All of Ben_West's Comments + Replies

Thanks! It's a new feature we are piloting – authors can turn it on in their settings:

We will have an announcement post about it coming out soon

Thanks! Lesswrong is currently experimenting with multidimensional voting; if you haven't already, I would suggest trying that out and giving them feedback.

Thanks!

I believe this is just the confusing way that Google handles anonymous forms. It states the account you are currently using, but then has a parenthetical indicating that the information won't be shared.

Thanks! All of our metrics are pretty well correlated with each other; you can see more information here.

Our primary metric is hours of engagement, which I didn't use for this post because the data doesn't stretch back as far. But the growth rate there is:

• 2020 (estimated): 90%
• 2021: 100.2%
• 2022 (so far): 140%
• Implied 3-year growth: 912%

More about how this is calculated and our historical data can be found here.

Thanks for sharing this! I hadn't really conceptualized the questions you asked as being "EA" – I had assumed that any sort of international development would need to benchmark on facts like the effectiveness of bed nets. Or at least international development professors would do better than the general public on this kind of thing; "international development is harder than you might think, even if people are pretty poor" seems like a very cold take in IDEV circles?

I guess my assumptions were wrong!

Thanks for the push back! I agree that 80k cares more about the use of their listener's time than most podcasters, although this is a low bar.

80k is operating under a lot of constraints, and I'm honestly not sure if they are actually doing anything incorrectly here. Notably, the fancy people who they get on the podcast probably aren't willing to devote many hours to rephrasing things in the most concise way possible, which really constrains their options.

I do still feel like there is a missing mood though.

Longform's missing mood

If your content is viewed by 100,000 people, making it more concise by one second saves an aggregate of one day across your audience. Respecting your audience means working hard to make your content shorter.

When the 80k podcast describes itself as "unusually in depth," I feel like there's a missing mood: maybe there's no way to communicate the ideas more concisely, but this is something we should be sad about, not a point of pride.^[1]

Thanks for the thoughtful response!

My anecdotal experience with GMA tests is that hiring processes already use proxies for GMA (education, standardized test scores, work experience, etc.) so the marginal benefit of doing a bona fide GMA test is relatively low.

It would be cool to have a better sense of when these tests are useful though, and an easy way to implement them in those circumstances.

Just wanted to nudge that I would find this write up very valuable. Even if the ranges are very wide, I often want to reference some sort of monetary estimate of the value of labor, and having a post like this to reference would be quite useful.

Thanks for doing this Sinclair! 

Thanks for writing this! I've personally struggled to apply academic research to my hiring, and now roughly find myself in the position of who "Stubborn Reliance" is criticizing, i.e. I am aware of academic research but believe it doesn't apply to me (at least not in any useful way). I would be interested to hear more motivation/explanation about why hiring managers should take these results seriously. 

Two small examples of why I think the literature is hard to apply:

GMA Tests
If you take hiring literature seriously, my impression is that the thing you... (read more)

On the sanity check: Reddit makes about four cents in revenue per user per month. It doesn't seem crazy to me that the average user gets two dollars of value per month, but a lot of this would depend on things like how many of their users are diehard versus casual users.

We are looking to hire, thanks! I put a link to our open positions at the bottom of the post.

My model was:

• It's kind of unclear what a marginal unit is for a software company like Reddit, but let's just say it's one user of their software
• Profit maximizing firms produce until marginal cost = marginal benefit
• Marginal benefit is 50 times higher
• Therefore firms will accept 50 times higher marginal cost
• Let's suppose that the percentage of this additional marginal cost which goes to labor remains unchanged, resulting in 50 times more employees per user

30 seconds of thought can identify a bunch of problems with this model, but I think the underlying insigh... (read more)

Furthermore, I'm not sure the information value alone was worth the millions spent on this campaign by the EA community. The 'lessons learned' listed in this forum post seem obvious.

The post author doesn't say anything about having a special connection to the campaign. I assume the "value of information" argument is that campaign staff/insiders gained knowledge they couldn't have gotten otherwise, and I'm not sure this post would shed much light on that argument either way. 

As a relatively trivial example of learning not available from a Google search: the campaign presumably learned things like how many people would show up to make calls, how much money they could raise, etc.

Can you clarify why you think it's "borderline illegal"? I assume you are referring to GDPR, but I'm not aware of any reason why the normal "legitimate interest" legal basis wouldn't apply to group organizers.

Hmm,  I think the question of whether employees should have objective requirements is somewhat orthogonal from the question of whether they should have a tour of service. For example, many salespeople have sales quotas, despite not being on a tour of service.

That being said: the alternative to having objective requirements is something like "you must fulfill the whims of your manager" and it's not obvious to me that this is actually better for job security.

Thanks for the question! The differences in my mind are:
 

1. A Schelling point for when the employment relationship might end
2. Clear discussion of the benefits the employee will receive even after leaving the employer

E.g. I've never had an employer pitch me something like "work for us for two years, after which you will be much more hirable by our competitor because of the portfolio you developed here." (Even though this is the strategy many employees have in practice.)

Thanks for doing this! The interviews are really interesting to read; the CEO example seems like something which perhaps should gain more prominence as a way to introduce AI-knowledgeable audiences to risks about alignment.

Thanks Charlie! If I understand your concern correctly, this is a misunderstanding of the approach. To quote the post

Note that the Tour of Service (both in the original version and CEA’s) is an informal and non-legally binding agreement. The legal structure of employment is unchanged... Like any other unusual hiring practice, people sometimes get confused. A decent fraction of our candidates think that a tour of service means that we are only hiring them for a limited-term engagement, and they worry about their job security. I’ve iterated on various ways

Thanks for sharing this! A couple quick thoughts:
 

1. One thing which I liked about Singer's original "famine, affluence, and morality" is that it included direct appeals for people to help. You start your first video off by saying that you are trying to get people to care about the developing world, which is great, but I think it would be cool if you did even more to signal "this is not an edgelord thought experiment but something which should actually change your behavior." One concrete thing is to include a donation button in the video (I think you hav

Several of the videos are tagged #effectivealtruism and the first video is currently the second highest video on the tag.

Thanks for the suggestion! I've added this to our backlog.

Thanks for the suggestion! I've added this to our backlog.

Thanks for the suggestion! I've added creating a feature like this to our backlog.

Thanks for the suggestion! I've added this to our backlog.

Thanks for the suggestion! I've added this to our backlog.

Thanks for the suggestion! This is on our roadmap.

Thanks for the suggestion! This is on our roadmap.

Thanks for the suggestion! I've added this to our backlog.

Thanks for the suggestion! Could you post it here?

Nope, it was written by me. I tried to explain that bit with footnote 2 – let me know if you have suggested wording changes to make it more clear!

This is a great counterexample/clarification, thanks!

I might do a future post about how important it is to hire value aligned people; I agree that this is a slightly different question.

I think you are saying something like: "outsourcing is a managerial task, therefore bottlenecks on outsourcing are by definition bottlenecked on management."

I think this is true, but I don't think it's the most helpful way of phrasing it. E.g. many biology labs can't outsource their research (or even have it be replicated by labs which are almost identical) because their work relies on a bunch of tiny things like "you should incubate the cells at 30°C except if you notice some of them starting to turn a little yellowish increase the heat to 32°C but then a... (read more)

Yeah, you can definitely get into something like the old joke about two economists who find a $20 bill on the ground: your activities must be core competencies, because if they weren't you would already have outsourced them.

FWIW I didn't interpret it is hostile, though I did change the title to make it more clear that I'm not suggesting CEA change

Hmmm, no, I think the ability to outsource well is not itself easily outsourceable. E.g. if you have some method of identifying whether an outsourced factory will produce high-quality products, I guess you could train an outsourced team to do that identification, but that doesn't seem remarkably easier than hiring staff and training them on your identification methods.

In general I do have a preference for hiring people involved in EA. Among other things, these people are easier to vet (as you remark).

When I hire contractors who are not EA's (including those in developing countries), it's usually because we need a specific skill set and can't easily find EA's with that skill set.

The breakdown you list at the bottom of your comment seems approximately similar to how I think about how easy work is to outsource.

Good question; it seems pretty similar to feature development.

• Can be outsourced: "make this button hidden on mobile."
• Hard to outsource: "investigate reports of sporadic data loss."

I think the reason the second one is hard to outsource is general problem.

(E.g. how you would verify that the outsourced firm investigated it correctly? Verifying that they actually fixed this hard to debug issue is often almost as complicated as just fixing it yourself. So you basically have to trust that the firm did it correctly, and getting an engineer who you trust to fix these problems is basically what it means to hire and onboard an engineer.)

Yeah I think that's right. "Having drivers in every major city in the world" is a core competency.

I would have said "network effect" (though I don't know much about either of them). This seems analogous to the claim that one of Walmart's core competencies is purchasing power.

An example is something like "having a CEO who is considered a prestigious thought leader in their field." The day-to-day operations of the business aren't really impacted by this, but it's also not something you can really outsource.

(That being said, maybe it would have been a lot simpler and almost as correct to just leave this axis off, like you suggest.)

This is a really interesting post, and I had not seen it before. Thanks for sharing!

Thanks! I'm glad to hear they are helpful.

Perhaps a dumb question, but do companies tend to have an accurate picture of what their core competencies are?

This is an interesting question! The world is certainly full of stories of outsourcing gone awry, which is some evidence that companies don't always have an accurate picture. But I don't know rigorous statistics on this.

Just following the example from the post: did CEA ever try to outsource some product-related tasks? Are there any circumstances in which it would it make sense for you to do so even if it's a core competency?

Yep, for example we hav... (read more)

Yeah, this is what footnote 2 was trying to explain.

I was considering marking that quadrant some third color to indicate that you should just not do those things, but thought that would overcomplicate the message of "only outsource the bottom right quadrant". Peter's comment that this is confusing is helpful feedback that maybe that was the wrong call – thanks! 

Is the first number a typo? Shouldn't it be ~54

Fixed, thanks!

I think the summary of this post should be updated to say something like "CEA is more competitive but in the same ballpark as industry"

I agree we hire a smaller percent of total applicants, but we hire a substantially greater percent of applicants who get to the people ops interview stage.

I think the latter number is probably the more interesting one because the former is affected a bunch by e.g. if your job posting gets put onto a random job board which gives you a ton of low-quality applicants.... (read more)