Michael_PJ

Karma: 2,526

Michael_PJ 18 Sep 2014 22:21 UTC
22 points
0 ∶ 0
in reply to: Erica_Edelman’s comment on: Supportive Scepticism

Hard decisions are (usually) the least important

A decision can be hard because the possible outcomes are finely balanced in expected payoff, or because you are quite lacking in knowledge about the possibly outcomes and/or their likelihood. If it’s the latter then it can be hard and matter a lot! For effective altruists there can be a bit of both. “Should I buy this pen or this other one? A better pen might help me write more effectively!” is probably the former, but “What career should I choose?” is probably the latter.

Plus, the latter kind of decision holds the promise of high value of information. If only you devote a bit more time to thinking about it or researching, you might improve your estimates a lot (or not). So that’s another incentive to worry about and delay such a decision.

Michael_PJ 24 Sep 2014 19:23 UTC
2 points
0 ∶ 0
on: Learning From Less Wrong: Special Threads, and Making This Forum More Useful
Many forums have a concept of a “stickied thread”, which is one that stays at the top of the list of topics permanently. I don’t know how good a fit that is for a forum designed like this, where there are very few visible topics, and also few subdivisions, but it does work well. It would be a good fit for things like the introductions thread.

I think LessWrong suffered from a lack of stickies—almost all the useful topics you posted saw extremely rapid declines in visibility as they dropped off the front page, which makes them effectively useless. The ones people liked—open threads, media threads, introduction threads—had to be reposted regularly for this reason. There’s no reason why a stickied introduction thread can’t last for a very long time (i.e. until it becomes unreasonably large).

Michael_PJ 2 Oct 2014 21:50 UTC
3 points
0 ∶ 0
on: Effective policy? Requiring liability insurance for dual-use research
I worry about the “that will never happen” effect. Mandating that researchers take out the insurance prevents it being dismissed on that front, but how do we make the insurance agencies take it seriously?

It seems all too plausible that the insurer will just say “this will never happen, and if it does it will be unusual enough that we can probably hold the whole thing up in court—just give them a random number”. For a big enough risk, if it happens then the insurer might expect to cease to exist in the upheaval, which also doesn’t give them much incentive to give a good estimate.

In general, I’m not sure whether insurers are quite robust enough institutions to be likely to have rational decision procedures over risks that are this big and unlikely.

Michael_PJ 2 Oct 2014 22:02 UTC
8 points
0 ∶ 0
on: Why is effective altruism new and obvious?
I think you’re quite right about this, and I would identify one of the key points of disagreement as cosmopolitanism. I think that post resonated with a lot of people precisely because it highlighted something that everyone kind of knew was an unusual feature of EA arguments, but nobody had quite put their finger on.

(And you yourself say something similar in your comment on that post.)

Cosmopolitanism is quite unconventional, but it’s also difficult to tackle head on, because it often involves a conflict between people’s stated moral views and their behaviour. Lots of people have views that, on the face of it, would make them cosmopolitan, but they rarely act in such a way. That’s partly because the implications can seem very demanding. So paying more attention to cosmopolitan arguments confronts people with a double threat—that of having an increased moral obligation, and that of being shown to be a hypocrite for not having accepted it before.

I could see this aversion just quietly diverting people away from really thinking about cosmopolitan ideas.

Michael_PJ 24 Oct 2014 5:27 UTC
3 points
0 ∶ 0
on: Should Giving What We Can change its Pledge?
There are two aspects to having more impact: giving more effectively and giving more. The GWWC pledge says something about both, but I think it is only the latter that really needs the behavioral support that you get from something like a pledge. Once people have got the idea of giving effectively, I think it is unlikely that they will stop giving effectively when they give. It’s a hard idea to unsee! But they might find it harder to keep giving as much.

So I think the most important bit is the 10%, when it comes to actually affecting people’s behavior.

That said, there is a signaling and outreach benefit to mentioning global poverty: it makes us look more credible, and a lot of people come to effective altruism via global health and poverty. I think we could have a cause-natural pledge but still make it clear that poverty is focus of the organization:

“While GWWC believes that alleviating global poverty is the best way to hero the world today, we live in hope that poverty may one day be eliminated, maybe even within our lifetimes. Hence the pledge does not oblige you to give to global poverty charities, but rather to those that you believe will best help the world. Even if you disagree with us about what the best approach is at the moment, do consider taking the pledge anyway, as we would love to have anyone who is serious about effective giving in our community.”

Michael_PJ 20 May 2015 22:28 UTC
2 points
0 ∶ 0
on: Log-normal lamentations
Thanks for writing this. I often feel quite similar—I can find being in contact with so many amazing people either inspiring or oddly demotivating!

That’s the old status-conscious monkey brain talking (everyone’s grasping vaingloriousness), and we shouldn’t feed it, but it’s good to acknowledge that it’s there from time to time.

Overall, I think the EA movement is pretty good at being positive. I’ve found such criticism as there is is usually self-criticism—if anything, I find people to be unusually generous with praise, which is lovely. I think you hit the nail on the head with your four sources of ego-damage. And yeah, I think the right thing to do is to try and not be bothered. For bonus points, remember to praise people when they do good things!

Michael_PJ 12 Jan 2016 23:59 UTC
1 point
0 ∶ 0
on: Announcing the Good Technology Project
To be clear, our current objective is to find problems that can be addressed with technology, along with at least some ideas about how to go about doing it. That might be something like “build an insurance product on top of mobile money”, or “build electricity distribution algorithms to allocate energy from renewable power sources” (disclaimer, these may not be good ideas!).

We then hope to actually be able to help get talented people actually doing some of these things!

Michael_PJ 17 Jan 2016 13:05 UTC
0 points
0 ∶ 0
in reply to: William_S’s comment on: Announcing the Good Technology Project
Working out what skills are needed for the problems is absolutely something we want to find out. I don’t know whether we can really effectively survey the pool of available talent, but we will hopefully be able to help individuals make decisions by telling them that, e.g. machine learning skills are particularly likely to be applicable to high-impact solutions

Michael_PJ 26 Jan 2016 21:22 UTC
8 points
0 ∶ 0
on: Beware surprising and suspicious convergence
This reminds me of Bryan Caplan: http://econlog.econlib.org/archives/2016/01/the_invisible_t.html

To continue the riff, we might say that the appropriate emotion for someone arguing that funding opera in fact maximises human welfare is surprise. “Now, I know you wouldn’t expect this, but amazingly it turns out that...”

Your interlocutor might believe that funding opera maximises human welfare for good reasons. But if they don’t seem to think that it’s a remarkable fact, that suggests the opposite.

Michael_PJ 27 Oct 2016 19:20 UTC
10 points
0 ∶ 0
on: Setting Community Norms and Values: A response to the InIn Open Letter
This concerns me because “EA” is such a vaguely defined group.

Here are some clearly defined groups:
- The EA FB group
- The EA forum
- Giving What We Can
All of these have a clear definition of membership and a clear purpose. I think it is entirely sensible for groups like this to have some kinds of rules, and processes for addressing and potentially ejecting people who don’t conform to those rules. Because the group has a clear membership process, I think most people will accept that being a member of the group means acceding to the rules of the group.

“EA”, on the other hand, is a post hoc label for a group of people who happened to be interested in the ideas of effective altruism. One does not “apply” to be an “EA”. Nor does can we meaningfully revoke membership except by collectively refusing to engage with someone.

I think that attempts to police the borders of a vague group like “EA” can degenerate badly.

Firstly, since anyone who is interested in effective altruism has a plausible claim to be a member of “EA” under the vague definition, there will continue to be many people using the label with no regards for any “official” definition.

Secondly (and I hope this won’t happen), such a free-floating label is very vulnerable to political (ab)use. We open ourselves up to arguments about whether or not someone is a “true” EA, or schisms between various “official” definitions. At risk of bringing up old disagreements, the arguments about vegetarian catering at last year’s EA Global were already veering in this direction.

This seems to me to have been a common fate for vague group nouns over the years, with feminism being the most obvious example. We don’t want to have wars between the second- and third-wave EAs!

My preferred solution is to avoid “EA” as a noun. Apart from the dangers I mentioned above, its origin as a label for an existing group of people gives it all sorts of connotations that are only really valid historically: rationalist thinking style, frank discussion norms, appreciation of contrarianism … not to mention being white, male, and highly educated. But practically, having such a label is just too useful.

The only other suggestion I can think of is to make a clearly defined group for which we have community norms. For lack of a better name, we could call it “CEA-style EA”. Then the CEA website could include a page that describes the core values of “CEA-style EAs” and some expectations of behaviour. At that point we again have a clearly defined group with a clear membership policy, and policing the border becomes a much easier job.

In practice, you probably wouldn’t want an explicit application process, with it rather being something that you can claim for yourself—unless the group arbiter (CEA) has actively decreed that you cannot. Indeed, even if someone has never claimed to be a “CEA-style EA”, declaring that they do not meet the standard can send a powerful signal.

EA should invest more in exploration

Michael_PJ5 Feb 2017 17:11 UTC

23 points

27 comments7 min readEA link

Michael_PJ 6 Feb 2017 23:24 UTC
0 points
0 ∶ 0
in reply to: Benjamin_Todd’s comment on: EA should invest more in exploration
I think I agree with this—it’s usually the case that one particular sub-problem in an area is particularly informative to work on.

However, I think it’s at least possible that some areas are systematically very informative to work on. For example, if the primary work is research, then you should expect to mainly be outputting information. AI research might be like this.

Michael_PJ 6 Feb 2017 23:27 UTC
1 point
0 ∶ 0
in reply to: Alex_Barry’s comment on: EA should invest more in exploration
I’m unsure where the balance should lie quantitatively. I think that $100 million would probably be too much, and $10 million is probably too low.

I agree that x-risk work doesn’t fit nicely into this: it’s not even clear whether you’d want to count the output of research as “actually” what you want or as VoI.

Michael_PJ 6 Feb 2017 23:28 UTC
1 point
0 ∶ 0
in reply to: Daniel_Eth’s comment on: EA should invest more in exploration
Exploration also carries significant risk of failure, which can be offputting. I don’t think there’s any way around that but to be somewhat tolerant of failure. But not so tolerant that people don’t try hard!

Michael_PJ 6 Feb 2017 23:35 UTC
0 points
0 ∶ 0
in reply to: Benjamin_Todd’s comment on: EA should invest more in exploration
1) I nearly added a section about whether exploration is funiding- or talent-constrained! In short, I’m not sure, and I suspect it’s different in different places. It sounds like OPP is probably talent-constrained, but other orgs may differ. In particular, if we wanted to try some of my other suggestions for improving exploration, like building institutions to start new orgs, then that’s potentially quite funding-intensive.

2) I’m not sure whether multi-armed bandits actually model our situation, since I’m not sure if you can incorporate situations where you can change the efficiencies of your actions. What does “improving exploration capacity” look like in a multi-armed bandit? There may also be complications because we don’t even know the size of the option set.

Michael_PJ 6 Feb 2017 23:37 UTC
0 points
0 ∶ 0
in reply to: tjmather’s comment on: EA should invest more in exploration
I think this is a case where we’re unlikely to be able to offer anything beyond what the academic community is going to do. I think the best way to improve exploration around schistosomiasis prevention would probably be to just fund some more PhD students!

Michael_PJ 7 Feb 2017 23:58 UTC
0 points
0 ∶ 0
in reply to: RobBensinger’s comment on: Anonymous EA comments
I agree that we’re in danger of having picked all the low-hanging fruit. But I think there’s room to fix this.

Michael_PJ 8 Feb 2017 0:06 UTC
3 points
0 ∶ 0
in reply to: RobBensinger’s comment on: Anonymous EA comments
There’s a lot of EA outside the Bay! The Oxford/London cluster in particular is quite nice (although I live there, so I’m biased).

Michael_PJ 9 Feb 2017 12:52 UTC
2 points
0 ∶ 0
on: Use “care” with care.
Thanks for this—I’m pretty sure I’m guilty of doing this carelessly, and I agree that it’s actually not great.

Michael_PJ 8 Mar 2017 16:19 UTC
5 points
0 ∶ 0
on: Introducing CEA’s Guiding Principles
This is pretty much exactly what I was hoping for! Thank you!

Michael_PJ

EA should in­vest more in exploration

EA should invest more in exploration