Welcome to The Nonlinear Library, where we use Text-to-Speech software to convert the best writing from the Rationalist and EA communities into audio. This is: TikTok EA Book Giveaway Intermediate Results, published by Ben West on November 11, 2021 on The Effective Altruism Forum.

Myself and Lacey (@benthamite) and Emma Williamson (@dailydosenews) piloted giving away copies of Doing Good Better on TikTok. This is a short write up of intermediate results.

Executive summary

We gave away 359 books across 5 videos

320 of those came from the most successful video

Each video took a couple of hours of work

I want to do a follow-up survey to see how many of these people actually read the book and got involved in EA. My guess is that a few of them did, which would make this a fairly cost-effective outreach strategy. I plan to publish a longer write up once those survey results are in.

Process

We created videos advertising various EA things, with a call to action for them to get a free book by accessing a link in our bio

People filled out a form with their contact information to receive the book. We did not require a credit card etc.

The form let them opt out of receiving a follow-up survey

Advertising

I tried purchasing promotion for one video. In TikTok, this means that the video was shown to more people than usual with small text indicating that it's being shown to the user via paid promotion.

As expected, this dramatically increased the number of people viewing the video, but very few of them signed up to get a book.

Probably someone should try this with a larger sample size though, and also with video which is inherently more compelling.

(I am unable to promote the most popular video as the sound is copyrighted.)

Thoughts on the most popular video

I filmed Sam doing a dance (as opposed to e.g. talking into the camera) because this let me decide later what text to put on top of it instead of having to decide that before filming.

I think this was the right call – my first attempt was not very successful, but the second one was.

Sam obviously has starpower but the fact that the first video was unsuccessful indicates that people aren't just watching because of his celebrity. This makes me hopeful that at least some of the people who got books did so because they were intrigued by the message.

The success of the video seems to have mostly been driven by having a higher watch percentage (as opposed to engagement through likes/comments etc.)

Thanks to the EABD team and the Open Philanthropy Project for making this project possible

16

1,543

136

5

5

21.0%

3-10

1,090

69

9

1

17.0%

4-7

14,200

247

2

0

23

2,580

84

18

7

31.7%

9

125,800

1,652

182

390

43.9%

320

Thanks for listening. To help us out with The Nonlinear Library or to learn more, please visit nonlinear.org.

Welcome to The Nonlinear Library, where we use Text-to-Speech software to convert the best writing from the Rationalist and EA communities into audio. This is: CGP Grey: The Fable of the Dragon-Tyrant, published by SamClarke on November 11, 2021 on The Effective Altruism Forum.

CGP Grey's animation of The Fable of the Dragon-Tyrant is probably the most moving piece of EA content I know of, by some margin.[1]

I'm linkposting it because (1) I've been surprised by the number of people who haven't seen it, and (2) I think it could be really cool if we had more content like this.

I'd also appreciate any comments linking to moving EA content I might have missed!

Thanks for listening. To help us out with The Nonlinear Library or to learn more, please visit nonlinear.org.

Welcome to The Nonlinear Library, where we use Text-to-Speech software to convert the best writing from the Rationalist and EA communities into audio. This is: CGP Grey: The Fable of the Dragon-Tyrant, published by SamClarke on November 11, 2021 on The Effective Altruism Forum.

CGP Grey's animation of The Fable of the Dragon-Tyrant is probably the most moving piece of EA content I know of, by some margin.[1]

I'm linkposting it because (1) I've been surprised by the number of people who haven't seen it, and (2) I think it could be really cool if we had more content like this.

I'd also appreciate any comments linking to moving EA content I might have missed!

Thanks for listening. To help us out with The Nonlinear Library or to learn more, please visit nonlinear.org.

Welcome to The Nonlinear Library, where we use Text-to-Speech software to convert the best writing from the Rationalist and EA communities into audio. This is: A Brief Introduction to Container Logistics, published by Vitor on November 11, 2021 on LessWrong.

Container logistics is an interesting but complicated topic, with a lot of implicit knowledge kept by industry insiders. In this post, I'll give a brief overview, based on my experiences having worked in the industry for three different shipping companies (all located in Chile) over a period of five years. Hopefully, this will allow people to get more accurate models of the heavy port congestion in Long Beach that has been theorized about extensively here on LW. I can't comment on circumstances specific to that port or the US more generally; but there's a set of underlying intuitions that should give people a solid, universal foundation to think about the issue. Based on those intuitions, I think people are way too optimistic about simple and easy solutions that claim to make major progress on the problem overnight.

Lifecycle of a Container Shipment

I'll start with the basics. Let's say you want to send some cargo from port A to B. Thanks to the invention of standardized shipping containers, this is a relatively easy process, vastly easier than 100 years ago. All the equipment used along the trip (ships, cranes, truck trailers, etc) are specialized for the task of moving containers. This reduces costs dramatically, and makes your cargo arrive faster and safer.

The main trunk lines of international transport are between Asia and the US west coast, and between Asia and Europe (through the Suez canal). There's a significant amount of traffic across the Atlantic too. All of this is done with massive container ships carrying 8'000 or more TEUs (twenty-foot equivalent units. 20' container = 1 TEU, 40' container = 2 TEUs). Then there are many, many feeder services with much smaller vessels, going along the coasts of every continent, delivering cargo to/from smaller ports. I was in charge of container logistics for one such feeder line along the South American West Coast, visiting Colombia, Ecuador, Peru and Chile.

From a user perspective the process of shipping from A to B looks like this: you make a booking with a shipping company. This booking allows you to pick up a standardized container at a depot, which is usually near A, but might be hundreds of kilometers inland as well. You fill the container with your goods, and apply a tamper proof seal. You then arrange to get that container to port A, which will give you a bill of lading (a physical or virtual document similar to a cheque for the goods). The container is then loaded onto a ship, and sometime later that ship arrives at port B, where it can be claimed by someone who holds the bill of lading. The container may or may not be on its original ship: it may have been a straight shot, or it may have been transferred at multiple ports between A and B, e.g., from a small feeder route to a larger trunk route.

After the recipient unloads the container at port B, it returns it to a depot. This depot is either directly owned by the shipping line, or it may be a separate company that offers their services to multiple shipping lines. A depot will typically contain thousands of containers. Here, containers are inspected, and minor damages like tears, dents and oil spills can be cleaned up between uses. Shipping lines will typically insist on a return to a depot even if the recipient immediately wants to ship something back out, for liability reasons.

These depots are concentrated near the port of course, but there are also many depots far away from any port, near important cities and industrial centers. In that sense, a port can have a very large area of influence, and the container fleet that a shipping company keeps will also be spread over this entire area.

Issues of Trust and Cooperation

The whole shipping process involves doze...

Welcome to The Nonlinear Library, where we use Text-to-Speech software to convert the best writing from the Rationalist and EA communities into audio. This is: A Brief Introduction to Container Logistics, published by Vitor on November 11, 2021 on LessWrong.

Container logistics is an interesting but complicated topic, with a lot of implicit knowledge kept by industry insiders. In this post, I'll give a brief overview, based on my experiences having worked in the industry for three different shipping companies (all located in Chile) over a period of five years. Hopefully, this will allow people to get more accurate models of the heavy port congestion in Long Beach that has been theorized about extensively here on LW. I can't comment on circumstances specific to that port or the US more generally; but there's a set of underlying intuitions that should give people a solid, universal foundation to think about the issue. Based on those intuitions, I think people are way too optimistic about simple and easy solutions that claim to make major progress on the problem overnight.

Lifecycle of a Container Shipment

I'll start with the basics. Let's say you want to send some cargo from port A to B. Thanks to the invention of standardized shipping containers, this is a relatively easy process, vastly easier than 100 years ago. All the equipment used along the trip (ships, cranes, truck trailers, etc) are specialized for the task of moving containers. This reduces costs dramatically, and makes your cargo arrive faster and safer.

The main trunk lines of international transport are between Asia and the US west coast, and between Asia and Europe (through the Suez canal). There's a significant amount of traffic across the Atlantic too. All of this is done with massive container ships carrying 8'000 or more TEUs (twenty-foot equivalent units. 20' container = 1 TEU, 40' container = 2 TEUs). Then there are many, many feeder services with much smaller vessels, going along the coasts of every continent, delivering cargo to/from smaller ports. I was in charge of container logistics for one such feeder line along the South American West Coast, visiting Colombia, Ecuador, Peru and Chile.

From a user perspective the process of shipping from A to B looks like this: you make a booking with a shipping company. This booking allows you to pick up a standardized container at a depot, which is usually near A, but might be hundreds of kilometers inland as well. You fill the container with your goods, and apply a tamper proof seal. You then arrange to get that container to port A, which will give you a bill of lading (a physical or virtual document similar to a cheque for the goods). The container is then loaded onto a ship, and sometime later that ship arrives at port B, where it can be claimed by someone who holds the bill of lading. The container may or may not be on its original ship: it may have been a straight shot, or it may have been transferred at multiple ports between A and B, e.g., from a small feeder route to a larger trunk route.

After the recipient unloads the container at port B, it returns it to a depot. This depot is either directly owned by the shipping line, or it may be a separate company that offers their services to multiple shipping lines. A depot will typically contain thousands of containers. Here, containers are inspected, and minor damages like tears, dents and oil spills can be cleaned up between uses. Shipping lines will typically insist on a return to a depot even if the recipient immediately wants to ship something back out, for liability reasons.

These depots are concentrated near the port of course, but there are also many depots far away from any port, near important cities and industrial centers. In that sense, a port can have a very large area of influence, and the container fleet that a shipping company keeps will also be spread over this entire area.

Issues of Trust and Cooperation

The whole shipping process involves doze...

Welcome to The Nonlinear Library, where we use Text-to-Speech software to convert the best writing from the Rationalist and EA communities into audio. This is: Discussion with Eliezer Yudkowsky on AGI interventions, published by RobBensinger on November 11, 2021 on The Effective Altruism Forum.

The following is a partially redacted and lightly edited transcript of a chat conversation about AGI between Eliezer Yudkowsky and a set of invitees in early September 2021. By default, all other participants are anonymized as "Anonymous".

I think this Nate Soares quote (excerpted from Nate's response to a report by Joe Carlsmith) is a useful context-setting preface regarding timelines, which weren't discussed as much in the chat transcript:

[...] My odds [of AGI by the year 2070] are around 85%[...]

I can list a handful of things that drive my probability of AGI-in-the-next-49-years above 80%:

1. 50 years ago was 1970. The gap between AI systems then and AI systems now seems pretty plausibly greater than the remaining gap, even before accounting the recent dramatic increase in the rate of progress, and potential future increases in rate-of-progress as it starts to feel within-grasp.

2. I observe that, 15 years ago, everyone was saying AGI is far off because of what it couldn't do -- basic image recognition, go, starcraft, winograd schemas, programmer assistance. But basically all that has fallen. The gap between us and AGI is made mostly of intangibles. (Computer Programming That Is Actually Good? Theorem proving? Sure, but on my model, "good" versions of those are a hair's breadth away from full AGI already. And the fact that I need to clarify that "bad" versions don't count, speaks to my point that the only barriers people can name right now are intangibles.) That's a very uncomfortable place to be!

3. When I look at the history of invention, and the various anecdotes about the Wright brothers and Enrico Fermi, I get an impression that, when a technology is pretty close, the world looks a lot like how our world looks.

Of course, the trick is that when a technology is a little far, the world might also look pretty similar!

Though when a technology is very far, the world does look different -- it looks like experts pointing to specific technical hurdles. We exited that regime a few years ago.

4. Summarizing the above two points, I suspect that I'm in more-or-less the "penultimate epistemic state" on AGI timelines: I don't know of a project that seems like they're right on the brink; that would put me in the "final epistemic state" of thinking AGI is imminent. But I'm in the second-to-last epistemic state, where I wouldn't feel all that shocked to learn that some group has reached the brink. Maybe I won't get that call for 10 years! Or 20! But it could also be 2, and I wouldn't get to be indignant with reality. I wouldn't get to say "but all the following things should have happened first, before I made that observation". I have made those observations.

5. It seems to me that the Cotra-style compute-based model provides pretty conservative estimates. For one thing, I don't expect to need human-level compute to get human-level intelligence, and for another I think there's a decent chance that insight and innovation have a big role to play, especially on 50 year timescales.

6. There has been a lot of AI progress recently. When I tried to adjust my beliefs so that I was positively surprised by AI progress just about as often as I was negatively surprised by AI progress, I ended up expecting a bunch of rapid progress. [...]

Further preface by Eliezer:

In some sections here, I sound gloomy about the probability that coordination between AGI groups succeeds in saving the world. Andrew Critch reminds me to point out that gloominess like this can be a self-fulfilling prophecy - if people think successful coordination is impossible, they won’t try to coordinate. I therefore remark in retrospective advance that it seems to me like at least som...

Welcome to The Nonlinear Library, where we use Text-to-Speech software to convert the best writing from the Rationalist and EA communities into audio. This is: Discussion with Eliezer Yudkowsky on AGI interventions, published by RobBensinger on November 11, 2021 on The Effective Altruism Forum.

The following is a partially redacted and lightly edited transcript of a chat conversation about AGI between Eliezer Yudkowsky and a set of invitees in early September 2021. By default, all other participants are anonymized as "Anonymous".

I think this Nate Soares quote (excerpted from Nate's response to a report by Joe Carlsmith) is a useful context-setting preface regarding timelines, which weren't discussed as much in the chat transcript:

[...] My odds [of AGI by the year 2070] are around 85%[...]

I can list a handful of things that drive my probability of AGI-in-the-next-49-years above 80%:

1. 50 years ago was 1970. The gap between AI systems then and AI systems now seems pretty plausibly greater than the remaining gap, even before accounting the recent dramatic increase in the rate of progress, and potential future increases in rate-of-progress as it starts to feel within-grasp.

2. I observe that, 15 years ago, everyone was saying AGI is far off because of what it couldn't do -- basic image recognition, go, starcraft, winograd schemas, programmer assistance. But basically all that has fallen. The gap between us and AGI is made mostly of intangibles. (Computer Programming That Is Actually Good? Theorem proving? Sure, but on my model, "good" versions of those are a hair's breadth away from full AGI already. And the fact that I need to clarify that "bad" versions don't count, speaks to my point that the only barriers people can name right now are intangibles.) That's a very uncomfortable place to be!

3. When I look at the history of invention, and the various anecdotes about the Wright brothers and Enrico Fermi, I get an impression that, when a technology is pretty close, the world looks a lot like how our world looks.

Of course, the trick is that when a technology is a little far, the world might also look pretty similar!

Though when a technology is very far, the world does look different -- it looks like experts pointing to specific technical hurdles. We exited that regime a few years ago.

4. Summarizing the above two points, I suspect that I'm in more-or-less the "penultimate epistemic state" on AGI timelines: I don't know of a project that seems like they're right on the brink; that would put me in the "final epistemic state" of thinking AGI is imminent. But I'm in the second-to-last epistemic state, where I wouldn't feel all that shocked to learn that some group has reached the brink. Maybe I won't get that call for 10 years! Or 20! But it could also be 2, and I wouldn't get to be indignant with reality. I wouldn't get to say "but all the following things should have happened first, before I made that observation". I have made those observations.

5. It seems to me that the Cotra-style compute-based model provides pretty conservative estimates. For one thing, I don't expect to need human-level compute to get human-level intelligence, and for another I think there's a decent chance that insight and innovation have a big role to play, especially on 50 year timescales.

6. There has been a lot of AI progress recently. When I tried to adjust my beliefs so that I was positively surprised by AI progress just about as often as I was negatively surprised by AI progress, I ended up expecting a bunch of rapid progress. [...]

Further preface by Eliezer:

In some sections here, I sound gloomy about the probability that coordination between AGI groups succeeds in saving the world. Andrew Critch reminds me to point out that gloominess like this can be a self-fulfilling prophecy - if people think successful coordination is impossible, they won’t try to coordinate. I therefore remark in retrospective advance that it seems to me like at least som...

Welcome to The Nonlinear Library, where we use Text-to-Speech software to convert the best writing from the Rationalist and EA communities into audio. This is: Business Coaching/Mentoring For EA Organisations, published by TheBusinessGrowthAgency on November 9, 2021 on The Effective Altruism Forum.

TL;DR: if you're a CEO or senior decision maker at an EA org I'd like to invite you to apply for our pro bono business coaching service. If you know somebody who could help connect EA orgs and business coaches/mentors please point me in their direction.

I'm the COO (Chief Operating Officer) at a business coaching practice in London. Over the past year I've been running a few EA side projects to try and use some of the influence I have at work to direct money and resources towards EA causes.

As a company we now donate a proportion of our profits to GiveWell charities and I've given presentations and worked individually with our CEO clients in London introducing them to basic EA concepts and supporting them in setting up giving pledges.

The next thing I want to try is providing pro bono coaching/mentoring to leaders at EA organisations.

Some reasons why we might be able to help...

We coach clients on productivity, goal setting and accountability but also practical business skills like management, operations, recruitment, culture, negotiation, marketing, ... etc. which I think could be relevant to many EA orgs as they grow

Each of our coaches have a decade plus of experience either in senior management positions at large organisations or founding and growing successful companies themselves

All our coaches have gone through rigorous coaching certification as well as coaching, between them, thousands of CEOs over the past decade

As a coaching practice with a team of 8 people we have a back office team and various different programmes that we've developed over the years. As well as 1-to-1 coaching sessions we do things like weekly group accountability coaching, workshops on specific business topics, coaching for emerging leaders in organisations and an entrepreneurs' book club. I'd be open to offering places at any of these free of charge to the right people.

If this sounds interesting to you and you would like to talk to one of our business coaches please fill out an application form at the bottom of this page.

If you know somebody who might benefit please forward them a link to this post or the application/explainer page on my website.

If you can put me in touch with someone who has a clever way of connecting coaches/mentors and EA orgs that I haven't thought of please let me know. (I've reached out to a few meta-EA organisations by email this morning but I'm sure there's others I'm not aware of or situations where I won't get anywhere without someone first making an initial introduction.)

If you have any feedback or advice on anything I've written above please let me know in the comments or via private message.

Thanks so much.

Thanks for listening. To help us out with The Nonlinear Library or to learn more, please visit nonlinear.org.

Welcome to The Nonlinear Library, where we use Text-to-Speech software to convert the best writing from the Rationalist and EA communities into audio. This is: Business Coaching/Mentoring For EA Organisations, published by TheBusinessGrowthAgency on November 9, 2021 on The Effective Altruism Forum.

TL;DR: if you're a CEO or senior decision maker at an EA org I'd like to invite you to apply for our pro bono business coaching service. If you know somebody who could help connect EA orgs and business coaches/mentors please point me in their direction.

I'm the COO (Chief Operating Officer) at a business coaching practice in London. Over the past year I've been running a few EA side projects to try and use some of the influence I have at work to direct money and resources towards EA causes.

As a company we now donate a proportion of our profits to GiveWell charities and I've given presentations and worked individually with our CEO clients in London introducing them to basic EA concepts and supporting them in setting up giving pledges.

The next thing I want to try is providing pro bono coaching/mentoring to leaders at EA organisations.

Some reasons why we might be able to help...

We coach clients on productivity, goal setting and accountability but also practical business skills like management, operations, recruitment, culture, negotiation, marketing, ... etc. which I think could be relevant to many EA orgs as they grow

Each of our coaches have a decade plus of experience either in senior management positions at large organisations or founding and growing successful companies themselves

All our coaches have gone through rigorous coaching certification as well as coaching, between them, thousands of CEOs over the past decade

As a coaching practice with a team of 8 people we have a back office team and various different programmes that we've developed over the years. As well as 1-to-1 coaching sessions we do things like weekly group accountability coaching, workshops on specific business topics, coaching for emerging leaders in organisations and an entrepreneurs' book club. I'd be open to offering places at any of these free of charge to the right people.

If this sounds interesting to you and you would like to talk to one of our business coaches please fill out an application form at the bottom of this page.

If you know somebody who might benefit please forward them a link to this post or the application/explainer page on my website.

If you can put me in touch with someone who has a clever way of connecting coaches/mentors and EA orgs that I haven't thought of please let me know. (I've reached out to a few meta-EA organisations by email this morning but I'm sure there's others I'm not aware of or situations where I won't get anywhere without someone first making an initial introduction.)

If you have any feedback or advice on anything I've written above please let me know in the comments or via private message.

Thanks so much.

Thanks for listening. To help us out with The Nonlinear Library or to learn more, please visit nonlinear.org.

Welcome to The Nonlinear Library, where we use Text-to-Speech software to convert the best writing from the Rationalist and EA communities into audio. This is: Relaxation-Based Search, From Everyday Life To Unfamiliar Territory, published by johnswentworth on November 10, 2021 on LessWrong.

Suppose I’m making plans to visit my parents for the holidays. I go online, and I book plane tickets. I do not worry about how I’m going to get to the airport - indeed, that question probably wouldn’t even enter my mind until a few days before the flight. Why not? Well, without thinking through any particular plan, I expect getting to the airport to be easy. There are many ways to get there (train, Uber/Lyft, drive & park) all of which I’ve used before and any of which would be fine.

This is relaxation-based search. I have a planning problem: visit my parents for the holidays. I’m searching for a plan which satisfies all my constraints: arrive at my parents’ house ~2-3 days before Christmas, arrive at the airport an hour before the plane takes off, don’t spend lots of time en route, don’t spend too much money, and many many other constraints (most of which are implicit in my head). In order to search efficiently, I relax the problem: I ignore many constraints which I expect will be easy to satisfy, like arriving at the airport on time or packing my bag. I throw out all but the hardest-to-satisfy constraints, which in this case mostly involve the plane tickets. I solve the “relaxed” search problem, with just the hardest constraints, then go back and tweak the plan to handle other constraints later.

Why This Matters: High-Dimensional Worlds

Our world is high-dimensional. Even a trip to visit my parents for the holidays involves hundreds of choices: dates, modes of transport, packing list, how early to arrive at the airport, etc. Multiply them all out, and we have an exponentially vast space of possibilities.

And that’s just a trip to visit my parents. Imagine the possibilities involved in designing a website or a scientific experiment!

When searching in spaces this large, we cannot afford to test every possibility. We are forced to rule out vast swaths of the search space without even looking at them. Relaxation-based search is one of the most general algorithms we know of which can do this.

The key is that we can usually express the problem-space using constraints which each depend on only a few dimensions. For instance, if I want to arrive at the airport an hour before the plane takes off, that constraint only involves two dimensions: my arrival time at the airport, and the takeoff time of the flight. It does not directly depend on what time I wake up, whether I pack a toothbrush, my parents’ plans, cost of the plane tickets, etc, etc.

So, if I can relax the problem to just a few of the most-difficult constraints, hopefully those constraints will involve a relatively small number of variables, and I can search through that smaller space more efficiently. When booking flights, for instance, I mostly just think about flight date, flight time, and cost (and maybe airline, sometimes, as a tiebreaker). I don’t worry about all the other dimensions, like packing my bag, getting to the airport, eating, buying presents for my parents, etc, etc.

Then, holding my solution for the most-taut constraints constant, I can go handle the other constraints. I plan around my flights to decide when to set my alarm, what/when to eat, and how to get to the airport. Each of those, in turn, only involves a few variables at a time, while holding the rest of the solution constant.

Tautness And Slackness

Imagine that I mistakenly think that the plane flight is an easy part of my holiday plans. I show up at the airport on December 22nd, find a service desk, and ask what flights are available.

This is likely to go poorly.

One crucial element to make relaxation-based search work well is to have an accurate idea of what’s hard, and what’s easy. Which constraints are “s...