Disqus Profile - Yvain

Sign Up Log In
Disqus
 disqus_l9FtbVgwO9
Yvain Follow 122
Recent Activity
Comments 33
Favorites
Followers 2
Following 1
Discussion on Overcoming Bias 18 comments
This Time Isn’t Different
Yvain
Yvain 24 days ago
"One way to understand this is in terms of the distribution over human jobs of how good machines need to be to displace humans. If this parameter is distributed somewhat evenly over many orders of magnitude, then continued steady exponential progress in machine abilities should continue to translate into only slow incremental displacement of human jobs. Yes machines are vastly better than they were before, but they must be much far vastly better still to displace most human workers."
Why can't jobs be arranged in a series of bottlenecks, such that solving the bottleneck opens a very large category of things to automation?
For example, human level visuospatial movement is a plausible near-term advance, and might allow the automation of waiters, janitors, tour guides, taxi drivers, et cetera.
Human level natural language recognition (like Watson) is another, and potentially hurts secretaries, call center employees, et cetera.
More generally, if you imagine jobs as stratified by IQ (with people able to substitute any job for any other job up to their IQ level), you see no unemployment once machines reach IQ 10, still no unemployment at IQ 20, still none at IQ 30, and then once you reach 80 or so suddenly a few more steps put nearly everyone out of work.
If you imagine machines improving one IQ point per year, then you can have the first eighty years of machine progress with minimal unemployment, and then the next thirty put almost everyone out of work.
Obviously IQ isn't a good way to model machine progress, but there could still be a similar pattern of relatively low unemployment until machines approach "human level" by some metric, and then very sudden unemployment of non-intellectual tasks.
Since a lot of the hard work is done by programming the machines and giving them lookup table type things for specific situations, I could imagine this happening well before the machines reach anything like general intelligence.
9 View Comment
Why I changed my mind and took the Giving What We Can Pledge
Discussion on The Uncredible Hallq 2 comments
Why I changed my mind and took the Giving What We Can Pledge
Yvain
Yvain 6 months ago
Before I took the pledge, I asked GWWC if it was okay to donate to x-risk charities instead of developing world charities. They said yes.
View Comment
How much you like someone is a poor predictor of their ethical behavior
Discussion on The Uncredible Hallq 52 comments
How much you like someone is a poor predictor of their ethical behavior
Yvain
Yvain 9 months ago
This sort of seems to conflate popularity and perceived niceness. I am very willing to believe that popularity doesn't provide any useful information about whether someone did a bad thing or not. I'm not prepared to completely suspend my judgments about whether.someone is a good person or not. Doing so seems like throwing out my priors for no reason when priors usually provide useful information.
I know some people who are so sketchy that the mere rumor that they've done something wrong would be enough to convince me - my thought process would be "Oh, someone finally caught them doing that thing that everyone already suspected they do."
I also know some people whom it would take extraordinary evidence to make me believe bad things about. If someone told me that Julia from The Whole Sky had - let's not even say committed a crime, let's say "behaved less than maximally nicely in any situation", I would want ten signed affadavits and video evidence before I even considered it.
I understand you're going from this framework where any time someone is accused of something in the news, their parents/friends/supporters say "Oh, he could never do that, he's so nice." But that's just an existence proof, saying that niceness isn't a 100% accurate filter for non-criminality - and it's one filtered by the media's man-bites-dog bias as well. My guess is that for most criminal acts, there are a bunch of people saying "Yup, we knew he was a jerk all along", that for most people who turn out to be innocent there are a bunch of people saying "Yeah, we knew he wouldn't do it, he's too nice", and that you can actually pick up substantial information from the percent of people who say these two things.
I think you recently called this the Litany of Hermione: "The thing that people forget sometimes, is that even though appearances can be misleading, they're usually not."
View Comment
Open letter to the defenders of Phil Robertson.
Discussion on What Would JT Do? 126 comments
Open letter to the defenders of Phil Robertson.
Yvain
Yvain a year ago
So I'm the person you are insisting doesn't exist - a completely pro-gay atheist who voted against Proposition 8 and thinks supporting gay marriage is a no-brainer, who also is kind of horrified at Phil Robertson being fired for his comments.
You are 100% correct that freedom-of-speech only binds the government and does not constrain private actors from punishing people whose speech they don't like.
But let's compare and contrast. Freedom-of-religion *also* only binds the government and does not constrain private actors from punishing people whose religion they don't like. If someone wants to picket a mosque while waving signs about how all Muslims are dirty terrorists who are going to Hell, Constitutional freedom-of-religion is a-ok with that. Heck, Constitutional freedom-of-religion is okay with Christian-owned businesses refusing to hire atheist employees or serve atheist customers - it's only unrelated more recent anti-discrimination laws that prevent that.
Point is, there's a big gap between "constitutional freedom of religion" and "the level of religious tolerance that is necessary to have a remotely civil society". Some of that gap can be filled in by laws, but a lot of it can't be. It's supposed to be filled in by basic human decency and understanding of the principles that made freedom of religion a good idea to begin with.
I think the same is true of freedom of speech. Constitutional freedom-of-speech is a necessary but not sufficient condition to have a "marketplace of ideas" and avoid de facto censorship. Beyond that people also have to understand that the correct response to "idea I disagree with" is "counterargument", not "find some way to punish or financially ruin the person who expresses it". If you respond with counterargument, then there's a debate and eventually the people with better ideas win (as is very clearly happening right now with gay marriage). If there's a norm of trying to quash the people with opposing views, then it doesn't really matter whether you're doing it with threats of political oppression, of financial ruin, or of social ostracism, the end result is the same - the group with the most money and popularity wins, any disagreeing ideas never get expressed.
Atheists may one day be the group with the most money and popularity, but that day isn't today and right now it's neither moral nor in our self-interest to encourage using greater resources to steamroll opponents. It's certainly not in gay people's self-interest either. Why shouldn't companies owned by Christians fire all gay people on the grounds that they are sinners? Right now it's because we have a mutual truce in which we agree businesses should employ people based on their skills and merit rather than to reward their political allies and punish our political opponents. Once you undermine that, gay people are in a pretty precarious position.
So I would turn your own hypothetical scenario in Part 2 of your post back on you. Suppose Robertson had, indeed, been a gay rights supporter - or a gay person! - who said on national news he thought that everyone should stand up for gay rights. But his company was going for the fundie demographic and decided to fire him for his statement. Would you be so quick to attack everyone who was disappointed in this action, so eager to stand up for the right of companies to fire anyone they disagree with?
I'm an atheist blogger and I work at a Catholic hospital. Employer tolerance for dissenting opinions is *personal* for me. I'm disappointed in the tone of this post and hope you reconsider.
31 View Comment
Spare a penny for the poor of patience, guv?
Discussion on Unequally Yoked 42 comments
Spare a penny for the poor of patience, guv?
Yvain
Yvain  Kristen inDallas a year ago
The problem is that there is a direct tradeoff between a charity's effectiveness at providing what you call "personal enrichment opportunities" and its effectiveness at doing other things like feeding the hungry or curing the sick.
If people just want personal enrichment that's great. But insofar as some people really cares about people starving to death, and want that to not happen - SERIOUSLY want it to not happen in the way you would want to make sure your own family didn't starve to death and wouldn't be too concerned about who did or didn't get personally enriched on the way to saving them - it's good that they have somewhere like GiveWell to help them out.
3 View Comment
Yvain
Yvain a year ago
I think GiveWell is operating on a utilitarian assumption - perhaps unexamined but probably not wrong - that for most reasonable definitions of utility the sorts of charities they look at will be more effective than the charities they don't look at.
I'm not sure what kinds of charities would make children more gratification-delaying, but I can't imagine any being so effective that the average person who really understands the concept of tradeoffs would prefer to invest in the gratification one than the public health one.
You'd pretty much have to say that if you had X kids, you'd rather one die and X-1 become better at gratification-delay than have all of them survive.
The only two cases in which I can imagine that being reasonable are first, astronomically high values of X which are pretty unlikely, or else if instilling people with all these virtues was then going to make them into supermen who could solve all public health problems themselves and end up saving more lives than the direct health investment. Which given the terrible track record of the public mass production of virtue also seems unlikely.
3 View Comment
How Do You Solve a Problem Like Atonement?
Discussion on Unequally Yoked 69 comments
How Do You Solve a Problem Like Atonement?
Yvain
Yvain  Iota a year ago
If you want.
View Comment
Yvain
Yvain  Iota a year ago
Thanks for the well thought out response. But:
a) I feel like if everyone is acting in good faith, calculability is a pretty reasonable assumption. We can make an analogy to trade. I want to sell you my old jacket. You want to buy the jacket. There is a tiny chance that sewn into the liner is a map to a priceless pirate treasure, or that it is contaminated with a chemical that will slowly kill whoever wears it, but usually the tiny chances of these things get factored into a relatively normal price without disrupting it too much.
b) Aside from very extreme cases like killing someone's loved one, I don't think this is true. Let's take your example of making someone work overtime when they have enough money but REALLY want to spend time with their family. But suppose we gradually raise our level of overtime pay. Maybe at $25/hour they still prefer time with their family. Maybe even at $100/hour this stays true. But if we offer (a normal middle class person) $100,000 hour, the chance that they turn this down is essentially zero. That suggests that in fact there *is* some amount we can give to compensate them for their time, and that it's somewhere between $100 and $100,000.
c) I kind of think it's true people should only atone for damage they could foresee. The classic example is the doctor who saves a dying child, and then that child grows up to be Hitler. As far as I can tell, that doctor does not need to "atone" for the Holocaust. This seems just as correct as the inverse argument: that if an axe murderer randomly kills a baby, and unbeknownst to him that baby would later have grown up to be Hitler, this doesn't suddenly make him a wonderful person or mean he gets to take credit for saving millions of lives.
d) Presumably these all cancel out. If I make you fifteen minutes late, it's *possible* I prevented you from meeting your future husband, but also *possible* that I saved you from being in a deadly car accident. I wouldn't want to demand restitution from you for all the car accidents I saved you from, so presumably you don't deserve any restitution from me for all the husbands I've prevented you from meeting.
e) This seems sort of like you're saying this is impossible or undesirable, but it seems to be the principle behind for example carbon taxes or cigarette taxes, both of which seem like good ideas.
I guess this kind of assumes good faith between everyone involved, but without that, this is a problem that far outstrips atonement. If a butterfly flapping its wings in China causes a hurricane in Florida, then probably every time you cough you're preventing someone in the year 2120 from meeting their future husband. I don't think you should feel guilty about this any more than you should feel triumphant about the fact that last time you stepped on a bug it prevented the Third Human-Mantis War in 2860. And the same sort of sanity-preserving mental heuristics you use to avoid those kinds of weird conclusions seem like they should suffice for the atonement problem as well.
1 View Comment
Yvain
Yvain  Christian H a year ago
This, and also that for many people there is no amount of money/value that they would accept as equal to having their loved one back.
1 View Comment
See More
[Turing 2013] Atheist Entry #8
Discussion on Unequally Yoked 15 comments
[Turing 2013] Atheist Entry #8
Yvain
Yvain  Slow Learner a year ago
"What atheist would describe the phrase "human dignity" as a Christian one?"
*tentatively raises hand*
Actually I and several other atheists I know have had whole conversations about Christian overuse of "human dignity". It's not that we're against human dignity per se, just that the Christian arguments we've read appear to, from our perspective, drop it kind of at random in places that don't seem to fit. If I were trying to parody or insult Christian argument at a level slightly above "WE HAVE TO BAN GAY MARRIAGE BECAUSE THE BIBLE SAID SO", I would try "WE HAVE TO BAN GAY MARRIAGE BECAUSE NOT DOING SO WOULD BE AN OFFENSE TO THE DIGNITY OF THE HUMAN PERSON"
If this entry is by a Christian, I give them + 1 Turing Test point for noticing.
5 View Comment
One-Sided Turing Tests and Privilege
Discussion on Unequally Yoked 41 comments
One-Sided Turing Tests and Privilege
Yvain
Yvain  Guest a year ago
Can you give some examples of what sort of studies you mean?
View Comment
Yvain
Yvain  JohnE_o a year ago
Can you tell me to whom I have to show my Y-chromosome to obtain this benefit?
View Comment
Yvain
Yvain  Robby Bensinger a year ago
I think privilege the way Leah means it here means something more like "You are luckier than I am and therefore so blind to the sorts of things unlucky people like me have to put up with that you can't even possibly understand my problems or arguments, and so have no right to participate in this discussion. On the other hand, I understand all of your problems and arguments just fine. Therefore, only my opinion on this issue can be truly valid."
...it sounds kind of evil the way I phrased it above, but it's sometimes true - a classic example might be a man who says "Are you kidding? I'd love to be sexually harassed by a pretty girl", which is reasonable from his perspective but only because he's never had to face any fear in sexual situations or deal with unwanted attention, so just understands them as potential free sex.
The counterargument would be that actually, members of privileged group X can understand the problems addressed by oppressed group Y just fine, they just don't agree with group Y's characterization or proposed solution. Or that there is reciprocal misunderstanding - an oppressed group also can't understand what it's like to be part of a privileged group, and therefore both groups have imperfect understanding of the situation. For example, a white person might not understand why black people want affirmative action, but a black person might not understand why white people *don't* want affirmative action, because she thinks white people can't possibly have any trouble getting jobs so if a white person loses out on one job by affirmative action she can just get another better one somewhere else.
What I mean by "empirically demonstrating privilege", is that oppressed/privileged Turing tests could distinguish between "oppressed group understands privileged group but not vice versa" (the position where 'privilege' makes a lot of sense), "both groups understand each other just fine" (naive position?), "both groups misunderstand each other" (some men's rights groups seem to say this, where feminists have good points but they don't understand that men have complicated problems that need addressing too), and "privileged group understands oppressed group, but not vice versa" (something I have never heard posited, although it might be possible if for example the previous one was true originally but the percolation of social justice memes into our culture has now made everyone super-sensitive to oppressed groups' concerns)
2 View Comment
See More
Discussion on Siris 8 comments
Aquinas on the Evil of Rape
Yvain
Yvain a year ago
See my response at http://slatestarcodex.com/2013...
1 View Comment
My Week at Rationalist Summer Camp [Sequence Index]
Discussion on Unequally Yoked 16 comments
My Week at Rationalist Summer Camp [Sequence Index]
Yvain
Yvain  URDUM a year ago
I wrote about my take on Less Wrong's belief in cryonics here in a way that I think might answer some of your objections.
1 View Comment
More
Feedback?
Help
Log Out