I can see where this post is pointing. But I find myself disagreeing. Say they replace the ticket machines at a train station. The new machines that have touch screens. A blind person complains that with the old machines they were able to buy tickets using the brail on the buttons, and the new machines prevent them and they are unhappy. Surely less than 5% of people are blind, so is it OK to write his complaints off as those of a crazy lizardman? The new machines may or may not be net-positive, but its clear that the impact they have had on that individual is a real negative to be really weighed up.
Maybe you trust yourself to tell the difference between the blind and the lizardmen. But I would ask why you should bother. If a person’s requests can be satisfied easily without much harm to anyone else then surely they should be, regardless of whether they seem maniac or not. And when we find ourselves in a adversarial situation where only one group or another can get what they want then we can do our best to weigh up the best decision for the greatest number. But we can do all that without stamping the losers with the “crazy lizardman” tag.
The building I work in has strange neon-tube lights which (I believe) flicker at a high frequency. They give me headaches. Lets say that in the future I raised this issue and asked for something to be done. If my employer told me that it was too expensive to change the lights for just one person, that would be fair. If they instead told me that I was a crazy lizardman and it wasn’t even worth considering a cost-benefit analysis to help a dumb lizardman, then that would not be fair. Indeed, if they wanted me to accept the decision the first response would be more effective.
A long winded way of saying: Do your utils multiplying, and act accordingly. But their is no need to pretend the negative utils with the smaller factor don’t exist, which I worry is what this post seems to be gesturing at.
I think that Duncan’s post implicitly is about complaints about issues we haven’t agreed as a society are “valid minority issues.”
We conventionally accept that disability issues are valid complaints even though only a small minority have any specific disability. It’s also conventionally fine to complain about things like flickering lights, just like it would be valid to complain if the bright sun shone directly into your eyes from a certain angle and you wanted to close the blinds.
Let’s say, on the other hand, that you didn’t like the color of your neighbor’s house. It is painted green, and you HATE green. Your neighbor would be justified in just ignoring you flat out, because it’s typically not valid to complain about the color of your neighbor’s house (at least not with the intent to get them to change it) unless there’s some kind of covenant.
By contrast, if your neighbor installed a tin roof that shone bright sun through your windows all day long and made it uncomfortable to be in your house, you’d have cause to complain (I’ve seen this before).
So I really do think we have valid complaint categories and invalid complaint categories, and the “Lizardman” concept does seem to encapsulate something of how we treat people making inappropriate/invalid complaints.
As another example of Lizardman, there’s a guy on my block who sometimes stands on the corner with a radar gun and screams at bicyclists who he thinks are riding too fast. He also screams at cars that idle on the side of the road for too long. He is another good example of Lizardman—I would never involve myself in hearing him out in order to decide whether his complaints have merit, and I don’t think anybody else needs to either. I think he needs therapy.
The phrase “valid minority issues” implies that their are “invalid minority issues”, which are issues that it has been decided are not real issues and should never be discussed. So how would anything ever (even in principle) move from the invalid category to the valid category? If it cannot be talked about their is no way to even have the discussion.
If medicine did hypothetically discover one day an exact shade of green that caused agonizing pain to 0.1% of the population when they saw it then I would want polices that limited the use of that shade of green for clothes and houses.
To me the cost/benefit analysis looks like this. First, downsides (-)’s.
(-) By dismissing the point without hearing the person out we might put something in the wrong bucket. (eg. “this crazy person has emailed because they hate touch screens. What a lizardman—oops, I missed the part where they said they were blind.”)
(-) We create dangerous argumentative incentives to portray our opposition as lizardmen, rather than actually putting forward a good argument against them. Because once we have them painted as lizardmen they have already lost and no one will listen to them.
(-) That one time that something really obscure is a problem for us we have to just live with it. (Does worrying about AI safety make you a lizardman? Certainly not on LW, but maybe elsewhere in the world it does)
(-) Maybe it is an easy problem to solve: the person who hates the green house is happy to pay for it to be repainted, and the house owner would prefer it was painted red but couldn’t justify the cost.
Upsides:
(+) We can save some time. That is a real saving. But I don’t know how much we save. Being ignored isn’t going to make the guy on your block chill out and stop shouting at you.
The phrase “valid minority issues” implies that their are “invalid minority issues”, which are issues that it has been decided are not real issues and should never be discussed. So how would anything ever (even in principle) move from the invalid category to the valid category? If it cannot be talked about their is no way to even have the discussion.
In your “agonizing shade of green” hypothetical, I think you’ve hit the nail on the head—the mechanism for resolving the issue is for medical providers to notice a pattern of complaints among patients about the suffering they experience from this agonizing shade of green, gather enough evidence that we can update from “this seems really unlikely to be a real issue” to “this is probably actually a serious issue for a minority of people and worth doing something about” and then creating a new social norm in which complaining about that particular shade of green is valid by default.
At some point, you’re trading off between the right of minority populations to have their rare but genuine concerns addressed, vs. the cost and side effects of fielding recreational whining and made-up complaints that service a deeper agenda. This is a fuzzy grey zone and there will be limits and tradeoffs to how sensitive and responsive we are in either direction.
Separately, I used to love the lizardmen monsters in Warhammer as a child. Here’s the one on your block (and you have my sympathies, he sounds like a real annoyance):
I can see where this post is pointing. But I find myself disagreeing. Say they replace the ticket machines at a train station. The new machines that have touch screens. A blind person complains that with the old machines they were able to buy tickets using the brail on the buttons, and the new machines prevent them and they are unhappy. Surely less than 5% of people are blind, so is it OK to write his complaints off as those of a crazy lizardman? The new machines may or may not be net-positive, but its clear that the impact they have had on that individual is a real negative to be really weighed up.
Maybe you trust yourself to tell the difference between the blind and the lizardmen. But I would ask why you should bother. If a person’s requests can be satisfied easily without much harm to anyone else then surely they should be, regardless of whether they seem maniac or not. And when we find ourselves in a adversarial situation where only one group or another can get what they want then we can do our best to weigh up the best decision for the greatest number. But we can do all that without stamping the losers with the “crazy lizardman” tag.
The building I work in has strange neon-tube lights which (I believe) flicker at a high frequency. They give me headaches. Lets say that in the future I raised this issue and asked for something to be done. If my employer told me that it was too expensive to change the lights for just one person, that would be fair. If they instead told me that I was a crazy lizardman and it wasn’t even worth considering a cost-benefit analysis to help a dumb lizardman, then that would not be fair. Indeed, if they wanted me to accept the decision the first response would be more effective.
A long winded way of saying: Do your utils multiplying, and act accordingly. But their is no need to pretend the negative utils with the smaller factor don’t exist, which I worry is what this post seems to be gesturing at.
I think that Duncan’s post implicitly is about complaints about issues we haven’t agreed as a society are “valid minority issues.”
We conventionally accept that disability issues are valid complaints even though only a small minority have any specific disability. It’s also conventionally fine to complain about things like flickering lights, just like it would be valid to complain if the bright sun shone directly into your eyes from a certain angle and you wanted to close the blinds.
Let’s say, on the other hand, that you didn’t like the color of your neighbor’s house. It is painted green, and you HATE green. Your neighbor would be justified in just ignoring you flat out, because it’s typically not valid to complain about the color of your neighbor’s house (at least not with the intent to get them to change it) unless there’s some kind of covenant.
By contrast, if your neighbor installed a tin roof that shone bright sun through your windows all day long and made it uncomfortable to be in your house, you’d have cause to complain (I’ve seen this before).
So I really do think we have valid complaint categories and invalid complaint categories, and the “Lizardman” concept does seem to encapsulate something of how we treat people making inappropriate/invalid complaints.
As another example of Lizardman, there’s a guy on my block who sometimes stands on the corner with a radar gun and screams at bicyclists who he thinks are riding too fast. He also screams at cars that idle on the side of the road for too long. He is another good example of Lizardman—I would never involve myself in hearing him out in order to decide whether his complaints have merit, and I don’t think anybody else needs to either. I think he needs therapy.
The phrase “valid minority issues” implies that their are “invalid minority issues”, which are issues that it has been decided are not real issues and should never be discussed. So how would anything ever (even in principle) move from the invalid category to the valid category? If it cannot be talked about their is no way to even have the discussion.
If medicine did hypothetically discover one day an exact shade of green that caused agonizing pain to 0.1% of the population when they saw it then I would want polices that limited the use of that shade of green for clothes and houses.
To me the cost/benefit analysis looks like this. First, downsides (-)’s.
(-) By dismissing the point without hearing the person out we might put something in the wrong bucket. (eg. “this crazy person has emailed because they hate touch screens. What a lizardman—oops, I missed the part where they said they were blind.”)
(-) We create dangerous argumentative incentives to portray our opposition as lizardmen, rather than actually putting forward a good argument against them. Because once we have them painted as lizardmen they have already lost and no one will listen to them.
(-) That one time that something really obscure is a problem for us we have to just live with it. (Does worrying about AI safety make you a lizardman? Certainly not on LW, but maybe elsewhere in the world it does)
(-) Maybe it is an easy problem to solve: the person who hates the green house is happy to pay for it to be repainted, and the house owner would prefer it was painted red but couldn’t justify the cost.
Upsides:
(+) We can save some time. That is a real saving. But I don’t know how much we save. Being ignored isn’t going to make the guy on your block chill out and stop shouting at you.
I’ll refer you to another comment of mine for my opinion on this issue.
In your “agonizing shade of green” hypothetical, I think you’ve hit the nail on the head—the mechanism for resolving the issue is for medical providers to notice a pattern of complaints among patients about the suffering they experience from this agonizing shade of green, gather enough evidence that we can update from “this seems really unlikely to be a real issue” to “this is probably actually a serious issue for a minority of people and worth doing something about” and then creating a new social norm in which complaining about that particular shade of green is valid by default.
At some point, you’re trading off between the right of minority populations to have their rare but genuine concerns addressed, vs. the cost and side effects of fielding recreational whining and made-up complaints that service a deeper agenda. This is a fuzzy grey zone and there will be limits and tradeoffs to how sensitive and responsive we are in either direction.
Separately, I used to love the lizardmen monsters in Warhammer as a child. Here’s the one on your block (and you have my sympathies, he sounds like a real annoyance):
This is probably how he perceives himself :D