnecessary, so that one can likely live a minimum-wage lifestyle without actually working. But we’ll still have a large amount of luxury available only to the lucky and productive (rich), and a larger amount of semi-luxury available only to those who are employed by the rich. In this reduced-scarcity world, those who find meaning in employment can partake, and will enjoy a bit of luxury as (part of) the reward.
I am confident that the percentage of people who work in fast food and retail that derive meaning from it is ~0.
This speaks to the work/employment distinction I raised. Do they suddenly stop helping people when they aren’t on the clock? I very much doubt it; they are probably the same kind of people who are happy to give strangers directions on the street. The meaning likely comes from helping people, not from helping people at their retail job.
This can be contrasted with something like being employed as an EMT or a 911 call operator, where the ability to help people is conditional on the resources and organization that comes with the employment relationship.
But your point also indicates a central challenge that reduced employment would produce: we have invested much of our social expectations and effort in employment. I would go as far as to say a hard majority of contact with other people, excluding family. It would be hard to replace that on short notice, which is to say on the order of years. The costs of social isolation accrue faster than that.
FYI, while I worked at a grocery store as a clerk, janitor and sometimes baker, I...
wrote multiple songs about it (always a bit ironic and self-deprecating but not entirely so, and some functioned as pseudo-anthems).
had a lot of fun talking to customers, and talked to like 1-2 orders of magnitude more of them than I would have talked to random strangers.
got a lot of physical exercise by default.
In my case I also had other longterm goals and aptitudes that made me not want to work there forever, but ever since then I’ve considered it surprisingly hard to beat “have a part time job that involves talking to customers and a lot of physical labor”, as a way to make sure a lot of basic needs are met at once, including certain kinds of tribal resonance.
(Up until starting work at LessWrong, I think I was generally less happy, or approximately as happy, at the programming jobs I worked at. I was also happier at other jobs when I intentionally shifted into “find meaning in the situation” stance)
This is largely because I have a property of “able to make meaning wherever I go”, which has pluses and minuses. I also did art and stuff that provides a lot of meaning, but I think at the time the meaning I got from my grocery clerk job to be… I dunno ranging from 10-60% of the meaning I was getting at a given time, depending on what other projects I had going on.
So if you got the same amount of exercise and talked to the same amount of people as you did working at the grocery store, but you did it not at the grocery store, do you think you would have gotten less meaning, more meaning, or about the same meaning? My expectation is about the same, and if correct that implies the meaning gained from being employed by the grocery store is about zero.
I assert that this:
is a default trait of humans. The meaning almost invariably comes from actions and experiences (like exercise and talking to people). The reason that people attribute meaning to having a job is because every systematic generator of actions and experiences has been converted to an employment relationship, or died. Employment doesn’t cause meaning, in general—it is just the dominant context in which it occurs. I think disentangling these things is important for thinking about this and similar problems (like automation).
Another counterpoint to the McDonald’s/Walmart example I used above might be people who are invested in the status that comes from working someplace prestigious, like Apple or Google, or anyone who is towards the top of their profession. Then people would derive meaning by the mere fact of the employment relationship. But I also think that such meaning is not threatened by making employment optional, because such people will simply choose to remain employed.
Hah! Score one more point for the shower stall as an indispensable writer’s tool.
For the last three days, I’ve known a few vague outlines of some ideas I want to write a story about, but couldn’t come up with anything better than those blurry notions. Today, while I was thinking about them while shampooing my head, I finally identified what I wanted out of them, and in enough focus to combine them into a three-word premise. (Or, come to think of it, a different three words, if I’m allowed to use published authours’ last names.) And now that I’ve done that, I’m extrapolating a host of details and new sub-ideas to play with.
I used an old outline as the basis for IO.SYS, and was starting to wonder if I’d need to look into creativity-workshop stuff to kick my brain back into gear. Now I’ve got my authorial confidence back. :)
(And /now/ all I have to worry about is whether I’ll be able to tell the difference between “not depressed” and “manic-to-hypomanic state”. But that’s a much more tolerable problem than before, so no complaints. ;) )
My SAD light seems to be doing some good; I’ve just finished a first draft of a quickie 6k-word story. Is anyone here willing to give me some private feedback on it before I make it public? If so, let me know and I’ll try to send a private message with a link to the GDoc. (Genres: hard-SF, at least in the general direction of rational, and abstract horror.)
New favorite example of ‘undecidable’:
My imagination is quite good. So when I recently read how people are liable to discount the likelihood of what they find hard to imagine and overestimate the likelihood of what they find easy to imagine, I tried to imagine what I would find hard to imagine… and found this hard to imagine.
Which means I find it easy to imagine something which I find hard to imagine, since I just imagined it! But then it is no longer hard to imagine, so I can’t have found it… But if I haven’t found it, then it’s hard to find; so I can imagine it...
My muses saddled me with this idea for doing subtitles in a different way. I don’t know if it’s ever been tried. I think it might end up being extremely good for language learning.
In short
see rest
For many many reasons I can’t be the one to implement or test this. Wondering if anyone could dismiss it as impractical and relieve me of my burden, or, failing that, reach out to some fansubbing communities and get some fine mapping subtitles rendered and see how they feel.
I’ve seen this done in children’s shows. There’s a song along with subtitles, and an object moves to each written word as it is spoken.
I considered the term “bouncing ball subtitles” yeah, but there are a couple of reasons that animation wouldn’t really work here
Sometimes a word in the voiceover language will share meaning with multiple words in the subtitle language (in which case the ball would have to split into multiple balls), or to parts of words (in which case it might not be clear that the ball is only supposed to be indicating only part of a word, or which part). Also it’s kind of just visually cluttered relative to other options.
I don’t think the research in that area would map either. Children are learning the subtitle language after learning the voiced language, whereas with adults watching subtitled video, they know the subtitled language extremely well.
It would probably work better when the speech is slow, so you have more time to notice which currently pronounced word corresponds to which highlighted word / word part / set of words.
Also, the subtitles would have to be a very literal translation, which I suspect is usually not the case. (At least, if I would make subtitles, I would sacrifice exactness in favor of shortness, because people need to be able to read the text in real time, and shorter is better.)
It doesn’t like, break, when a non-literal translation is used. When the translation doesn’t map directly, this is communicated to the viewer quite clearly as certain words in the VO produce no pulses and certain words in the subtitle fail to pulse at all.
So you don’t have to do a literal translation at all. It sort of imposes a mild pressure towards doing more literal translations; the demographic for fine mapping kinda want them. You don’t have to give it to them all of the time. The most important thing is making sure that they understand what’s being communicated.
Thanks for the explanation, and I agree now that the two are too different to infer much.
A Flash of Colour in the Mind:
Some say to remember that the finger pointing at the moon is not the moon. And some say that every time you call up a memory, you change it. But here’s the best I can express what remains of a split-second of thought earlier today:
I was enjoying reading a classic SF novel for the first time, and as my thoughts went over expanding on an idea from one line, I had a combination of seeing that expansion in the form of some Avatar-like glowing blue text, combined with an odd sensation. It took me some time to nail it down, which was a combination of thinking that the expansion was new-to-me, interesting… and what I now realize was the actual emotional sensation of hope.
I’m not sure if I can describe what it’s like to realize that I’d literally forgotten what hope feels like. I’ve cobbled together an intellectual approximation, so that, as a hyperbolic-to-the-unrealistic-extreme example, I can analyse the pros and cons of suicide, taking into account that I know my mind is prone to certain biases, and come to the logical conclusion that even if I don’t anticipate anything ever getting anything better, staying alive is most likely the better choice. But that’s an entirely different thing than actually /feeling/ “hey, that sounds like something better that just might happen”.
Sure, I’ve now been going over that split-second so many times that by now I mostly only remember remembering it. But I’m still taking it as a /very/ good sign I’m still on an upswing. (Sure, one step back every few steps forward, and there are days as blah as before… but there are days that /aren’t/.)
About the only downside is that re-thinking my latest story idea, I’m now realizing how bleak and depressing my outline is; so I’m going to have to change it so much that I might as well be coming up with something from scratch. Which is such a ridiculously contrived “downside” that I’m grinning lopsidedly to myself as I type this.
Of course, given past experience, I may only be peaking before a return to previous depression; I’ve had such before. But… it may not be. And I’m looking forward to hoping my mental state will improve further.
Try not to cling to the past. Instead appreciate what was, and find what is new.