An alternative to the blender framing:
Everyone must press a red or blue button. If you press red you will survive. If you press blue you will survive if, and only if, more than 50% of people press blue. Everyone receives the same instructions.
I suspect that with that framing the red vote would increase significantly.
If we are talking decision theory may I suggest the following way of looking at it.
Let’s say 101 participants with 50% threshold. Your individual decision has the following effect depending on how the other 100 vote.
If 0-49 others vote red then your vote has no effect as majority will be blue however you vote.
If 51-100 others vote red then you voting blue kills 1 extra person (you!) compared to voting red.
If exactly 50 others vote red then you voting red kills 50 extra people (all the blue voters).
So, there are 50 scenarios where you can save 1 life by voting red and 1 scenario where you can save 50 lives by voting blue.
You decision therefore you be based on:
a) whether you value you own life more than others in the game (technically how much you value your life compared to the average blue voter!)
b) how you expect others to vote. If you expect more blue voters it should move you towards voting blue as the 50:50 scenario is more likely than each of the individual 51-100 red scenarios. Likewise, if you expect more reds it should shift you towards red.
So now it looks very like the 2 player version—a coordination game with the possibility of guaranteeing personal survival.
I have a friend who is a dermatologist and their main advice on sunscreen is to use high SPF because almost no-one applies it right. If you use SPF50+ then you’ll still get good protection when you inevitably do it wrong!
Lesswrong needs to be monetised. Every post is sponsored by a relevant partner.
Sometimes dishonesty is the right thing?
In this post John describes a method by which functioning democracies can attempt to prevent tyranny of the majority—giving each major faction a de-facto veto over new legislation.
Whilst this is a method used in some countries, it is by no means the only, or indeed the most common, method for achieving this. I am only properly familiar with the UK as a counter-example but Dumbledore’s army lists Frace, Germany, Italy and Canada as some others.
The method described in the post is likely more useful in situations where there are 2-3 major factions, whose values are incompatible with each other (slavery in pre civil-war US, religion/ethnic background in Iraq, religion/national identity in Northern Ireland—an example mentioned by Michael Roe). In Northern Ireland, for example, you cannot both want NI to be part of independant Ireland and want it to be part of the UK and you are very unlikely to change your position on the topic. Similarly slave vs non-slave states in pre civil-war america.
If such countries can maybe be thought of as bimodal (or trimodal) distributions of beliefs, other countries can be imagined as having a normal, or at least a relatively flat distribution.
In situations without a solid dichotomy, other methods for preventing tyranny of the majority are used. Some countries have regular coalition building (e.g. Germany), others are just kinda ok with regular opportunities to vote for a new leader. Possibly these solutions are less resilient if the country implementing them becomes more internally polarised and the de facto veto would be a useful backstop but I’m not sure.
Unfortunately this post doesn’t tackle these issues, instead focussing only on the single solution. To improve this post it would need to consider what different methods are used and the situations which make them more or less useful.
Footnote: Even the example of the US isn’t so obvious. Historically (before Nixon) the president and the 2 houses were united about 2/3rds of the time. More recently, power has been shared, but there is no tendency towards house or senate between Democrats & Republicans—each party has won each about 50% of the time since 1981. Before that Democrats controlled both for all but 4 of 48 years.
The main reasons I didn’t use that title were that the piece I’m responding to doesn’t (Lack of social grace is also not always epistemic virtue! Sometimes it’s pure unforced error!)
IUUC, the “Epistemic” in the title of LoSGiaEV is intended as the qualifier—LoSG isn’t an unqualified virute. In the first 3 paragraphs LoSGiaEV lays out what it means by that qualifier. I feel like LoSGiaLoS would be improved if it engaged more with the nuances.
I assume you mean “having the potential to become disastrous; at a point of crisis”
Yes
“Hey moron there’s a car coming” is worse than “There’s a car coming” right?
Yes, but both are worse than just shouting “CAR”. Being shouted at feels bad, along with the implications that you weren’t being careful enough / are a bad driver / are stupid for not looking. When I was younger I was in a car with someone who I respected who was texting as they drove. I saw a car coming but was initially socially nervous and tried to politely get their attention. This didn’t work and in the end I had to shout at them. They felt pretty bad about it but it worked.
Or take expressing disapproving comments or judgements; I’ve often seen people give feedback that only points out one or two flaws with no further commentary. I later learned that the person giving feedback loved most of the piece.
I think this should mainly be orthogonal to social grace. In my initial comment I deliberately put in the bits I agreed with, not because I wanted to be socially gracious but because I wanted to accurately portray what I thought.
There’s a great scene where a scientist is in a room full of people congratulating themselves on the Chernobyl disaster being not that big a deal and being well in hand, and stands up to say things are not fine.
Unfortunately I can’t view the video because it’s copyright blocked. It sounds like a good example of when being socially gracious is useful.
My counterpoint would be to consider what culture led to the scientist having to waste his time on multiple explanations. Should a less socially adept scientist have been ignored?
In pushing the culture of LessWrong we should be wary of such problems. My personal feeling is that LessWrong at the moment is in about the right place on the “Gracious” vs “Honest” pareto front for a given social skill level.
Tying this in with the car example, my wife and I have an agreement when one of us driving. If the passenger sees something dangerous that they think the driver might not have seen, they’re allowed to shout this with no concern for social grace and the driver should not take this as a personal slight.
I feel like I agree with almost everything that is said here, while finding myself concerned about the overall feeling of the post and the effect on LW if applied as is.
I think this stems from the title which isn’t quite what the article argues for.
I think that the article persuasively argues that:
The play off between social grace and honesty can change with skill at social grace
With high social grace skill the cost of being socially graceful is lower than you might think
In many circumstances being socially graceful is worth it
This doesn’t quite add up to Lack of Social Grace is a Lack of Skill.
This is especially true in critical situations. If I acknowledge that being socially graceful creates even a slight loss of honesty/accuracy then this could have very large costs if the implication of the decision is critical enough.
If LW conversations are intended to be more often about critical situations then we should hope that our conversations are closer to the bottom right of our own Pareto front than the top left.
I think without specifically acknowledging this and with the title as it is the post could easily end up pointing people in the wrong direction.
One of my favourite LW post types is “something that I had a vague sense of being identified and fleshed out”
This is a great example of that.
In my experience, people thinking badly about me isn’t upstream of the social punishment of losing resources, people thinking badly of me IS the social punishment.
Evolutionarily speaking I would guess that worrying about what people think of me is important in order to allow me to get resources but my emotions are implemented on the social level, not the resource level.
(E.g. I don’t know any LWers IRL and will get very little resource from them but it would still feel bad if people had a bad opinion of me)
The Co-Op’s purchase of Somerfield (£1.57b) would seem like an interesting case study—why did they decide to go ahead with this? Did it actually benefit staff in the long run?
Yeah, I think that clarifies my thoughts—IMO using the word purpose is not ok in these other perspectives unless you actually mean a person had that purpose. it brings in connotations that someone has this purpose for which we then assign moral blame when the real question may be competence. And then, when there is moral blame, separating that from a competence issue is harder.
Much better IMO to say “this system isn’t achieving its purpose” than to say “this system has this other purpose” unless that’s what you mean—you are claiming that some designer(s) have this other purpose.
I’m not familiar specifically with HSR but my guess is that there are multiple purposes from multiple people. Some designers have the purpose to improve transport links, some want a big project to make their name and some want a jobs program. Systems can have multiple purposes and trying to narrow it down to a single one oversimplifies things.
I don’t think this works for general systems. For POSIWID the purpose of a car is to to get you for A to B. I don’t think POSIWIR returns a result—cars aren’t rewarded.
Having said that, I do think it’s much more interesting than POSIWID and, for human systems, more likely to yield sensible answers.
My take would be that purpose comes from conscious minds—the purpose of a system is what the designer intends. This is true even if the designer is rubbish at their job and their system doesn’t achieve the purpose they intend. When we don’t know what the designer intends we try to work it out by looking at what the system does or what it rewards.
This declares by fiat that non-engineered systems have no purpose but I think I’m ok with that. The purpose of a rabbit is not to make baby rabbits—rabbits just are. Saying rabbits have a purpose is to anthropomorphise evolution.
Quick note that I can’t open the webpage via my institution (same issue on multiple browsers). Their restrictions can be quite annoying and get triggered alot. I can view it myself easily enough on phone but if you want this to get out beware trivial inconveniences and all that...
Firefox message is below.
An error occurred during a connection to ifanyonebuildsit.com. Cannot communicate securely with peer: no common encryption algorithm(s).
Error code: SSL_ERROR_NO_CYPHER_OVERLAP
The page you are trying to view cannot be shown because the authenticity of the received data could not be verified.
Please contact the web site owners to inform them of this problem.
Do dragon unbelievers accept this stance? My impression is that dragon agnosticism would often be considered almost as bad as dragon belief.
I’m confused as to how the fits in with UK politics. I don’t think the minority party has any kind of veto?
I guess we have the House of Lords but this doesn’t really have a veto (at least not long term) and the House of Commons and House of Lords aren’t always or even usually controlled by different factions.
One extra thing to consider financially is if you have a smart meter then you can get all of your hot water and a chunk of your heating done at off peak rates. Our off peak electricity rates are about equal per kWh to gas rates.
Without this I think our system would be roughly the same cost per year as gas or slightly more, with it we save £200 per year or so I think. (This would be a very long payback time but there was a fully funded scheme we used).
If it helps anyone we are in Scotland and get average COP=2.9
In the UK there is a non-binding but generally observed rule that speed cameras allow you to drive 10% + 2mph above the speed limit(e.g. 35mph in a 30mph zone) before they activate.
This is a bit more of a fudge but better than nothing.
Someone in your company gets fired by a boss you don’t know/particularly like without giving any reason
You are mad with the boss and want the decision overturned
You have a credible, attractive BATNA (the Microsoft offer)
These 3 items seem like they would be sufficient to cause something like the Open Letter to happen.
In most cases number 3 is not present which I think is why we don’t see things like this happen more often in more organisations.
None of this requires Sam to be hugely likeable or a particularly savvy political operator, just that people generally like him. People seem to suggest he was one or both so this just makes the letter more likely.
I’m sure this doesn’t explain it all in OpenAI’s case—some/many employees would also have been worried about AI safety which complicates the decision—but I suspect it is the underlying story.
This is an interesting result. Gemini would perhaps approve of Kirk’s approach to the Kobayashi Maru?
One thing which would increase my confidence in the conclusion would be to split the “Low signs of awareness” into the 4 categories of eval. At the moment it isn’t quite clear whether the difference in behaviour in the different eval categories is due to:
Gemini realising it is being evaluated or
Gemini being better aligned for certain eval categories