Players of Battle for Wesnoth often accuse random number generator of being broken, e.g. when their unit has 3 attacks, each of them has independently 70% chance to hit, and all three attacks happen to miss. But the chance of that happening is actually 2,7%, and if a level takes twenty or more turns, and in each turn several units attack, this is likely to happen several times per level.
Viliam
Karma: 11,391
The market starts producing as soon as it suspects there might be panic and shortage, I don’t think that shops running out are actually needed for industries getting the message.
There is also uncertainty, and the producers don’t want to oversupply. Starting the panic “collapses” the uncertainty. If some families are going to buy 3 years worth of disinfectants, I want the market to know this, not as a possibility, but as a fact. So that the result is that some families have 3 years worth of disinfectant, and the remaining families have enough.
I agree with the political backlash. Doing the right thing on object level may be a mistake on political level. When you do something, you become responsible for everything that happens, and in situation of pandemic, all outcomes are bad, so the optimal political strategy is probably to stay in background until the situation gets really bad, and then come and save the day. One does not get political points for preventing problems. (After Slovakia successfully defeated covid during spring, many people concluded that it was just a hoax and that all measures were unnecessary. This probably contributed to the reluctance of government to do anything when the numbers started growing exponentially in autumn.)
Depending on the cause of sleep apnea, a removable wire in mouth may help. Feel free to ask me if you have any questions; I got sleep apnea from being fat and having a narrow throat, and this helps. I like that it’s purely mechanical, no electric power needed.
I attributed the Slovak reaction to cultural experience from socialism. Like, when the authorities told you: “comrades, do no panic, there is no shortage of toilet paper”, you knew you had to run immediately to the nearest shop and stock up, because in the evening it will already be sold out. So when the authorities told us not to buy face masks… :D
People also stocked up with disinfectants. (I don’t remember whether authorities mentioned these, or it was just common sense.) This seemed more tricky, because making disinfectants at home… well, you could burn some strong alcohol, you wouldn’t even have to worry about toxicity if you do not intend to drink it; and some people actually did this (I think there was a guy in Czechia who started mass-producing alcohol-as-disinfectant, got into legal trouble, there was a public outrage and he was pardoned)… anyway, what happened was that shelves were empty for two or three weeks, and then the shops resupplied.
Which again makes me think that if there is a risk of panic and shortage, you might want it to happen sooner rather than later, so that the market has enough time to adapt before the worst happens. As a government, you could even contribute to the shortage, by buying tons of stuff… and later redistributing it to the places of greatest need: sell it to hospitals for the original price, thus shielding them from shortage and price hikes.
I remember clearly I had concluded from the start that masks had to help reduce the spread of the virus because they would reduce how far you’d breathe.
My reasoning was: people in Asia seem to have more experience with this type of situation, and they wear masks; case closed. Also, this song (note: February 2020; English version made in April).
But yes, common sense also seemed on the side of wearing the masks. Like, maybe they won’t filter all viruses, but at least they should filter some. I guess you needed a “gears model” of infection, so you knew it was not “one particle or millions of particles, you are doomed either way” but rather “fewer particles = smaller chance of infection (and smaller expected damage)”. This probably wasn’t obvious to most people; I have doctors in my family so I knew.
Thinking through it now, my guess is that instructions on how to make a mask at home or what to use as a quick fix would have likely contained the pandemic more and prevented more deaths.
Definitely. This is one of the things that public television is supposed to exist for, isn’t it?
this team was part of a bigger organization and platform where multiple teams had to work together to something done, e.g. agree on interfaces with other teams. And in the absence of clear joint goals that didn’t happen.
I have seen this happen also in a small team. Two or three guys started building each his own part independently, then it turned out those parts could not be put together; each of them insisted that others change their code to fit his API, and refused to make the smallest change in his API. It became a status fight that took a few days. (I don’t remember how it was resolved.)
In another company, there was a department that took care of everyone’s servers. Our test server crashed almost every day and had to be restarted manually; we had to file a ticket and wait (if it was after 4PM, the server was restarted only the next morning) because we did not have the permission to reset the server ourselves. It was driving us crazy; we had a dedicated team of testers, and half of the time they were just waiting for the server to be restarted; then the week before delivery we all worked overtime… that is, until the moment the server crashed again, then we filed the ticket and went home. We begged our manager to let us pool two hundred bucks and buy a notebook that we could turn into an alternative testing environment under our control, but of course that would be completely against company policy. Their manager refused to do anything about it; from their perspective, it meant they had every day one support ticket successfully closed by merely clicking a button; wonderful metric! From the perspective of our manager’s manager, it was a word against a word, one word coming from the team with great metrics and therefore more trustworthy. (The situation never got solved, as far as I know.)
...I should probably write a book one day. Except that no one would ever hire me afterwards. So maybe after I get retired...
So, yes, there are situations that require to be solved by greater power. In long term it might even make sense to fire a few people, but the problem is that these often seem to be the most productive ones, because other people are slowed down by the problems they cause.
Where did the project you did implement in your first case come from? Somebody figure out that it would provide value to the company. Otherwise, you might have built a beautiful project that didn’t actually provide value. I think we agree that the company you worked in did have some management that provided value (I hope it was no moral maze).
Yeah, but we have two different meanings of the word “management” here. Someone who decides which project to do—this is useful and necessary. Or someone who interrupts you every day while you are trying to work on that project—I can imagine that in some teams this may also be necessary, but arguably then your problem is the team you have (at least some parts of it). Motte and bailey of management, sort of.
From epistemic perspective, I guess the problem is that if you keep micro-managing people all the time, you can never learn whether your activity actually adds or removes value, simply because there is nothing to compare to. (I guess the usual null hypothesis is “nobody ever does anything”, which of course make any management seem useful; but is it true?) Looking at the incentives and power relations, the employee at the bottom doesn’t have an opportunity to prove they could work just as well without the micro-management, and the manager doesn’t have an incentive to allow the experiment. There is also the “heads I win, tail you lose” aspect where bad employee performance is interpreted as necessity of more management, but good employee performance is interpreted as good management, so either way management is perceived as needed.
This is really not a secret art. The principles are well-known (for a funny summary see e.g. Leadersheep).
Yep. That’s a very good summary. Heh, I fail hard at step 1 (creating, or rather communicating a strong vision).
But it turns out that building a big organization is hard. Politics is real and professional management is still mostly a craft. It rarely approaches something you can call engineering much less hard science.
Seems analogical to social sciences: in theory, they are much more difficult than math or physics, so it would make sense if smarter people studied them; in practice, it’s the other way round, because if something is too difficult to do properly, it becomes easy to bullshit your way to the top, and intelligent people switch to something where being intelligent gives you a clear comparative advantage.
Good luck to you! I suppose your chances will depend on how much autonomy you get; it is hard to do things right, if the sources of problem are beyond your control. However, if you become a great manager and your people will like you, perhaps in the future you can start your own company and give them a call whether they would like to work for you again.
Thank you for the comment!
Specifically, I strongly agree that flagging is an underused tool in debate moderation.
if people knew masks would have made them safer from day one then there would have been more deaths, not fewer, because masks would have not been available where they’d have saved most lives.
In Slovakia in March 2020, people spontaneously started wearing masks in public despite authorities telling them not to. People started making masks at home, and selling them to their neighbors. A week after the pandemic became a common knowledge here, my wife bought the first dozen masks made of cloth from some lady in neighborhood who advertised them on Facebook. The authorities then changed their minds, and masks became mandatory. So we were fully masked already in March, and there was no shortage. On internet I found a project where volunteers with 3D printers shared recipes for transparent face shields, and donated the printed ones to hospitals. Between March and July, less than 30 people out of 5 500 000 died of covid. (Then we completely screwed up in autumn.)
It was quite surreal watching the rest of internet argue how nothing can be done, and our only choices are letting the sick and elderly die, or destroy the entire economy, because a mask that isn’t N95 looted from a hospital is certainly useless (and somehow there is no way to ramp up their production).
So, I remain firmly convinced that discouraging people from wearing masks caused deaths. In short term, by making the pandemic spread faster. In long term, by undermining public trusts in experts.
If some potentially powerful ideas were suppressed successfully, we wouldn’t know about them, would we? Seeing only the cases where censorship failed, we may underestimate its impact.
Second, ideas can be killed without being completely erased. For example, you can read on Wikipedia about the Cathar religion, but how many active Cathars do you know? The crusade was a success.
Third, driving an idea into underground can harm its development. If posts are getting deleted, you can’t have the three levels of response (can’t find a link now, but the idea is that quality debate requires at least: someone stating a hypothesis, an opponent refuting it, the original author explaining why the refutation was wrong, and the opponent explaining why the author’s objection was wrong). If adherents can’t build their reputations, it is hard to distinguish in a debate between actual believers, trolls, and false flag operations; there is no common knowledge of what the idea actually is (so everyone is free to replace it with a strawman).
I admit I am worried about the slippery slope, which is often considered to be a bad argument. The exceptional treatment, you first try with the worst scumbags, because there most people will support you. Later, by analogy and precedent, you proceed towards lesser scumbags, and then finally it becomes normal for ordinary people. At least this is how the story goes; I don’t know if this is empirically true.
I also worry that by the coordinated removal of Trump, tech monopolies sent a strong signal to all politicians, that they are capable and willing to do it again, if necessary. (Kinda like by throwing the first nuke you send the “I have nukes and I am not afraid to use them” signal to the whole world, and no one is going to forget that ever again.) Will it have a chilling effect e.g. on politicians who were toying with the idea of applying the antitrust law to these companies? Would you as a politician pick a fight with someone who can remove you from the sight of your voters literally overnight? Until now, you could have reasoned “well, if platform X removed me, I would certainly tell my voters about that on platform Y, some of them would as a result move from X to Y, so it would not be in X’s financial interest, yay free market and competition”, but now you know that if such thing happened, it would be coordinated across platforms… and the few platforms who wouldn’t join, would in turn find themselves in trouble. (And that’s just the overt bans. What about coordinated shadowbans? Coordinated removal of positive mentions, and promoting of negative mentions on the front pages?) Influence is not just about what you do, but also about what the others don’t do because they are afraid of what you might do in return.
I guess what I am trying to say is that from a larger perspective, I see this as a shift of power: away from elected politicians, towards colluding social media monopolies. People applaud, because the first nuke landed on Trump. Okay, cool. Who is next? And next? Who decides? Quis custodiet ipsos custodes?
“Decision algorithm” was here my metaphor for humans. Like a rulebook for human censors, who themselves are smart and educated, although not extremely.
In a hypothetical situation where a government would officially appoint censors, or if the social networks would decide to optimize for something other than low costs. In a hypothetical situation, where this would be relatively high-status job… I mean, you’d have the power to shape the public discourse, and that is no small thing; give it a decent salary and many university-educated people would compete for it. But you wouldn’t hire the smartest ones, because they have a better use of their skills; you wouldn’t want to hire people with contrarian ideas; and to censor the overwhelming amounts of text online, you would need to hire lot of people, so you couldn’t afford to be too picky even if your budget was unlimited.
The average censor, if such job existed today, would realistically be some bored bureaucrat, who doesn’t give a fuck about ideas, and just applies rules mechanically, trying to cover his ass. But an idealized censor would be a smart and educated person, passionately opposing pseudoscience… kinda like I imagine the people who write on RationalWiki, except maybe less politically mindkilled. And with a button that would allow them to delete content anywhere online. And I wondered what would happen as a consequence.
I agree both with you and jimrandomh. I originally started writing this as a shortform, but then I got carried away and it became too long for shortform, so I posted it as an article. I agree that this doesn’t belong to the frontpage, and perhaps to LW in general, regardless of how much karma it may get (although the karma of course makes me happy, as a reflection of the fact that my text made someone else happy).
The outgroup in my opinion completely deserves the sneer, for reasons explained in Matthew 26:52. LW is not the right place to put it. I don’t have a personal blog currently, and this seemed worth writing somewhere. Apologies to everyone who feels bad about it; I will not do this often.
I was a team leader twice. The first time it happened by accident. There was a team leader, three developers (me one of them), and a small project was specified. On the first day, something very urgent happened (I don’t remember what), the supposed leader was re-assigned to something else, and we three were left without supervision for unspecified time period. Being the oldest and most experienced person in the room, I took initiative and asked: “so, guys, as I see it, we use an existing database, so what needs to be done is: back-end code, front-end code, and some stylesheets; anyone has a preference which part he would like to do?” And luckily, each of us wanted to do a different part. So the work was split, we agreed on mutual interfaces, and everyone did his part. It was nice and relaxed environment: everyone working alone at their own speed, debating work only as needed, and having some friendly work-unrelated chat during breaks.
In three months we had the project completed; everyone was surprised. The company management assumed that we will only “warm up” during those three months, and when the original leader returns, he will lead us to the glorious results. (In a parallel Everett branch, where he returned shortly before we finished the product, I wonder whether he got a bonus and promotion.) Then everything returned to normal: more micromanagement, lower productivity, people burned out.
The second time, we were a small group working together for some time already. Then our manager quit. No one knew who would get the role next, and in an attempt to deflect a possible danger, I volunteered to do it on top of my usual work. What happened was that everyone worked exactly the same as they did before, only without the interruptions and extra stress caused by management, and I got some extra paperwork which I gradually reduced to minimum. The work progressed so well—no problems, no complaints from users, the few tickets we got almost always turned out to be a problem outside our project—that higher management concluded that there is apparently too litle work to do on our project, so the team members were assigned to also work on extra projects in parallel.
Perhaps my short experience is not representative, but it suggests that a manager, merely by not existing, could already create a top-decile work environment in terms of both work satisfaction and productivity. The recommended mantra to recite every day is: “first, do no harm”. My experience also suggests that this approach will ultimately get punished, despite the increased productivity: the expected outcome is more work for no pay raise until you break, or just being told to return to the old ways without any explanation why. I assume I am missing some crucial maze-navigating skills; for someone trying to be a professional manager this would be fatal; luckily I do not have this ambition.
It is quite possible that this approach only works when there is a good team: in both cases I worked with people who were nice above average. If you had a dominant asshole in the team, this could easily become a disaster: the power vacuum left by a passive manager would simply be replaced by an ambitious alpha male, who would probably soon be promoted into the role of formal leader. So perhaps the companies play it safe by using a widely applicable strategy that happens to be inferior in the case of good employees who also happen to be good people; quite likely this is because the companies simply cannot recognize such people.
Is there a leadership level beyond this? Sure, but in my quarter century of career I have only met such manager once. What he did was basically meeting each of his people once a day in the morning (this was long before I heard about “daily standups” and such) and talking with him for 5 or 10 minutes; with each team member separately, in the manager’s room. He asked the usual questions “what did you do yesterday?”, “what is your plan for today?”, “are there any obstacles to your work?”, but there was zero judgment, even if you said things like “yesterday I had a really bad day, I tried some things but at the end it was wrong and I had to throw it all away, so today I am starting from scratch again”; essentially he treated you like an adult person and assumed that whatever you did, there was a good reason for that. Before and after the report, a very short small talk; it helped that he was extremely intelligent and charismatic, so for many people this was the best part of the day. Also, the obstacles in work that you mentioned, he actually did something about them during the day, and always reported the outcome to you the next morning. Shortly, for the first and the last time so far in my career, I had a regular feeling that someone listens to me and cares about what I do (as opposed to just whipping me to run faster in order to meet arbitrary deadlines, randomly interrupting me for no good reason, second-guessing my expert opinion, etc.).
So yes, there is a level beyond “not doing harm” and it is called “actually motivating and helping”, but I guess most managers dramatically overestimate their ability to do it… and when they try regardless, and ignore the feedback, they actively do harm.
RationalWiki on face masks
I guess we agree that limited processing capacity means that interfering with the work of your underlings—assuming they are competent and spending enough of their processing capacity on their tasks—is probably a bad move. It means taking the decision away from the person who spends 8 hours a day thinking about the problem, and assigning it to a person who spent 30 seconds matching the situation to the nearest cliche, because that’s all they had time for between the meetings.
This might work if the person is such a great expert that their 30 seconds are still extremely valuable. That certainly is possible; someone with lots of experience might immediately recognize a frequently-made mistake. It is also is the kind of assumption that Dunning and Kruger would enjoy researching.
I might be wrong and it works less and less the further up you go
That would make sense. When you are a lowest-level manager, if you stop interfering, it allows the people at the bottom to focus on their object-level tasks. But if you are a higher-level manager, how you interact with the managers below you does not have a direct impact on the people at the bottom. Maybe you manage your underlings less, and they copy your example and give more freedom to the people at the bottom… or maybe you just gave them more time to interfere.
So sometimes the smart move is to refuse promotion
So you have more time to scheme… but you have to stay low in the pyramid. Not sure what you scheme about then. (Trying to get to the top in one huge jump? Sounds unlikely.)
Yeah. How long I can work depends on how crappy the work is. This could also explain why CEOs and similar are happy to spend long hours at job… they probably have way more power over their working conditions than I do. Pretty sure they have a room where they can close the door for a moment.
If there is only one thing you would reliably do, it probably makes sense to establish a regular time to reflect on things. Not sure what is the optimal amount. One hour a week? Maybe make it a walk in nature/park, with pen and paper, so that you are not tempted to cut it short or start doing something else. Don’t use this time to force yourself to do anything. (Don’t do anything that would make you hate this ritual.) Just think, observe, and dream.
Establish good habits. (This sounds like a contradiction: “habits” vs “agenty”. The idea is that you are agenty about choosing and updating your habits, but you use the habits as the raw power to achieve the goals. For example, if the goal is writing a novel, the agenty thing is to establish the habits of regularly reading and writing, and reflecting on progress.) Regularity beats volume. If you can make yourself do one push-up consistently every morning, it is relatively simple to increase the amount later.
For inspiration, talk to other smart and agenty people. (May be difficult to find them. Perhaps I am prejudiced but I believe that “read/watch inspiring blogs/videos” is a wrong answer: famous bloggers will optimize for making you an addicted reader and potential customer, not for making you more agenty. Look for a relationship where you talk with someone as a friend, not where someone is trying to sell you things.) This gives you inspiration and peer pressure. Overcome your ego: people who are ahead of you are the best to learn from, but it may make you feel like a loser. (Maybe too much ahead is not good, because they would be too difficult to copy.)
Perhaps try meditation; one of the first achievements is the ability to turn off unwanted distractive thoughts when they come. Then, whenever you face an unpleasant task, precommit to work on it seriously for five minutes, and use the skill to turn off distracting thoughts during that time period. (You may be surprised to find out that it is actually quite easy to continue doing the task when the five minutes are over. Yes! But don’t push yourself. Precommit to the five minutes only; everything afterwards is optional.)
Read Don’t Shoot The Dog by Karen Pryor to get a “gears model” how conditioning works. (Sometimes things that work for some people don’t work for others, but for me this was the best book on the topic, and it explained so many important things, including some common mistakes people make. If there is one book people would ever choose to read based on my recommendation, it is this one.)
Most people are naturally pro-social. (No, this can’t be applied to AI.) Given a task, they will try to do it well, especially if they feel like their results are noticed and appreciated.
A cynical hypothesis is that most of the things managers do are actively harmful to the project; they are interfering with the employees trying to do their work. The less the manager does, the better the chances of the project. “Delegation” is simply when manager stops actively hurting the project and allows others to do their best.
The reason for this is that most of the time, there is no actually useful work for the manager. The sane thing would be to simply sit down and relax, and wait for another opportunity for useful intervention to arise. Unfortunately, this is not an option, because doing this would most likely get the manager fired. Therefore managers create bullshit work for themselves. Unfortunately, by the nature of their work, this implies creating bullshit work for others. In addition to this, we have the corrupted human hardware, with some managers enjoying power trips and/or believing they know everything better than people below them in the hierarchy.
When you create a manager role in your company, it easily becomes a lost purpose after the original problems are solved but the manager wants to keep their job.
I have no hard data to prove that my rational thinking is improving, so maybe it isn’t. But here are the things I am trying recently:
Reduce the social media: I almost stopped reading Reddit, and there are days when I don’t use Facebook. Now my major distraction is Hacker News, but I try to reduce that, too.
I downloaded and read a few textbooks, mostly on math. The idea is that if I am curious about something, e.g. set theory, it is more efficient to get the fundamentals right first, and only then proceed to more controversial parts. (Here, the rational thing is not the information I get from the textbook per se, but rather the habit to look at the textbook first.)
I tried to meditate again… but this was only a week or two ago, so no results to report yet. (The expected benefit is to be able to focus better and reduce distracting thoughts.)
Hi Amir, I believe we have two different topics here. Comparative advantage in traditional sense only cares about what is good choice right now; it does not include the considerations of future growth in skills. It kinda says, if you are 3× worse at X but only 2× worse at Y, don’t despair; you should produce Y and trade it for X, and this will be beneficial to both sides.
Actually, the considerations of future growth can be used as an argument against following the comparative advantage blindly. Like, maybe the reason why others are so much better at X than at Y is because “Y doesn’t have a future”. Perhaps practicing X improves your X skills significantly, but practicing Y doesn’t improve your Y skills at all. (X = technology, when you do it, you invent new methods, build new machines, train your employees; Y = agriculture, doing the same thing over and over again; of course this is oversimplification.) In which case perhaps you should sacrifice the short-term gains from possible trade and keep doing X anyway, until you level up enough that X becomes your comparative advantage.
But the full analysis would be quite complicated here, because we would need to know the learning curves for X and Y, your optimal strategy would depend on what other players would do… and if we are assuming a dynamic world, who knows, tomorrow a new Z might enter the game and completely change the situation. (Like, if the reason for your disadvantage is simply lack of practice, e.g. you are younger than other players, it might make sense for you to immediately switch to the new Z where everyone starts from zero.)
I wish I had some simple personal example of “not using comparative advantage because of strategic considerations”, but nothing specific comes to my mind right now.
Seems to me that there must be more about pain and pleasure than mere −1 and +1 signals, because there are multiple methods how to make some behavior more or less likely. Pain and pleasure is one such option, habits are another option, unconscious biases yet another. Each of them make some behavior more likely and some other behavior less likely, but feel quite differently from inside. Compared to habits and unconscious biases, pain and pleasure have some extra quality because of how they are implemented in our bodies.
The simple RL agents, unless they have the specific circuits to feel pain and pleasure, are in my opinion more analogical to the habits or unconscious biases.
I don’t understand the article enough to decode what “the right kind of state” means, but this feels like circular explanation. The three-dimentional space can “emerge” from a graph, but only assuming it is the right kind of graph. Okay, so what caused the graph to be exactly the kind of graph that generates a three-dimensional space?