Use and tags to enclose your code to get indentation (you can then just write the indentation with spaces).
Also, you're right that I didn't conform exactly to the specifications; I probably should have added a "return list;". But when not doing functional programming, there's often no reason to bother.
Lambda functions in Python don't really have any advantages over inner classes in Java except that the syntax is a little shorter and Java's standard libraries don't use the model enough.
I also think you need to have more complex examples to see the power of closures or lambda functions.
For example, int n = 0; for (Integer i : list) n += i*i;
is shorter than the closure code to do the same thing (and is easier on one's working memory). Likewise for void AddL(List<AtomicInteger> list,int increment) {
for (AtomicInteger i : list) i.set( i.get()+increment ); }
And these would be prettier still in a weakly typed language, or a bit prettier if Java would assume sensible defaults.
Often, languages that have lambda functions need them because they have low-level types that are accessed by value (or other immutable types). For example, in Python, numbers = [1,2,3,4,5] for i in numbers: i = i+5
does not do what you want because i is not accessed by reference. So you write numbers = [1,2,3,4,5] numbers = map(lambda i:i+5,numbers)
and feel all pleased, except that you only needed to do that because "for" only refers to objects, not base types, by reference.
Lambda calculus is very powerful, but I've not seen any simple examples that actually demonstrated the power beyond what you could achieve with standard programming constructs. Closures are nice for things like sorting, UI events, and such--but inner classes are even nicer because they can carry data too in a more obvious way (instead of simply by scoping rules).
All right, I think we are done, because this post was pretty unhelpful for me--and when it comes to posting on the internet, I am going to be selfish to a large degree. But I'll throw out a few more comments anyway.
(3) Selfless. Acting to improve the welfare of others without regard for my own welfare.
This does not exist. Whatever the motivation, we choose our actions. And thus we choose our end-goals.
If one's goal is to "make sure that Sally has the best life possible, whatever the consequences to me", then it's a selfless choice. I'm not saying that the person isn't making the decision, and they may have reasons/feelings for doing so, but if they're basically leaving themselves out of the equation, that is sensibly called selfless.
Defining an attempt to achieve ones' end goals as selfish (even though it may benefit others) makes this (3) an impossibility.
I don't think this is a useful definition--then everyone would be selfish except for people with such major cognitive deficits as to be unable to form or pursue goals.
The interesting distinction regards the nature of those goals, which is why my definition included the "without regard for the welfare of others" caveat.
The only axiom of ethics that I put forward is "do not do to others what you wouldn't have them do to you".
What is your justification for adopting this axiom? Note, for example, that under your ethical system, it's perfectly okay to leave someone bleeding to death to die when you could easily save them.
If one wants to elicit an action from anyone else, one must do so through an agreement.
That's a little difficult for the bleeding person if they're already unconscious.
Anyone who claims to care to much for me is assumed to be attempting to compel me to act without paying for it (ie, without eliciting my consent).
This is generally a good guess when dealing with salespeople. It's a rather weird way to look at things when the "anyone" is, say, a parent or spouse.
Reasonable behavior is a choice and (as such) it is a conscious one. It is not always a psychological need. Given the restrictions put forward above, it will produce a better society.
That doesn't mean that it's attainable in practice. Communism didn't fail because it was a bad idea in principle--it failed because it was unattainable in practice because humans don't act like Marx wanted them to. Expecting reasonable behavior without emotional support seems, well, unreasonable.
And having a "civil" society (a society in which it is not acceptable to take proactive steps to hurt each other) is a worthwhile goal. Confusing it with a society in which people are compelled to take care of the people they don't care to take care of will quickly deteriorate into a society of caprices and whims and will erase civility.
Indeed. Sweden is so uncivil. Japan, too.
All the "community-oriented" thinking produces is xenophobia and lack of privacy.
Indeed, which is why so many people say they feel so good after volunteering.
You might want to pay more attention to reality before making such sweeping statements.
With one reminder, once personal freedom is taken away, a group of people (acting in the name of a "greater good") becomes a tyrannical entity... an entity that rules over adults as parents do over children. Taking up arms against such an entity is not immoral.
Are children justified in taking up arms against parents? Why or why not?
The degree to which a violent revolution is justified is largely controlled by the degree to which the personal freedoms have been infringed.
I largely agree, but there is a weighting factor that I apply to various different aspects of freedom--some are more important than others.
But this isn't terribly relevant to whether one shou
But if one only looks out for one's own interests, then having this weird rule about "don't use other people" is suboptimal--it is in effect a hard-wired unselfish motive dropped into the middle of a selfish framework. There's nothing wrong with that, but at least recognize it for what it is!
Ok, sure, but unselfishness, in this stretched definition, is defined negatively. It's sort of equivalent to defining "apathy" as "care" because it is characterized by lack of hate.
Not at all. Here are the three definitions I'm using: (1) Selfish. Acting to improve my welfare, and ignoring how my actions affect others (except inasmuch as that will then affect me). (2) Unselfish. Acting to improve the overall welfare of some group (or with regard to many groups), of which I may be a part. (3) Selfless. Acting to improve the welfare of others without regard for my own welfare.
Both (2) and (3) are forms of altruism; I am more concerned with (2) than (3).
But if one only looks out for one's own interests, then having this weird rule about "don't use other people" is suboptimal
No, it isn't. You forgot the "renormalize" step at the end of each iteration.
If you could actually do this reliably, it would be a good idea. I will explain below why in practice one cannot. Your analysis isn't far off in certain ideal cases, but it fails to cover all commonly encountered cases.
But the principles you cited do not justify your choice of honesty as a virtue.
I think the above [now removed] should explain it.
It explains it in certain sets of interpersonal relationships. There are piles of exceptions. Here are three.
(1) Relationships that can be presumed to be one-time-only. For example, if you see someone drop their wallet and not notice, the selfish thing to do is wait until they're out of sight and take all the cash out of the wallet. The unselfish thing to do is to tell them they've dropped their wallet. These type of win/lose situations carry essentially no cost to generate, do not (if done halfway carefully) affect your reputation, and benefit you. They often hurt the social system overall, though, as people have to spend lots of extra effort locking up all their stuff, replacing all their credit cards when their wallets disappear for a bit (and appear without cash), and so on.
(2) Relationships which are not easily broken once made. A typical example of this is vendor lock-in: you engage in a mutually beneficial relationship to the point where the other party is dependent on you but you are not dependent on them. You then shift the relationship from a win/win situation to a win/lose situation where they accept the loss because breaking the relationship would entail taking an intolerably high loss.
(3) Relationships which are unavoidable. Chances are that you only have one set of power lines into your house. In fact, chances are that there are only one set of power lines within any distance that you can reasonably cover. Assuming that you like electricity, you have very little walk-away power when dealing with your electricity provider. Likewise with cable, telephone, etc.; there is a tiny bit of cross-monopoly competition, but basically, if one has a monopoly over a very important resource, one can impose huge win/tiny win conditions on everyone. Some resources naturally monopolize due to physical constraints (e.g. roadways).
Do you stop if it's win/lose, with you doing the winning? If so, why?
I am pretty sure I explained that above... Mostly because it's not worth the time. Probably, because it also raises the cost for starting new win/win games (you get a "reputation").
Reputation is essentially a societal metric for measuring the fairness (or other qualities) of your interactions. Effectively, it's a "do you
If interactions are fully optional, I agree that you can turn any negative interaction into a zero interaction on the next trial.
But using them leaves you an empty shell of a person (even though you get what you need). I don't understand the justification, at least not if it's independent of unselfishness. I can understand why if one values other people that using them has an emotional cost associated with it. But if one only looks out for one's own interests, then having this weird rule about "don't use other people" is suboptimal--it is in effect a hard-wired unselfish motive dropped into the middle of a selfish framework. There's nothing wrong with that, but at least recognize it for what it is!
(Again, note the selfish / unselfish / selfless distinction that I'm making here.)
Better than what? Isn't it suboptimal--"destructive"--to not cheat when you have the opportunity to do so and the expected costs are less than the expected reward? Why all the language about esteem and desperation? Now you are mixing your paradigms. In the one I presented, "constructive" solution arises from selfish behavior moderated by honesty. But the principles you cited do not justify your choice of honesty as a virtue.
In the one you presented, optimal is defined as the one maximizing outcome of the PD. Your axiomatize "optimal". I axiomatize "ethical". You were the one who was using optimality as the argument against unselfishness. If you want to switch to ethical, I think unselfishness does pretty well there, too.
My argument is that we have an internal sense of discomfort at being dishonest. Your argument is that we have an internal sense of discomfort at being insensitive. Do you not think it is an empirical observation that for some people both are true, for some people only one is true, and for some people neither seems to be true, and that these things are affected somewhat by development?
At that point, the question is not whether these things exist, but whether it is useful to encourage or discourage them.
Mine is based on using one's internal judgement. Yours comes from judging oneself through the eyes of others (you compare yourself to others to get a sense of "how you are doing" in term of resources gained). If you're in a social environment, it's generally worthwhile to pay some attention to how you are interacting with the group, since otherwise you won't be able to predict how others will react to you. For example, you won't know whether they think you're dishonest.
My paradigm excludes possibility of the win/loss solution. And as soon as the solution becomes loss/win, it stops the game. As soon as dishonesty occurs, I don't to play anymore. That's win/win or no deal. Do you stop if it's win/lose, with you doing the winning? If so, why?
PD doesn't allow for this scenario -- it insists that one keep playing even after loss/win in order to maximize one's gain. Not all interactions are optional if you require resources from others for survival, or if the situation otherwise does not allow withdrawal at no cost. There are many cases where your strategy works; I'm concerned about those and those where it doesn't.
Exhausting where? Not to the photoreceptor cells. Do you mean to bipolar cells? Ganglion cells? What is the evidence that retinal fatigue, as opposed to processes in the LGN, V1, superior colliculus, etc., has anything to do with subjective fatigue?
Greed for power is also a problem, and it's also a case of not being unselfish, and it's not something that monetized interactions particularly help with. Since you seemed to be favoring monetized selfish interactions, I focused more on the problems of financial selfishness.
I'm not assuming greed is bad. I've been giving examples where greed leads to suboptimal outcomes for everyone involved (PD-style). All you've managed to say is something like, "Nuh-uh, it doesn't work that way," with much explanation but mostly by bringing up specific cases where selfishness is okay, not addressing the central problem. I've never claimed that selfishness is always bad, just that unselfishness is superior in some situations (not necessarily for the one being unselfish, but for everyone overall).
I'm familiar with objectivism and Marxism/communism. They're both overly naive, but good to know about if one hasn't yet encountered the ideas. Marxism is particularly naive about human motivations (having fallen entirely victim to the "blank slate" theory of human psychology); Objectivism is also naive about human motivations (having fallen victim to the "rational actor" theory of human psychology) and doesn't quite grasp the consequences of its own premises either as applied to humans or to ideal rational actors.
Both the Communist Manifesto and Atlas Shrugged (or some other synthesis of Rand's ideas) are worth reading, but one has to constantly ask, "Really?" when evaluating the claims and arguments therein.
Better than what? Isn't it suboptimal--"destructive"--to not cheat when you have the opportunity to do so and the expected costs are less than the expected reward? Why all the language about esteem and desperation?
Or have you just noticed a case where unselfishness is a good policy because of PD-type considerations?
First, what is more tiring, some glow, when most of the retina remains inactive picking 'dark', or a full blast from a CRT tube against your eyes? Actually, your photoreceptor cells are fully active in the dark, not the light. The rest of the retina is about half light-responders and half dark-responders and light-responders (and a good number of those only respond to changes in the favored direction).
So your premise is wrong. The retina is highly active even in the dark.
Free society (and that means people personally invested in treating each other fairly rather than nicely) is a better world than a communist(capitalization intentional) society. It encourages cooperation by making it unavoidable (due to efficiencies of specialization). Fairly? Wouldn't it make more sense to treat people as sources of potential benefit to you, and act accordingly to how that benefit plays out? Why be fair, unless you're going to get caught and being caught is worse than being fair?
At this point, I think that all I can say is that you have a really odd view of the world. You might want to think more about the nature of human motivations and social interactions, and also about less drastic ways of curbing human abuses than monetizing all interactions (not that this really helps, as Kerviel's actions at Societe Generale attest).
You've made a number of additional points, but you seem to have so poorly grasped the essence of the previous points that I don't expect it would be fruitful to continue the discussion. You're quite good at explaining your position and reasoning through it, but have displayed almost no imagination when it comes to considering any other hypotheses. Thus, your objections to my points of view have been largely lacking in insight. For example, even if nepotism were this horrible evil that we had to combat at every opportunity, a solution would be to have additional transparency and third-party reviews of performance. That is, pit public shame against greed expressed through social ties. Fringe benefit: that works for dampening individual greed, too.
You're right that there's a second system there, but I'm not introducing it. That's the original system. We introduced a monetary system on top of our social interaction system when we started living in such large groups that it became too difficult to keep track of personal relationships.
We have various instincts regarding social exchange that make it a superior system in some contexts--most notably that we tend to get stronger feelings that we should act a certain way, and that motivates us; when that way is a way that solves PD problems, it's a great benefit.
When it's a way that creates more problems than it solves (if any), it's better to use money.
I'm not saying not to use money; it's just that we have the other system and may as well use it for what it's good for.
I agree that businesses are in danger of nepotism, but that's a danger regardless of the system used. The problem with nepotism is that favoritism leads to bad decisions that hurt the overall performance of the business as people selfishly siphon off resources to themselves and those they want to benefit. If people weren't selfish (in a broader sense which includes "cronies"), nepotism would be a danger. One way to solve the problem is to remove any illusion of unselfishness, so everyone knows to watch everyone else closely. Another way is to collect people who are less selfish.
I'm afraid the "those people are an extension of myself" argument is not very helpful. Why not extend that same feeling, at a weaker level, to others who you're less desperately and unavoidably in love with? Besides, whether you can help it or not, if it's destructive it's still not something a perfectionist would want to accept.
Something is constructive if it alleviates the harshness of mother nature (and thus promotes pleasantness of life) and destructive if it does the opposite.
Okay, we agree on the standard.
Under this standard (which you are at least tempted to consider right now), exchanging lesser value for higher value is destructive.
Fair enough.
Because the assumption has to be made that those who create value have value to exchange and those who consume it (and don't create) require sacrifice.
If by "sacrifice" you mean "effort expended on someone other than myself", I suppose the assumption is somewhat relevant. But if the overall situation is such that the value of the help being received to the person receiving it is greater than the lost value to you, then overall the interaction is not destructive. It's just potentially destructive-to-you if you can't benefit from such interactions in the opposite direction often enough.
Being an unselfish person in a selfish environment is a bad idea.
Extroverts would be the people that seek interaction. A human being creates value by thinking (doing physical work is also result of some thought process, so, yes, even a human being performing manual labor). What is established during interaction with other human beings is a good will (a feeling of family, comradery, tribalism, etc) that allows for an honest exchange at a future time. But the value which can be exchanged (the one that results from tranformation of pristine natural things into human-usable things) must still be created by individuals alone with their perception. Perfectionism must necessitate introverted effort of keeping ones concentration on the subjects of ones imagination. "Sensitivity" to others means allowing consideration of how other people feel to interrupt ones own train of thought on persistent basis. One so easily distracted cannot be so deeply concentrated.
Why do you assume it's such a horrible distraction? You can always shut the door, focus on what you need to concentrate on when that's what you're supposed to be doing, and only actively attend to others' needs after you're done or while taking a break. I don't see why people should be any more of a distraction than the fact that you could send an instant message to a friend, play Crysis, read a Slashdot article, or any of the other things that might be a distraction from "perfection".
There are two exceptions.
First, if one has a significant attentional deficit, having to deal with interruptions of any source can be a significant impediment, and people are a frequent source of distractions.
Second, if one has a significant degree of Asperger's Syndrome / high functioning autism, attending to the needs of other people can be a very difficult and demanding task, also making it very distracting.
For the large majority of the population--even those on the sides of the distribution that tend towards but don't reach the level of clearly diagnosable disorders--neither of these is relevant.
For example, [selfishness is] a dreadful way to try to raise children.
Raising children and caring for loved ones actually doesn't fit this discussion at all. "Sociopath" is how we describe the people who do not show sensitivity to the considerations of strangers (or relative strangers). And humans are hard-wired to love their young (I would argue that evolution wouldn't let be any other way). Caring for the ones we love is always a loss (in terms of resources -- not in terms of emotional rewards).
I agree that it's often a loss of resources--which sounds an awful lot like the "requiring sacrifice" line you gave originally. So it's destructive, and therefore bad, by your analysis, isn't it?
If not, what's the distinction? "Oh, but it's love," is no explanation. What are the key features that distinguish the utility of love from the utility of unselfishness?
I don't see why, if a strong motivation for personal gain can lead to perfectionism, a strong motivation for the well-being of others cannot also. Because it necessitates sacrifice. And sacrifice is exchange of useless for useful. That is not optimal behavior (as anyone emotionally unattached can see). Putting value on such behavior conditions one away from valuing optimal behavior in other areas of one's life. Whether it is useful or not depends on your motivations. If you gain personal satisfaction from contributing to the satisfaction of others--and this doesn't imply extroversion, just sensitivity (e.g. you can still want to not be bothered for 99.999999% of people)--then perfectionism fits right in.
It is good, by any reasonable definition of the word, to have the capacity for unselfish behavior, since if one fails to have that capacity, one is considerably less likely to gain the support of others. Self-confident people who need each other for personal gains cooperate all the time. Sure. But there are certain regimes where this works well, and others where it works poorly. For example, it's a dreadful way to try to raise children. It's also typically not a good way to run a business with ongoing partners, actually, since you have to devote considerable extra resources to policing. Businesses generally realize this, and tend to try to be perceived as "trusted", "honest", etc., even though it temporarily hurts their bottom line; even from a purely selfish perspective, it's a good idea to fake selfishness/caring.
why is it tyrannical to encourage personality traits that aid social interaction? Because that produces forced interactions. Interactions that are natural do not need to be encouraged. They occur out of necessity. You don't think that necessary interactions can go better if the parties have personality traits that aid social interaction?
Surely you've heard of the Prisoner's Dilemma. Better yet, do you happen to know why it doesn't apply? It doesn't apply because many situations in life have the distribution of weights that is different. The prisoner's dilemma relies on the profit from confess/confess (both people acting selfishly) configuration to be lower than the profit from the do-not-confess/do-not-confess (both people act selflessly) configuration. Right. And you're claiming that there do not exist situations like this in real life?
But this makes some assumptions which do not hold in the real world. One of those assumptions are that work is a zero-sum game. No, this is entirely wrong. The scenario has nothing to do with work, and the point totals do not sum to a constant.
In fact, it's exactly the other way around: if work is a zero-sum-game, there are no Prisoner's Dilemma issues until some other factor intervenes to create local non-zero-sums.
Of course, the only way to "shove off" work in the real world is by paying someone with the money gained from your own work. Really? So people never do favors for each other out of friendship?
So you have a situation where both people act selfishly (try to maximize their gain without regard for the gain of the other party) and yet both end up with higher reward than if they acted selflessly (and did their own work). This just doesn't follow--people can selflessly do each other's work too. If that's a strong enough motivation, it's more optimal than having a monetary exchange since you don't need to waste time and energy figuring out and conducting the exchange. Generally, one can't rely upon it as a motivation, so one has to fall back to the next best thing, usually efficient monetary transactions.
I don't see why, if a strong motivation for personal gain can lead to perfectionism, a strong motivation for the well-being of others cannot also. Either way, it's a motivation to improve. If you soften "cannot" to "is less effective", then sure, self-interest is usually a better motivating factor than interest in the well being of others.
It is good, by any reasonable definition of the word, to have the capacity for unselfish behavior, since if one fails to have that capacity, one is considerably less likely to gain the support of others.
If one is going to enjoy the benefits of a society (specialization, reciprocal aid between individuals who are temporarily in dire straits, collective action to achieve goals too large for individuals to manage, etc.), why is it tyrranical to encourage personality traits that aid social interaction?
Surely you've heard of the Prisoner's Dilemma. You don't get out of that by being rational or selfish.
The overwhelming consensus on both this forum and the comments attached to the Nature page are negative. Okay, fine, the dictator stuff is highly speculative--there's some blame to be apportioned for that.
But the nearly universal negative reaction suggests to me that something else is going on too--some sort of visceral objection to the content, perhaps?
The vasopressin/vasopressin receptor story is an interesting one, worked out best in prairie voles and mice, and the results of this study are basically consistent with previous studies. There isn't much new here to get excited about, except to note that, gee, it looks like humans are mammals too (who'd have thought!)--and the mildly interesting result that dictator games are one place where the difference in vasopressin signaling correlates with behavior.
So shouldn't the appropriate stance be something like, "Huh, kinda interesting, wish they hadn't written up so much over-the-top speculation. (Wish they hadn't done that on this cancer story, either, or this one on biofuels, or that one on....)"?
If the problem is that one is worried that genes control your behavior--sorry, news flash, your genetic makeup is important! It's one of the consequences of being a physically implemented being. This doesn't mean that you are completely at the whim of your genetic makeup. That is the whole point of having a cortex, you know, to modify your behavior in ways that are too complex for genes to do alone!
Almost everyone is not a ruthless genocidal dictator, so if you happen upon someone with an overly short AVPR1a gene, maybe you should realize it's a little harder for them to be unselfish, so they need a little more encouragement for it to become a habit. And if you have an overly short AVPR1 gene, maybe you can remember if you're feeling particularly greedy that your feelings in this matter might be poorly calibrated (especially if they seem to have been badly calibrated in the past), so you should think about what you're about to do a little more carefully.
Brains are really cool, especially when they're used.
The Economist, despite its obvious corporatist bias (although I would rank it as less so than the WSJ), does a pretty good job at reporting facts accurately as far as I can tell. They don't always dig as deeply on all sides for difficult-to-uncover facts, but their bias appears to me mostly as an interpretation that they add on top of the facts, not in presenting non-facts or highly biased facts. This makes it a very useful information source for me, because I can not only tell what they think about an event, but also why, and disagree if I feel so inclined. With many other news sources, the accuracy is low enough so that by default I don't even know what the facts are.
I've seen some examples where they've not gotten their facts right and been absolutely nailed in letters to the editor (and they published such letters too, though probably only the more polite ones). Seeing that happen in areas that I know about--and seeing them get quite a lot right in areas that I know about--has raised my confidence that a lot of the information is actually accurate, even if I disagree with their opinions on the facts.
I've been less impressed with the NYT in this regard.
The issue is not about censoring a plausible idea. The question is whether highly hyped papers with extraordinarily little data are a good idea.
The problem with publications that have n=1 and consist of anecdotal observations is that they are highly unreliable. Since most "negative result" papers never get published, allowing unreliable anecdotes to be published leaves a trail of debris in the literature--you see various promising-sounding papers with no follow-up. You can surmise that the claims were probably false, but it could also be just that no-one was paying attention. It would be good if more negative-result papers were published, but until then, sloppy, unreliable work shouldn't be published in scientific journals. It's a distraction.
I'm really not sure why this "study" was worth publishing. Where are the statistics of patient status after injection of drug vs. injection of drug-free control? How about a timecourse? Or anything besides anecdotes from one patient?
The hype on the article compared to what is shown is shocking. Even if the compound is a silver bullet that instantly and completely reverses Alzheimer's, you'd never know it from a paper like this. So this is an essentially useless bit of PR.
Well, I would happily voluntarily donate to help people in distress, but maybe not if they're only in distress because they do stupid stuff like not wear a seatbelt, and only fail to wear one because they can count on my soft-heartedness to save them after they suffer the consequences of their bad choices.
There are certain categories of mistake that people tend to make (e.g. dramatically under- or over-valuing rare events, e.g. brain damage that can be avoided by wearing a seatbelt, or dying in a plane crash). It's not really reasonable to expect people to make the right decisions for the right reasons--we're not built to reason that way instinctively, and doing it consciously is unreliable--so it makes sense to me to, once we identify such situations, simply say, "do it that way or else".
Exactly what the else entails--a loss of sympathy from me, or a fine from a police officer--should be decided on a case-by-case basis depending on the ease of keeping track of what the person has done, how onorous the demands would be, whether enforcement would be easy or hard, etc..
But even then, it's not really fair to give people the freedom to be stupid if you can't give the freedom to not care to the people who end up around the stupid ones. If we were ideal rational agents, maybe something like that would work. But last I checked, we're not.
If you would rather be left to painfully die in a car wreck (or to be saved and then left to painfully die on the street outside a hospital) when you can't immediately afford the cost of care, then sure, you should be allowed to avoid seatbelt laws.
Just figure out a way to avoid imposing too much of an emotional cost on people who are bothered by (your) suffering, and a way to keep track of who is like you and wants to live as if they're not part of a social group, and who likes the normal methods of non-explicit reciprocal responsibilities of individuals and the group to which they belong, and I'll be all for it.
And I hope that when you were growing up you went to a private school--not a public one that I partially funded--and if you went to university you also went to a private university, and so on; or, you've already repaid these costs in taxes and/or donations that you've made since then.
The author is either using a very peculiar form of Condorcet with reversals, or is just confused.
Suppose 50 people prefer C to B, 51 people prefer B to A, and 776 people prefer A to C. If you prefer C to B to A, there's *nothing you can do* to make C win (the author claims you can make C win)--the 776 margin is too large to be reversed, regardless of how the margin is counted. All you can do is influence whether A or B wins, since either the CB contest will get reversed (making B the winner over both C and A), or the AB contest will get reversed (making A the winner over both C and B).
It is true that if you count the number of places between the two and use that to score the preference--i.e. C vs. A would give two points in favor of C if B is in between--you create a pressure to place the candidates farther apart from each other than you really believe.
But the solution is trivial. Don't do that! If C beats A, it counts as one person preferring C to A.
This yields a system where it is extremely unlikely that dishonest voting will help the voter, and thus everyone should vote honestly, and thus by the author's measures, Condorcet yields the best outcome when the voters vote maximally in their own interest.
Laws that protect one person from themselves are useful if the rest of society invests resources in education or takes care of hurt people. In that case, hurting oneself does, in practice, impose a cost on society; you can't hurt yourself "for free".
Slashdot Mirror
Use
andtags to enclose your code to get indentation (you can then just write the indentation with spaces).Also, you're right that I didn't conform exactly to the specifications; I probably should have added a "return list;". But when not doing functional programming, there's often no reason to bother.
Lambda functions in Python don't really have any advantages over inner classes in Java except that the syntax is a little shorter and Java's standard libraries don't use the model enough.
I also think you need to have more complex examples to see the power of closures or lambda functions.
For example,
int n = 0;
for (Integer i : list) n += i*i;
is shorter than the closure code to do the same thing (and is easier on one's working memory). Likewise for
void AddL(List<AtomicInteger> list,int increment)
{
for (AtomicInteger i : list) i.set( i.get()+increment );
}
And these would be prettier still in a weakly typed language, or a bit prettier if Java would assume sensible defaults.
Often, languages that have lambda functions need them because they have low-level types that are accessed by value (or other immutable types). For example, in Python,
numbers = [1,2,3,4,5]
for i in numbers: i = i+5
does not do what you want because i is not accessed by reference. So you write
numbers = [1,2,3,4,5]
numbers = map(lambda i:i+5,numbers)
and feel all pleased, except that you only needed to do that because "for" only refers to objects, not base types, by reference.
Lambda calculus is very powerful, but I've not seen any simple examples that actually demonstrated the power beyond what you could achieve with standard programming constructs. Closures are nice for things like sorting, UI events, and such--but inner classes are even nicer because they can carry data too in a more obvious way (instead of simply by scoping rules).
(3) Selfless. Acting to improve the welfare of others without regard for my own welfare.
This does not exist. Whatever the motivation, we choose our actions. And thus we choose our end-goals.
If one's goal is to "make sure that Sally has the best life possible, whatever the consequences to me", then it's a selfless choice. I'm not saying that the person isn't making the decision, and they may have reasons/feelings for doing so, but if they're basically leaving themselves out of the equation, that is sensibly called selfless.
Defining an attempt to achieve ones' end goals as selfish (even though it may benefit others) makes this (3) an impossibility.
I don't think this is a useful definition--then everyone would be selfish except for people with such major cognitive deficits as to be unable to form or pursue goals.
The interesting distinction regards the nature of those goals, which is why my definition included the "without regard for the welfare of others" caveat.
The only axiom of ethics that I put forward is "do not do to others what you wouldn't have them do to you".
What is your justification for adopting this axiom? Note, for example, that under your ethical system, it's perfectly okay to leave someone bleeding to death to die when you could easily save them.
If one wants to elicit an action from anyone else, one must do so through an agreement.
That's a little difficult for the bleeding person if they're already unconscious.
Anyone who claims to care to much for me is assumed to be attempting to compel me to act without paying for it (ie, without eliciting my consent).
This is generally a good guess when dealing with salespeople. It's a rather weird way to look at things when the "anyone" is, say, a parent or spouse.
Reasonable behavior is a choice and (as such) it is a conscious one. It is not always a psychological need. Given the restrictions put forward above, it will produce a better society.
That doesn't mean that it's attainable in practice. Communism didn't fail because it was a bad idea in principle--it failed because it was unattainable in practice because humans don't act like Marx wanted them to. Expecting reasonable behavior without emotional support seems, well, unreasonable.
And having a "civil" society (a society in which it is not acceptable to take proactive steps to hurt each other) is a worthwhile goal. Confusing it with a society in which people are compelled to take care of the people they don't care to take care of will quickly deteriorate into a society of caprices and whims and will erase civility.
Indeed. Sweden is so uncivil. Japan, too.
All the "community-oriented" thinking produces is xenophobia and lack of privacy.
Indeed, which is why so many people say they feel so good after volunteering.
You might want to pay more attention to reality before making such sweeping statements.
With one reminder, once personal freedom is taken away, a group of people (acting in the name of a "greater good") becomes a tyrannical entity... an entity that rules over adults as parents do over children. Taking up arms against such an entity is not immoral.
Are children justified in taking up arms against parents? Why or why not?
The degree to which a violent revolution is justified is largely controlled by the degree to which the personal freedoms have been infringed.
I largely agree, but there is a weighting factor that I apply to various different aspects of freedom--some are more important than others.
But this isn't terribly relevant to whether one shou
But if one only looks out for one's own interests, then having this weird rule about "don't use other people" is suboptimal--it is in effect a hard-wired unselfish motive dropped into the middle of a selfish framework. There's nothing wrong with that, but at least recognize it for what it is!
Ok, sure, but unselfishness, in this stretched definition, is defined negatively. It's sort of equivalent to defining "apathy" as "care" because it is characterized by lack of hate.
Not at all. Here are the three definitions I'm using:
(1) Selfish. Acting to improve my welfare, and ignoring how my actions affect others (except inasmuch as that will then affect me).
(2) Unselfish. Acting to improve the overall welfare of some group (or with regard to many groups), of which I may be a part.
(3) Selfless. Acting to improve the welfare of others without regard for my own welfare.
Both (2) and (3) are forms of altruism; I am more concerned with (2) than (3).
But if one only looks out for one's own interests, then having this weird rule about "don't use other people" is suboptimal
No, it isn't. You forgot the "renormalize" step at the end of each iteration.
If you could actually do this reliably, it would be a good idea. I will explain below why in practice one cannot. Your analysis isn't far off in certain ideal cases, but it fails to cover all commonly encountered cases.
But the principles you cited do not justify your choice of honesty as a virtue.
I think the above [now removed] should explain it.
It explains it in certain sets of interpersonal relationships. There are piles of exceptions. Here are three.
(1) Relationships that can be presumed to be one-time-only. For example, if you see someone drop their wallet and not notice, the selfish thing to do is wait until they're out of sight and take all the cash out of the wallet. The unselfish thing to do is to tell them they've dropped their wallet. These type of win/lose situations carry essentially no cost to generate, do not (if done halfway carefully) affect your reputation, and benefit you. They often hurt the social system overall, though, as people have to spend lots of extra effort locking up all their stuff, replacing all their credit cards when their wallets disappear for a bit (and appear without cash), and so on.
(2) Relationships which are not easily broken once made. A typical example of this is vendor lock-in: you engage in a mutually beneficial relationship to the point where the other party is dependent on you but you are not dependent on them. You then shift the relationship from a win/win situation to a win/lose situation where they accept the loss because breaking the relationship would entail taking an intolerably high loss.
(3) Relationships which are unavoidable. Chances are that you only have one set of power lines into your house. In fact, chances are that there are only one set of power lines within any distance that you can reasonably cover. Assuming that you like electricity, you have very little walk-away power when dealing with your electricity provider. Likewise with cable, telephone, etc.; there is a tiny bit of cross-monopoly competition, but basically, if one has a monopoly over a very important resource, one can impose huge win/tiny win conditions on everyone. Some resources naturally monopolize due to physical constraints (e.g. roadways).
Do you stop if it's win/lose, with you doing the winning? If so, why?
I am pretty sure I explained that above... Mostly because it's not worth the time. Probably, because it also raises the cost for starting new win/win games (you get a "reputation").
Reputation is essentially a societal metric for measuring the fairness (or other qualities) of your interactions. Effectively, it's a "do you
(Again, note the selfish / unselfish / selfless distinction that I'm making here.)
At that point, the question is not whether these things exist, but whether it is useful to encourage or discourage them. Mine is based on using one's internal judgement. Yours comes from judging oneself through the eyes of others (you compare yourself to others to get a sense of "how you are doing" in term of resources gained). If you're in a social environment, it's generally worthwhile to pay some attention to how you are interacting with the group, since otherwise you won't be able to predict how others will react to you. For example, you won't know whether they think you're dishonest. My paradigm excludes possibility of the win/loss solution. And as soon as the solution becomes loss/win, it stops the game. As soon as dishonesty occurs, I don't to play anymore. That's win/win or no deal. Do you stop if it's win/lose, with you doing the winning? If so, why? PD doesn't allow for this scenario -- it insists that one keep playing even after loss/win in order to maximize one's gain. Not all interactions are optional if you require resources from others for survival, or if the situation otherwise does not allow withdrawal at no cost. There are many cases where your strategy works; I'm concerned about those and those where it doesn't.
Exhausting where? Not to the photoreceptor cells. Do you mean to bipolar cells? Ganglion cells? What is the evidence that retinal fatigue, as opposed to processes in the LGN, V1, superior colliculus, etc., has anything to do with subjective fatigue?
Greed for power is also a problem, and it's also a case of not being unselfish, and it's not something that monetized interactions particularly help with. Since you seemed to be favoring monetized selfish interactions, I focused more on the problems of financial selfishness.
I'm not assuming greed is bad. I've been giving examples where greed leads to suboptimal outcomes for everyone involved (PD-style). All you've managed to say is something like, "Nuh-uh, it doesn't work that way," with much explanation but mostly by bringing up specific cases where selfishness is okay, not addressing the central problem. I've never claimed that selfishness is always bad, just that unselfishness is superior in some situations (not necessarily for the one being unselfish, but for everyone overall).
I'm familiar with objectivism and Marxism/communism. They're both overly naive, but good to know about if one hasn't yet encountered the ideas. Marxism is particularly naive about human motivations (having fallen entirely victim to the "blank slate" theory of human psychology); Objectivism is also naive about human motivations (having fallen victim to the "rational actor" theory of human psychology) and doesn't quite grasp the consequences of its own premises either as applied to humans or to ideal rational actors.
Both the Communist Manifesto and Atlas Shrugged (or some other synthesis of Rand's ideas) are worth reading, but one has to constantly ask, "Really?" when evaluating the claims and arguments therein.
Better than what? Isn't it suboptimal--"destructive"--to not cheat when you have the opportunity to do so and the expected costs are less than the expected reward? Why all the language about esteem and desperation?
Or have you just noticed a case where unselfishness is a good policy because of PD-type considerations?
So your premise is wrong. The retina is highly active even in the dark.
At this point, I think that all I can say is that you have a really odd view of the world. You might want to think more about the nature of human motivations and social interactions, and also about less drastic ways of curbing human abuses than monetizing all interactions (not that this really helps, as Kerviel's actions at Societe Generale attest).
You've made a number of additional points, but you seem to have so poorly grasped the essence of the previous points that I don't expect it would be fruitful to continue the discussion. You're quite good at explaining your position and reasoning through it, but have displayed almost no imagination when it comes to considering any other hypotheses. Thus, your objections to my points of view have been largely lacking in insight. For example, even if nepotism were this horrible evil that we had to combat at every opportunity, a solution would be to have additional transparency and third-party reviews of performance. That is, pit public shame against greed expressed through social ties. Fringe benefit: that works for dampening individual greed, too.
You're right that there's a second system there, but I'm not introducing it. That's the original system. We introduced a monetary system on top of our social interaction system when we started living in such large groups that it became too difficult to keep track of personal relationships.
We have various instincts regarding social exchange that make it a superior system in some contexts--most notably that we tend to get stronger feelings that we should act a certain way, and that motivates us; when that way is a way that solves PD problems, it's a great benefit.
When it's a way that creates more problems than it solves (if any), it's better to use money.
I'm not saying not to use money; it's just that we have the other system and may as well use it for what it's good for.
I agree that businesses are in danger of nepotism, but that's a danger regardless of the system used. The problem with nepotism is that favoritism leads to bad decisions that hurt the overall performance of the business as people selfishly siphon off resources to themselves and those they want to benefit. If people weren't selfish (in a broader sense which includes "cronies"), nepotism would be a danger. One way to solve the problem is to remove any illusion of unselfishness, so everyone knows to watch everyone else closely. Another way is to collect people who are less selfish.
I'm afraid the "those people are an extension of myself" argument is not very helpful. Why not extend that same feeling, at a weaker level, to others who you're less desperately and unavoidably in love with? Besides, whether you can help it or not, if it's destructive it's still not something a perfectionist would want to accept.
Something is constructive if it alleviates the harshness of mother nature (and thus promotes pleasantness of life) and destructive if it does the opposite.
Okay, we agree on the standard.
Under this standard (which you are at least tempted to consider right now), exchanging lesser value for higher value is destructive.
Fair enough.
Because the assumption has to be made that those who create value have value to exchange and those who consume it (and don't create) require sacrifice.
If by "sacrifice" you mean "effort expended on someone other than myself", I suppose the assumption is somewhat relevant. But if the overall situation is such that the value of the help being received to the person receiving it is greater than the lost value to you, then overall the interaction is not destructive. It's just potentially destructive-to-you if you can't benefit from such interactions in the opposite direction often enough.
Being an unselfish person in a selfish environment is a bad idea.
Extroverts would be the people that seek interaction. A human being creates value by thinking (doing physical work is also result of some thought process, so, yes, even a human being performing manual labor). What is established during interaction with other human beings is a good will (a feeling of family, comradery, tribalism, etc) that allows for an honest exchange at a future time. But the value which can be exchanged (the one that results from tranformation of pristine natural things into human-usable things) must still be created by individuals alone with their perception. Perfectionism must necessitate introverted effort of keeping ones concentration on the subjects of ones imagination. "Sensitivity" to others means allowing consideration of how other people feel to interrupt ones own train of thought on persistent basis. One so easily distracted cannot be so deeply concentrated.
Why do you assume it's such a horrible distraction? You can always shut the door, focus on what you need to concentrate on when that's what you're supposed to be doing, and only actively attend to others' needs after you're done or while taking a break. I don't see why people should be any more of a distraction than the fact that you could send an instant message to a friend, play Crysis, read a Slashdot article, or any of the other things that might be a distraction from "perfection".
There are two exceptions.
First, if one has a significant attentional deficit, having to deal with interruptions of any source can be a significant impediment, and people are a frequent source of distractions.
Second, if one has a significant degree of Asperger's Syndrome / high functioning autism, attending to the needs of other people can be a very difficult and demanding task, also making it very distracting.
For the large majority of the population--even those on the sides of the distribution that tend towards but don't reach the level of clearly diagnosable disorders--neither of these is relevant.
For example, [selfishness is] a dreadful way to try to raise children.
Raising children and caring for loved ones actually doesn't fit this discussion at all. "Sociopath" is how we describe the people who do not show sensitivity to the considerations of strangers (or relative strangers). And humans are hard-wired to love their young (I would argue that evolution wouldn't let be any other way). Caring for the ones we love is always a loss (in terms of resources -- not in terms of emotional rewards).
I agree that it's often a loss of resources--which sounds an awful lot like the "requiring sacrifice" line you gave originally. So it's destructive, and therefore bad, by your analysis, isn't it?
If not, what's the distinction? "Oh, but it's love," is no explanation. What are the key features that distinguish the utility of love from the utility of unselfishness?
In fact, it's exactly the other way around: if work is a zero-sum-game, there are no Prisoner's Dilemma issues until some other factor intervenes to create local non-zero-sums. Of course, the only way to "shove off" work in the real world is by paying someone with the money gained from your own work. Really? So people never do favors for each other out of friendship? So you have a situation where both people act selfishly (try to maximize their gain without regard for the gain of the other party) and yet both end up with higher reward than if they acted selflessly (and did their own work). This just doesn't follow--people can selflessly do each other's work too. If that's a strong enough motivation, it's more optimal than having a monetary exchange since you don't need to waste time and energy figuring out and conducting the exchange. Generally, one can't rely upon it as a motivation, so one has to fall back to the next best thing, usually efficient monetary transactions.
I don't see why, if a strong motivation for personal gain can lead to perfectionism, a strong motivation for the well-being of others cannot also. Either way, it's a motivation to improve. If you soften "cannot" to "is less effective", then sure, self-interest is usually a better motivating factor than interest in the well being of others.
It is good, by any reasonable definition of the word, to have the capacity for unselfish behavior, since if one fails to have that capacity, one is considerably less likely to gain the support of others.
If one is going to enjoy the benefits of a society (specialization, reciprocal aid between individuals who are temporarily in dire straits, collective action to achieve goals too large for individuals to manage, etc.), why is it tyrranical to encourage personality traits that aid social interaction?
Surely you've heard of the Prisoner's Dilemma. You don't get out of that by being rational or selfish.
The overwhelming consensus on both this forum and the comments attached to the Nature page are negative. Okay, fine, the dictator stuff is highly speculative--there's some blame to be apportioned for that.
But the nearly universal negative reaction suggests to me that something else is going on too--some sort of visceral objection to the content, perhaps?
The vasopressin/vasopressin receptor story is an interesting one, worked out best in prairie voles and mice, and the results of this study are basically consistent with previous studies. There isn't much new here to get excited about, except to note that, gee, it looks like humans are mammals too (who'd have thought!)--and the mildly interesting result that dictator games are one place where the difference in vasopressin signaling correlates with behavior.
So shouldn't the appropriate stance be something like, "Huh, kinda interesting, wish they hadn't written up so much over-the-top speculation. (Wish they hadn't done that on this cancer story, either, or this one on biofuels, or that one on....)"?
If the problem is that one is worried that genes control your behavior--sorry, news flash, your genetic makeup is important! It's one of the consequences of being a physically implemented being. This doesn't mean that you are completely at the whim of your genetic makeup. That is the whole point of having a cortex, you know, to modify your behavior in ways that are too complex for genes to do alone!
Almost everyone is not a ruthless genocidal dictator, so if you happen upon someone with an overly short AVPR1a gene, maybe you should realize it's a little harder for them to be unselfish, so they need a little more encouragement for it to become a habit. And if you have an overly short AVPR1 gene, maybe you can remember if you're feeling particularly greedy that your feelings in this matter might be poorly calibrated (especially if they seem to have been badly calibrated in the past), so you should think about what you're about to do a little more carefully.
Brains are really cool, especially when they're used.
The Economist, despite its obvious corporatist bias (although I would rank it as less so than the WSJ), does a pretty good job at reporting facts accurately as far as I can tell. They don't always dig as deeply on all sides for difficult-to-uncover facts, but their bias appears to me mostly as an interpretation that they add on top of the facts, not in presenting non-facts or highly biased facts. This makes it a very useful information source for me, because I can not only tell what they think about an event, but also why, and disagree if I feel so inclined. With many other news sources, the accuracy is low enough so that by default I don't even know what the facts are.
I've seen some examples where they've not gotten their facts right and been absolutely nailed in letters to the editor (and they published such letters too, though probably only the more polite ones). Seeing that happen in areas that I know about--and seeing them get quite a lot right in areas that I know about--has raised my confidence that a lot of the information is actually accurate, even if I disagree with their opinions on the facts.
I've been less impressed with the NYT in this regard.
The issue is not about censoring a plausible idea. The question is whether highly hyped papers with extraordinarily little data are a good idea.
The problem with publications that have n=1 and consist of anecdotal observations is that they are highly unreliable. Since most "negative result" papers never get published, allowing unreliable anecdotes to be published leaves a trail of debris in the literature--you see various promising-sounding papers with no follow-up. You can surmise that the claims were probably false, but it could also be just that no-one was paying attention. It would be good if more negative-result papers were published, but until then, sloppy, unreliable work shouldn't be published in scientific journals. It's a distraction.
The prior study was interesting, and was published. The new one doesn't add much.
I'm really not sure why this "study" was worth publishing. Where are the statistics of patient status after injection of drug vs. injection of drug-free control? How about a timecourse? Or anything besides anecdotes from one patient?
The hype on the article compared to what is shown is shocking. Even if the compound is a silver bullet that instantly and completely reverses Alzheimer's, you'd never know it from a paper like this. So this is an essentially useless bit of PR.
Well, I would happily voluntarily donate to help people in distress, but maybe not if they're only in distress because they do stupid stuff like not wear a seatbelt, and only fail to wear one because they can count on my soft-heartedness to save them after they suffer the consequences of their bad choices.
There are certain categories of mistake that people tend to make (e.g. dramatically under- or over-valuing rare events, e.g. brain damage that can be avoided by wearing a seatbelt, or dying in a plane crash). It's not really reasonable to expect people to make the right decisions for the right reasons--we're not built to reason that way instinctively, and doing it consciously is unreliable--so it makes sense to me to, once we identify such situations, simply say, "do it that way or else".
Exactly what the else entails--a loss of sympathy from me, or a fine from a police officer--should be decided on a case-by-case basis depending on the ease of keeping track of what the person has done, how onorous the demands would be, whether enforcement would be easy or hard, etc..
But even then, it's not really fair to give people the freedom to be stupid if you can't give the freedom to not care to the people who end up around the stupid ones. If we were ideal rational agents, maybe something like that would work. But last I checked, we're not.
If you would rather be left to painfully die in a car wreck (or to be saved and then left to painfully die on the street outside a hospital) when you can't immediately afford the cost of care, then sure, you should be allowed to avoid seatbelt laws.
Just figure out a way to avoid imposing too much of an emotional cost on people who are bothered by (your) suffering, and a way to keep track of who is like you and wants to live as if they're not part of a social group, and who likes the normal methods of non-explicit reciprocal responsibilities of individuals and the group to which they belong, and I'll be all for it.
And I hope that when you were growing up you went to a private school--not a public one that I partially funded--and if you went to university you also went to a private university, and so on; or, you've already repaid these costs in taxes and/or donations that you've made since then.
The author is either using a very peculiar form of Condorcet with reversals, or is just confused.
Suppose 50 people prefer C to B, 51 people prefer B to A, and 776 people prefer A to C. If you prefer C to B to A, there's *nothing you can do* to make C win (the author claims you can make C win)--the 776 margin is too large to be reversed, regardless of how the margin is counted. All you can do is influence whether A or B wins, since either the CB contest will get reversed (making B the winner over both C and A), or the AB contest will get reversed (making A the winner over both C and B).
It is true that if you count the number of places between the two and use that to score the preference--i.e. C vs. A would give two points in favor of C if B is in between--you create a pressure to place the candidates farther apart from each other than you really believe.
But the solution is trivial. Don't do that! If C beats A, it counts as one person preferring C to A.
This yields a system where it is extremely unlikely that dishonest voting will help the voter, and thus everyone should vote honestly, and thus by the author's measures, Condorcet yields the best outcome when the voters vote maximally in their own interest.
Laws that protect one person from themselves are useful if the rest of society invests resources in education or takes care of hurt people. In that case, hurting oneself does, in practice, impose a cost on society; you can't hurt yourself "for free".