I have a specific drawback in mind that with probability .88 cancels out any benefits for the vast majority of potential users.
What drawback did you have in mind? I’m confused why you would allude to this without specifying what it was, and curious as to whether it’s one I’ve already thought of and tried to mitigate, one that I’ve thought of but can’t mitigate, or one that I haven’t thought of.
I will explain the drawback I have in mind if you answer my two questions in grandparent to my satisfaction.
My reason for holding back on the explanation of the drawback is that it will require an appeal to one or two highly technical concepts (from the field of neuropsychology roughly speaking), and I know you well enough by now to know that you will probably latch onto the highly technical concepts and refuse to write or think about the only part of this thread of conversation that has any significant interest to me: namely, the fact that, unless I am extremely mistaken about your plan, you already have enough information to know that your plan of creating yet another RSVP software is a waste of time even if I had made no mention of any drawback. I predict that you will persist in not updating on this information you already have.
You seem to be under the impression that I’m implementing naive RSVP with a slightly improved UI. I’ve made improvements that I haven’t told you about, and I’m experimenting with different variations on the concept. I believe that people don’t persist with RSVP because the software is bad—in addition to the inconvenience of getting started, they periodically force task loads into timeslices where they don’t fit, which derails the reader badly. I wrote my own software because I believe I’ve found ways to reduce how often that happens, and to reduce the disruption when it does.
If you think you know something about RSVP that I don’t, just say it. Don’t play games.
I would rather just drop it because grandparent’s score is currently at minus 3, and by convention downvotes mean, “I don’t want to see more like this”.
(Upvoted back up to 0 because this is an entirely reasonable response! Why on earth would someone be expected to want to continue a conversation in which they were getting downvoted?)
Why on earth would someone be expected to want to continue a conversation in which they were getting downvoted?
You may not have read the context of that comment. Here’s how the conversation went, from my perspective:
rhollerith: Hey, I know something very important about the project you’re working on right now. Me: And what’s that? rhollerith: Not telling. You’d just ignore me anyways. (-2) Me: Don’t play games. ... Me: Are you going to follow up on this? rhollerith: I’d like to drop this conversation now because I was downvoted.
Needless to say, I am quite unhappy with how this conversation went. But ending it there would make it much worse, since I don’t know whether rhollerith actually noticed something important, or is defecting by accident, or if I offended him somehow and forgot about it.
You may not have read the context of that comment.
I read the context but just don’t object to rhollerith exiting the conversation. You’ve already had your chance to punish him (via vote and social positioning) for what you perceive to be not paying you your due respect and it isn’t necessarily wise for him to make himself vulnerable to another salvo if he anticipates an aggressive response. Even paying the courtesy of responding to “are you going to continue?” was probably a mistake since it allowed you to draw attention to his earlier faux pas. Either don’t respond at all or respond with something thoroughly polished.
Basically I am reluctant to expect people to do things that are clearly bad for them unless I have a damn good reason to. (Even if this is a rather trivial instance with respect to degree of ‘demand’ and ‘bad for’.)
since I don’t know whether rhollerith actually noticed something important, or is defecting by accident
In rhollerith’s story (if I am modelling him correctly) it would be you who was defecting by not answering the questions he had already asked. His mistake was that by social convention it is usually ok to ignore an adversary’s questions, especially if witnesses aren’t paying close attention. But including “to my satisfaction” was never going to be received well, even by a casual observer—it is a defection that just seems banal and almost cringe-worthy. He could have drawn attention back to his questions in a more effective manner. Perhaps by first giving the example and then hammering down the questions again.
In rhollerith’s story (if I am modelling him correctly) it would be you who was defecting by not answering the questions he had already asked. His mistake was that by social convention it is usually ok to ignore an adversary’s questions, especially if witnesses aren’t paying close attention.
No, the mistake was that I wasn’t actually interested in convincing him (or anyone) of anything—I was only trying to use the thread to collect ideas.
No, the mistake was that I wasn’t actually interested in convincing him (or anyone) of anything—I was only trying to use the thread to collect ideas.
When I speak of rhollerith’s mistake I speak of the deviation of his actions from an optimal path for achieving his goals. In this case particularly with respect to social goals. I didn’t speak of where he violated your expectations or preferences be they professed or actual (or both). Having an accurate model of you would be relatively insignificant (and entirely redundant) when it came to predicting how his “to my satisfaction” comment would be received in the social context.
Looks to me like he was willing to talk about what you wanted to talk about if you were willing to talk about what he wanted to talk about. Unless I’m missing one of your comments, you didn’t acquiesce to his request, so he didn’t acquiesce to yours.
Parent is right: I wanted to extract some sort of concession or token act of reciprocity from him because I have been feeling that my efforts on LW are unappreciated. I considered extracting a promise that he would email me not more than 3 months from now and not less than 4 weeks from now with a report on where his work on RSVP has led, but I settled on, “first answer my questions, and then I will answer yours”.
I have an intellectual commitment to trying to be helpful to people who are trying to improve the world and who are not destructive or irresponsible about it (e.g., who are not trial-and-erroring AGI designs) but lately I have been having trouble summoning the actual short-term motivation to do so. Either upvotes or some indication from the person I am trying to help that he considers me more than just an information-supplying machine whould probably have provided that short-term motivation in the present case.
It is no big deal because this period during which I do not feel sufficiently good about myself and my status to continue to attempt small thankless good deeds on the internet will probably pass in a few months. Alternatively I will replace thankless good deeds on the net with some other way for me to try to improve the world.
Speaking of thanks, I hereby thank wedrifid for his analysis. In my defense, I knew the downvoted comment was unsatisfactory, and I was in the process of rewriting it from scratch when I noticed I had accidentally posted it.
I have PMed jimrandomh the information he wanted. (800 words of it.) He is sincerely trying to improve the world and is doing so in a responsible non-destructive way, so I want to help him for some value of “want”.
That’s too general a question to answer in a reasonable amount of time!
One way to limit the question is to replace it with this next:
Do you think this is specific to you, or is it just that trying to improve the world by pointing out errors in comments on LW is less rewarding than it ought to be?
Would you like an answer to the more specific question?
Yes, I’d be interested to know if you think your own efforts on LW are less appreciated than other people’s, and why.
Also, I would tend to think that pointing out errors in LW comments is one of the better deals currently available in terms of social reward per unit of effort. I’m curious what you had in mind when you said “replace … with some other way for me to try to improve the world”.
I’d be interested to know if you think your own efforts on LW are less appreciated than other people’s.
If you are asking whether I believe my status here or my karma score is unfairly low, the answer is no.
When I wrote in great great grandparent that “I have been feeling that my efforts on LW are unappreciated,” I meant that my karma score and my perception of my own status here is insufficiently high to continue to motivate me to point out errors here, give thoughtful answers to questions here, etc.
Parenthetically, I would probably have an easier time motivating myself to continue participating on LW if LW were not growing with the result that comments by people I have gotten to know and like, such as jimrandomh and yourself, are mixed in with many comments by people I know nothing about. Of course there is significant global expected utility in attracting many new people to LW that trumps my personal preferences. ADDED. “Continuing to participate” was the wrong phrasing because I will almost certainly keep on using LW to help me learn. The thing I am having trouble motivating myself to continue to do (partly because there’s no objective indication I’m doing significant good) is to use LW help others learn, help others realize there is a flaw in their plan to improve the world, etc.
I’m curious what you had in mind when you said “replace … with some other way for me to try to improve the world”.
Well, I could spend the time I would have spent trying to help people on LW on open-source software development. I am confused about what you are after with this question, so I won’t say more yet.
I would tend to think that pointing out errors in LW comments is one of the better deals currently available in terms of social reward per unit of effort
When I started to do it, I believed that it is one of the better deals in terms of expected increase in global utility per unit of my effort, but the lowness of the rate at which I am being upvoted contradicts that original belief, and it would be unwise of me to ignore that metric of how effective I am at achieving what I set out to achieve given that I can find other ways of improving the world that also have good metrics attached to them that I might do better at.
Interesting. There seem to be two separate issues here: karma as motivation, and karma as metric of effectiveness.
Re: karma as motivation. Since you’ve been here from the OB days and have been sufficiently motivated to participate for the past several years, I suggest that the problem isn’t your absolute karma or the rate of increase of karma, but habituation of your dopaminergic motivation system to the amount of karma you get per comment. I suggest (partly from theory and partly from personal experience) that if you go away from LW for a few months (or maybe just a few weeks), the habituation will reverse itself, and you’ll be sufficiently motivated to participate again.
Re: karma as metric of effectiveness. This seems a much more difficult problem. Karma is a dimensionless quantity. We know that the sign of your total karma is positive, which suggests that you’re doing more good than harm on LW, but how can we convert a karma value into an estimate of how much good you’re doing, and compare that to how much good you might do elsewhere? You seem to have some intuition about that, but I’m not sure what it’s based on. (Could it just be the affect heuristic? I.e., your habituated dopaminergic system makes you feel bored about participating on LW, and you translate that into a low estimate of effectiveness?) This might be a good question for a discussion post, unless there’s an obvious solution that I’m missing...
Speaking of thanks, I hereby thank wedrifid for his analysis. In my defense, I knew the downvoted comment was unsatisfactory, and I was in the process of rewriting it from scratch when I noticed I had accidentally posted it. … I have PMed jimrandomh the information he wanted. (800 words of it.) He is sincerely trying to improve the world and is doing so in a responsible non-destructive way, so I want to help him for some value of “want”.
Ok, now I feel bad about having been dickish about it. Sorry for being so impatient! I got your email. I’ve got some followup research to do before I respond to it at length, which might take a few days, but that was valuable. Thank you.
The current status is that I have a reasonably good implementation, and am setting up server-side stuff to market and sell, with a 30-day free trial version. Whether people like it and stick with it will ultimately be revealed by the free-to-paid conversion rate.
Thanks for the kind words. Part of the reason this thread did not go smoothly is that my communication skills and helping skills are not where they probably need to be for me to be effective at consistently helping LWers over the internet. Heck, I still have trouble consistently writing a short email to my ex-girlfriend that does not inadvertently cause her to get upset (and stay upset for more than a day).
What drawback did you have in mind? I’m confused why you would allude to this without specifying what it was, and curious as to whether it’s one I’ve already thought of and tried to mitigate, one that I’ve thought of but can’t mitigate, or one that I haven’t thought of.
I will explain the drawback I have in mind if you answer my two questions in grandparent to my satisfaction.
My reason for holding back on the explanation of the drawback is that it will require an appeal to one or two highly technical concepts (from the field of neuropsychology roughly speaking), and I know you well enough by now to know that you will probably latch onto the highly technical concepts and refuse to write or think about the only part of this thread of conversation that has any significant interest to me: namely, the fact that, unless I am extremely mistaken about your plan, you already have enough information to know that your plan of creating yet another RSVP software is a waste of time even if I had made no mention of any drawback. I predict that you will persist in not updating on this information you already have.
You seem to be under the impression that I’m implementing naive RSVP with a slightly improved UI. I’ve made improvements that I haven’t told you about, and I’m experimenting with different variations on the concept. I believe that people don’t persist with RSVP because the software is bad—in addition to the inconvenience of getting started, they periodically force task loads into timeslices where they don’t fit, which derails the reader badly. I wrote my own software because I believe I’ve found ways to reduce how often that happens, and to reduce the disruption when it does.
If you think you know something about RSVP that I don’t, just say it. Don’t play games.
Are you going to follow up on this?
I would rather just drop it because grandparent’s score is currently at minus 3, and by convention downvotes mean, “I don’t want to see more like this”.
The downvotes are for the “I will only explain if you answer my questions first” thing. If you have insights about RSVP, I do want to hear them.
(Upvoted back up to 0 because this is an entirely reasonable response! Why on earth would someone be expected to want to continue a conversation in which they were getting downvoted?)
You may not have read the context of that comment. Here’s how the conversation went, from my perspective:
rhollerith: Hey, I know something very important about the project you’re working on right now.
Me: And what’s that?
rhollerith: Not telling. You’d just ignore me anyways. (-2)
Me: Don’t play games.
...
Me: Are you going to follow up on this?
rhollerith: I’d like to drop this conversation now because I was downvoted.
Needless to say, I am quite unhappy with how this conversation went. But ending it there would make it much worse, since I don’t know whether rhollerith actually noticed something important, or is defecting by accident, or if I offended him somehow and forgot about it.
I read the context but just don’t object to rhollerith exiting the conversation. You’ve already had your chance to punish him (via vote and social positioning) for what you perceive to be not paying you your due respect and it isn’t necessarily wise for him to make himself vulnerable to another salvo if he anticipates an aggressive response. Even paying the courtesy of responding to “are you going to continue?” was probably a mistake since it allowed you to draw attention to his earlier faux pas. Either don’t respond at all or respond with something thoroughly polished.
Basically I am reluctant to expect people to do things that are clearly bad for them unless I have a damn good reason to. (Even if this is a rather trivial instance with respect to degree of ‘demand’ and ‘bad for’.)
In rhollerith’s story (if I am modelling him correctly) it would be you who was defecting by not answering the questions he had already asked. His mistake was that by social convention it is usually ok to ignore an adversary’s questions, especially if witnesses aren’t paying close attention. But including “to my satisfaction” was never going to be received well, even by a casual observer—it is a defection that just seems banal and almost cringe-worthy. He could have drawn attention back to his questions in a more effective manner. Perhaps by first giving the example and then hammering down the questions again.
No, the mistake was that I wasn’t actually interested in convincing him (or anyone) of anything—I was only trying to use the thread to collect ideas.
When I speak of rhollerith’s mistake I speak of the deviation of his actions from an optimal path for achieving his goals. In this case particularly with respect to social goals. I didn’t speak of where he violated your expectations or preferences be they professed or actual (or both). Having an accurate model of you would be relatively insignificant (and entirely redundant) when it came to predicting how his “to my satisfaction” comment would be received in the social context.
Looks to me like he was willing to talk about what you wanted to talk about if you were willing to talk about what he wanted to talk about. Unless I’m missing one of your comments, you didn’t acquiesce to his request, so he didn’t acquiesce to yours.
Parent is right: I wanted to extract some sort of concession or token act of reciprocity from him because I have been feeling that my efforts on LW are unappreciated. I considered extracting a promise that he would email me not more than 3 months from now and not less than 4 weeks from now with a report on where his work on RSVP has led, but I settled on, “first answer my questions, and then I will answer yours”.
I have an intellectual commitment to trying to be helpful to people who are trying to improve the world and who are not destructive or irresponsible about it (e.g., who are not trial-and-erroring AGI designs) but lately I have been having trouble summoning the actual short-term motivation to do so. Either upvotes or some indication from the person I am trying to help that he considers me more than just an information-supplying machine whould probably have provided that short-term motivation in the present case.
It is no big deal because this period during which I do not feel sufficiently good about myself and my status to continue to attempt small thankless good deeds on the internet will probably pass in a few months. Alternatively I will replace thankless good deeds on the net with some other way for me to try to improve the world.
Speaking of thanks, I hereby thank wedrifid for his analysis. In my defense, I knew the downvoted comment was unsatisfactory, and I was in the process of rewriting it from scratch when I noticed I had accidentally posted it.
I have PMed jimrandomh the information he wanted. (800 words of it.) He is sincerely trying to improve the world and is doing so in a responsible non-destructive way, so I want to help him for some value of “want”.
Do you think this is specific to you, or is it just that trying to improve the world is in general less rewarding than it ought to be?
That’s too general a question to answer in a reasonable amount of time!
One way to limit the question is to replace it with this next:
Do you think this is specific to you, or is it just that trying to improve the world by pointing out errors in comments on LW is less rewarding than it ought to be?
Would you like an answer to the more specific question?
Yes, I’d be interested to know if you think your own efforts on LW are less appreciated than other people’s, and why.
Also, I would tend to think that pointing out errors in LW comments is one of the better deals currently available in terms of social reward per unit of effort. I’m curious what you had in mind when you said “replace … with some other way for me to try to improve the world”.
If you are asking whether I believe my status here or my karma score is unfairly low, the answer is no.
When I wrote in great great grandparent that “I have been feeling that my efforts on LW are unappreciated,” I meant that my karma score and my perception of my own status here is insufficiently high to continue to motivate me to point out errors here, give thoughtful answers to questions here, etc.
Parenthetically, I would probably have an easier time motivating myself to continue participating on LW if LW were not growing with the result that comments by people I have gotten to know and like, such as jimrandomh and yourself, are mixed in with many comments by people I know nothing about. Of course there is significant global expected utility in attracting many new people to LW that trumps my personal preferences. ADDED. “Continuing to participate” was the wrong phrasing because I will almost certainly keep on using LW to help me learn. The thing I am having trouble motivating myself to continue to do (partly because there’s no objective indication I’m doing significant good) is to use LW help others learn, help others realize there is a flaw in their plan to improve the world, etc.
Well, I could spend the time I would have spent trying to help people on LW on open-source software development. I am confused about what you are after with this question, so I won’t say more yet.
When I started to do it, I believed that it is one of the better deals in terms of expected increase in global utility per unit of my effort, but the lowness of the rate at which I am being upvoted contradicts that original belief, and it would be unwise of me to ignore that metric of how effective I am at achieving what I set out to achieve given that I can find other ways of improving the world that also have good metrics attached to them that I might do better at.
Finally, I declare Crocker’s rules.
Interesting. There seem to be two separate issues here: karma as motivation, and karma as metric of effectiveness.
Re: karma as motivation. Since you’ve been here from the OB days and have been sufficiently motivated to participate for the past several years, I suggest that the problem isn’t your absolute karma or the rate of increase of karma, but habituation of your dopaminergic motivation system to the amount of karma you get per comment. I suggest (partly from theory and partly from personal experience) that if you go away from LW for a few months (or maybe just a few weeks), the habituation will reverse itself, and you’ll be sufficiently motivated to participate again.
Re: karma as metric of effectiveness. This seems a much more difficult problem. Karma is a dimensionless quantity. We know that the sign of your total karma is positive, which suggests that you’re doing more good than harm on LW, but how can we convert a karma value into an estimate of how much good you’re doing, and compare that to how much good you might do elsewhere? You seem to have some intuition about that, but I’m not sure what it’s based on. (Could it just be the affect heuristic? I.e., your habituated dopaminergic system makes you feel bored about participating on LW, and you translate that into a low estimate of effectiveness?) This might be a good question for a discussion post, unless there’s an obvious solution that I’m missing...
Ok, now I feel bad about having been dickish about it. Sorry for being so impatient! I got your email. I’ve got some followup research to do before I respond to it at length, which might take a few days, but that was valuable. Thank you.
The current status is that I have a reasonably good implementation, and am setting up server-side stuff to market and sell, with a 30-day free trial version. Whether people like it and stick with it will ultimately be revealed by the free-to-paid conversion rate.
Thanks for the kind words. Part of the reason this thread did not go smoothly is that my communication skills and helping skills are not where they probably need to be for me to be effective at consistently helping LWers over the internet. Heck, I still have trouble consistently writing a short email to my ex-girlfriend that does not inadvertently cause her to get upset (and stay upset for more than a day).