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: Apply to Negotiating for Good [Feb 26 - Mar 12], published by Cillian Crosson on January 20, 2022 on The Effective Altruism Forum.

Description

Training for Good is offering a free training programme in salary negotiation, to help you increase the amount you donate to effective charities.

This programme will consist of two workshops: a 3 hour salary negotiation workshop on Feb 26 2022 and a 1.5 hour practice session on Mar 12 2022. The application deadline is January 30th.

All training is free and will be run online. We therefore ask that all attendees pledge to donate at least 50% of their salary increase to effective charities. (Although we understand that people have very different circumstances and are happy to discuss any and all cases).

Apply here.

Overview

This training programme will consist of:

Salary Negotiation Workshop (part one). A 3 hour interactive workshop in salary negotiation covering the following topics:

Prepare your case: Using the pre-session worksheets, we’ll build a compelling case for your raise. You’ll practice presenting this case & receive feedback from other trainees about the strongest elements and how best to improve it.

Develop a plan: You’ll develop a step-by-step approach for seeking a salary increase, from the initial message you’ll send to the exact counter-offers that you expect to make at each stage of the negotiation process.

Build the skill: We’ll discuss the shape that we want the conversation with your manager to take and provide an overview of the specific tactics that you can use to make this happen. In small groups, you’ll do a short 5-minute role play, practicing the conversation with your manager.

Practice Session (part two). A 1.5 hour practice session where you'll practice your negotiation skills with other trainees, receiving feedback and refining your approach.

Pre-session worksheets. These are to be completed and returned ahead of both sessions and are expected to take 1-2 hours per session.

Outcomes

By the end of the workshop, you will have

Developed a clear plan for navigating the negotiation process from start to finish.

Determined a precise salary goal and a compelling rationale for why you deserve it.

Improved your negotiation skills to drive up your earning (and giving) potential.

Increased confidence to ask for and successfully negotiate a significant salary increase.

When

We’ll run two separate programmes to accommodate different time zones:

Europe/Asia

Salary Negotiation Workshop (part one): 26th February, 10am - 1pm (UTC)

Practice Session (part two): 12th March, 10am - 11.30am (UTC)

Americas

Salary Negotiation Workshop (part one): 26th February, 3pm - 6pm (UTC)

Practice Session (part two): 12th March, 4.30pm - 6pm (UTC)

Testimonials

~25 people attended the first iteration of this workshop at the end of 2021 (quite a few of whom have already negotiated sizable salary increases).

From 15 respondents, our workshop received a net promoter score of 4.8 / 5. Below, I've copied the testimonials we received from our final session to give a flavour of the sessions:

It gives you all the tools you need, and the confidence to use them.

Negotiating a higher salary is one of the best ways to advance your career, be more valued, and do more good in the world and I'm so glad I took advantage of Training for Good's free salary negotiation program. The session was well-organized, engaging, and provided numerous key insights about the salary negotiation process. Highly recommend.

Steve & Cillian ran a 3 hour 'earning to give' negotiation workshop which was very helpful in convincing that you can & should negotiate! They are friendly & experienced with a lot of interesting & useful resources. Thanks for having me, and excited to see how the workshops & Training for Good evolve!

I can probably write that better after some time ...

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: 80,000 Hours is hiring!, published by Michelle Hutchinson on January 20, 2022 on The Effective Altruism Forum.

My team at 80,000 Hours is hiring for advisors. In the past I've found this a difficult role to hire for, because it isn't at traditional path for people to take. I love the role though: It's just lovely to spend every day talking to smart, like minded people and helping them problem solve about how to really do the most good with their career. I enjoy the breadth of the problems I get to think about, and that it keeps me constantly wanting to learn more about all the possible ways of having an impact and how to compare between them. If you think you might enjoy that too, please consider applying to work with me!

If you're interested, there are more details below. Please do comment or DM me if you have questions about the role or whether you might be suitable for it.

Our advisors talk one-on-one to talented and altruistic applicants in order to help them find the highest impact career they can.

We’ve found that experience with coaching is not necessary – everything from management consulting to global priorities research has helped someone be a good fit.

London-based role with starting salary around £65,000.

Please apply here by February 20th.

80,000 Hours’ mission

The EA community is growing in reach and now includes funding bodies with over $40 billion to allocate in total. How to turn those resources into long-term impact is the problem 80,000 Hours is trying to solve.

We’ve had over 8 million visitors to our website, and more than 3,000 people have now told us that they’ve significantly changed their career plans due to our work. 80,000 Hours is the largest single source of people getting involved in the effective altruism community, according to the most recent EA Survey.

The 1on1 team

The 1on1 team at 80,000 Hours takes people from “interested in the ideas and want to help” to “actually working to solve pressing world problems.” For example, Sophie Rose applied for advising in 2019. I chatted to her and helped her decide to focus on biosecurity and start working in the field. She co-founded One Day Sooner and is now working at the Johns Hopkins Center for Health Security – doing research to help prevent catastrophic pandemics. In 2021 we had more than 2,000 applications to speak with us from talented and altruistic people who aren’t sure how to use their skills to help the world.

Some of the most important ways our conversations help people increase their impact are:

Introductions to experts in relevant fields, hiring managers and other members of the EA community.

Talking through and reframing decisions in order to pin down people’s key uncertainties.

Suggesting new ideas — whether that’s specific jobs, new paths to explore, or ways of getting funding.

We also help with recruitment for high impact organisations, sending them lists of people we think might be a great fit for them.

While our impact per call is high, the world needs far more people in total working on the world’s most pressing problems. We therefore want to greatly increase the number of calls we have by doubling our team of advisors over the next year. Our aim is to build out an excellent system for finding the best candidates for the most impactful jobs as they are created.

What we’re looking for

It’s a great sign you’d enjoy being an 80,000 Hours advisor if you’ve enjoyed managing, mentoring or teaching. For example, Habiba Islam joined us from being a manager at the Future of Humanity Institute, Alex Lawsen was a Maths teacher, and Matt Reardon was previously a corporate lawyer.

But it’s also particularly useful for us to have a broad range of experience on the team for comparing different types of roles, so we’re excited to hear about people with all kinds of backgrounds.

The core o...

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: What's Up With Confusingly Pervasive Consequentialism?, published by Raymond Arnold on January 20, 2022 on The AI Alignment Forum.

Fictionalized/Paraphrased version of a real dialog between me and John Wentworth.

Fictionalized Me: So, in the Eliezer/Richard dialogs, Eliezer is trying to get across this idea that consequentialism deeply permeates optimization, and this is important, and that's one reason why Alignment is Hard. But something about it is confusing and slippery, and he keeps trying to explain it and it keeps not-quite-landing.

I think I get it, but I'm not sure I could explain it. Or, I'm not sure who to explain it to. I don't think I could tell who was making a mistake, where "consequentialism is secretly everywhere" is a useful concept for realizing-the-mistake.

Fictionalized John: [stares at me]

Me: Okay, I guess I'm probably supposed to try and explain this and see what happens.

Me: Okay, so the part that's confusing here is that this is supposed to be something that Eliezer thinks thoughtful, attentive people like Richard (and Paul?) aren't getting, despite them having read lots of relevant material and paying attention and being generally on board with "alignment is hard."

...so, what is a sort of mistake I could imagine a smart, thoughtful person who read the sequences making here?

My Eliezer-model imagines someone building what they think is an aligned ML system. They've trained it carefully to do things they reflectively approve of, they've put a lot of work into making it interpretable and honest. This Smart Thoughtful Researcher has read the sequences and believes that alignment is hard and whatnot. Nonetheless, they'll have failed to really grok this "consequentialism-is-more-pervasive-and-important-than-you-think" concept. And this will cause doom when they try to scale up their project to accomplish something actually hard.

I... guess what I think Eliezer thinks is that Thoughful Researcher isn't respecting inner optimizers enough. They'll have built their system to be carefully aligned, but to do anything hard, it'll end up generating inner-optimizers that aren't aligned, and the inner-optimizers will kill everyone.

John: Nod. But not quite. I think you're still missing something.

You're familiar with the arguments of convergent instrumental goals?

Me: i.e. most agents will end up wanting power/resources/self-preservation/etc?

John: Yeah.

But not only is "wanting power and self preservation" convergently instrumental. Consequentialism is convergently instrumental. Consequentialism is a (relatively) simple, effective process for accomplishing goals, so things that efficiently optimize for goals tend to approximate it.

Now, say there's something hard you want to do, like build a moon base, or cure cancer or whatever. If there were a list of all the possible plans that cure cancer, ranked by "likely to work", most of the plans that might work route through "consequentalism", and "acquire resources."

Not only that, most of the plans route through "acquire resources in a way that is unfriendly to human values." Because in the space of all possible plans, while consequentialism doesn't take that many bits to specify, human values are highly complex and take a lot of bits to specify.

Notice that I just said "in the space of all possible plans, here are the most common plans." I didn't say anything about agents choosing plans or acting in the world. Just listing the plans. And this is important because the hard part lives in the choosing of the plans.

Now, say you build an oracle AI. You've done all the things to try and make it interpretable and honest and such. If you ask it for a plan to cure cancer, what happens?

Me: I guess it gives you a plan, and... the plan probably routes through consequentialist agents acquiring power in an unfriendly wa...

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: Covid 1/20/22: Peak Omicron, published by Zvi on January 20, 2022 on LessWrong.

The Omicron and Weekly posts are now combined, so this includes the last day’s Omicron developments, which will be how it works going forward. Next week’s will include the probability updates section as well. When there is urgent news, I’ll supplement as appropriate.

This week was the peak. From here, it would be highly surprising if Omicron cases didn’t start declining. The other news is mostly more of the same, with several developments worth analyzing: The Supreme Court throws out one of Biden’s mandates but upholds the other, Djokovic gets deported a second time, a great proposal to expand Manhattan (which got its own post), and other neat stuff like that.

Executive Summary

If you live in USA and haven’t done so yet, order your free Covid-19 tests.

We are at or past the peak in Omicron infections in the United States and UK, but that doesn’t mean it’s over yet, next few weeks still rough.

UK lifts all Covid-19 restrictions, midnight is perhaps temporary after all.

Let’s run the numbers.

The Numbers

Predictions

Prediction from last week: 5.6mm cases (+13%) and 15,000 deaths (+30%).

Results: 4.9mm cases (+0%) and 12,606 deaths (+7%).

Prediction for next week: 4.4mm cases (-10%) and 14,500 deaths (+15%).

The deaths number is confusingly low this week. Cases increased by a huge amount with the three week lag, yet the deaths number isn’t increasing much. There’s about to be a much bigger three-week-lagged increase in cases, but the pattern is clear so my estimated size increase needs to adjust. I am fully prepared to miss low reasonably often here, if I wasn’t the estimate would be wrong.

For cases, it seems clear that we are peaking, so we should expect things to begin to decline. The question is how fast, with some areas not yet peaked, and the Northeast already having a much smaller share of total cases. The decline could easily be much faster than this, but things could also stay mostly steady for one more week.

Deaths

Even with Omicron being milder, the lack of more deaths is very good news here. The next week is the time when deaths should by all rights go much higher, so if they stay under control this week, we’re fully in the clear.

Cases

The Northeast already peaked, as probably did a number of other large states like Illinois, Florida and California. The South might have one more week left in aggregate, it also might not.

Cases at Yale in similar freefall to rest of east coast.

Thread pointing out that cases are declining in many places.

Trevor Bedford thread on us having peaked nationally but often not locally, including good state-by-state graphs. Expects about a symmetrical march down to the path up.

Ashish Jha thinks we have peaked.

Vaccinations

US announces (MR secondary source, StatNews article) that.

.if the Food and Drug Administration decides to update Covid-19 vaccines to take better aim at Omicron or other variants, it is unlikely to go it alone.

Instead, a senior FDA official told STAT, the agency expects to take part in an internationally coordinated program aimed at deciding if, when, and how to update Covid-19 vaccines. The approach would ensure decisions are not left solely to individual vaccine manufacturers.

“We can’t have our manufacturers going willy-nilly [saying], ‘Oh well, the EMA decided they wanted this composition, but FDA wanted that composition,’” the official said, referring to the European Medicines Agency. “So we are very much of the mind that we would like to be part of a more global process in helping to come to what vaccine composition there should be now.”

Tyler is kind:

Designed for flexbility and speedy response? I guess we’ll see. Here is the full StatNews article. And obviously, the entire public health community is up in arms about this.

Th...

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: Estimating training compute of Deep Learning models, published by Lennart Heim on January 20, 2022 on The AI Alignment Forum.

by Jaime Sevilla, Lennart Heim, Marius Hobbhahn, Tamay Besiroglu, and Anson Ho

You can find the complete article here. We provide a short summary below.

In short: To estimate the compute used to train a Deep Learning model we can either: 1) directly count the number of operations needed or 2) estimate it from GPU time.

Method 1: Counting operations in the model

2×# of connectionsOperations per forward pass×3Backward-forward adjustment×# training examples×# epochsNumber of passes

Method 2: GPU time

training time×# cores×peak FLOP/s×utilization rate

We are uncertain about what utilization rate is best, but our recommendation is 30% for Large Language Models and 40% for other models.

You can read more about method 1 here and about method 2 here.

Other parts of interest of this article include:

We argue that the ratio of operations of backward and forward pass of neural networks is often close to 2:1. More.

We discuss how the formula of method 1 changes for recurrent models. More.

We argue that dropout does not affect the number of operations per forward and backward pass. More.

We have elaborated a table with parameter and operation counts for common neural network layers. More.

We give a detailed example of method 1. More.

We discuss commonly used number representation formats in ML. More.

We share an estimate of the average performance of GPU cards each year. More.

We share some reported GPU usages in real experiments. More.

We compare both methods and conclude they result in similar estimates. More.

We discuss the use of profilers to measure compute. More.

Complete Article

You can find the article here.

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: Land Ho!, published by Zvi on January 20, 2022 on LessWrong.

I love this modest proposal so much I am making an exception to my no-New-York-Times rule, and split it off from what was going to be a bonus section in the weekly Covid post. Time to think big.

Both Mayors Bill de Blasio and Michael Bloomberg offered climate-change plans that included extending the shoreline along the East River in Lower Manhattan. But these proposals, while admirable, would be small steps and would hardly make a dent with problems of such big scale.

This new proposal offers significant protection against surges while also creating new housing. To do this, it extends Manhattan into New York Harbor by 1,760 acres. This landfill development, like many others in the city’s past, would reshape the southern Manhattan shoreline. We can call the created area New Mannahatta (drawn from the name the Lenape gave to Manhattan).

It quickly became fashionable to mock this proposal. This is a sign of civilizational decline and inadequacy, because we used to do this kind of thing all the time and (unless there’s some logistical problem I’m missing that on one is talking about) it’s obviously insanely great.

The objections seem to be:

Ha ha.

Get a load of the nerd.

Who thinks America can do things, like, ever.

New York, in particular, ha ha.

Also, how would more housing help with a shortage of housing?

And how dare you build things when climate change is coming?

Evil bunch of idiot nerds.

I volunteer to be the mayoral candidate of the New Amsterdam (which is the obviously correct neighborhood name if we’re not making a multi-billion-dollar endorsement deal) party. My other platforms will include free and expanded subway service paid for at least in part by a 3%/year property tax surcharge on vacant apartments and houses, ending rent control on all new or vacant apartments, approval of all new construction everywhere, permanent outdoor dining, and ending of all city-based occupational licensing requirements. Construction costs will be paid for temporarily by a bond issue, then repaid with interest by a 100% tax on the unimproved value of the newly created land.

This was a proposal in The New York Times, so of course it was necessary to lead with talk about protection against climate change. That did not of course stop stupid people from saying how stupid it would be to build things at sea level, no matter how much this was used as part of a project to protect the city, because guarding against climate change doesn’t involve the proper moral repentance, so it doesn’t count. The link is a New York Post article where this is the best objection they could find.

Not everyone is on board, however.

In a response piece for Curbed, writer Willy Blackmore pointed out that the “closest counterpart” to Barr’s New Mannahatta are the landfill-based neighborhoods that Robert Moses constructed along the Jamaica Bay waterfront.

“How have those fared over the years?” Blackmore quipped. “Queen’s Broad Channel — with its houses on stilts extending over marsh-grass-dotted shallows — is arguably the most flood-prone neighborhood in the city, and has the highest proportion of repeat flood-insurance claims.”

Did you know that you can use better designs instead of worse designs, and you can be more compatible with life in your creations than Robert Moses? Literally the previous sentence is this:

“Building the land at a higher elevation would further improve its protective ability, and the new peninsula could recreate historic ecologies and erect environmental and ecological research centers dedicated to improving the quality of New York’s natural world,” he said.

Ah, indulgences and offsets that make it clear that people’s true objections always lied elsewhere.

Urbanization looks like humans doing useful things, so even though it ...

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: Positly covid survey: long covid, published by KatjaGrace on January 18, 2022 on LessWrong.

Here are some more careful results from a survey I ran the other day on Positly, to test whether it’s trivial to find people who have had their lives seriously impacted by long covid, and to get a better sense of the distribution of what people mean by things like ‘brain fog’, in bigger, vaguer, research efforts.

Respondents are from the US, aged 20-40, and decided to take the survey—entitled ‘Lifestyle 2022’ with a longer description mentioning the pandemic—when offered good compensation for doing so, a couple of days ago.

Most basic results

228 responses

56 people thought they have had covid; seven or eight of them on multiple occasions

12 had ongoing health problems or symptoms after recovering from acute illness (21% of those infected)

This was the basic long covid question:

Do you seem to have ongoing health problems or symptoms after recovering from acute covid illness?

In retrospect, it seems misinterpretable, but looking at other responses (see very bottom for all questions, or section ‘How bad are these long covid cases’ for more details on most of them), I don’t think they misinterpreted it much.

Long covid from cases over two months ago, by vaccination status

The most basic results are not so telling, because maybe all of those people with ongoing problems are among the masses recovering in recent weeks, and will recover from everything before long. So here we’ll just look at people who were infected more than two months ago.

Also, let’s divide up rates by vaccination status at the time of infection.

Minor details:

These results are about cases rather than people, i.e. people with two cases are registered twice below (I asked about the date, vaccination status and health consequences for each, so e.g. one person says they got long covid from the second case, so the first is counted under ‘no LC’).

‘Vaccinated’ = at least a full one or two shot course for two weeks

Results:

All Vac No vac (Vac N/A) All 39 6 31 2 LC 9 2 7 0 no LC 30 4 24 2 P(LC) 0.23 0.33 0.23

In sum, in this (small) group:

Around quarter of cases of covid >2 months ago were reported to precede some kind of ongoing health problems or symptoms

Around a sixth of covid cases happened to people vaccinated at the time

For vaccinated cases, the rate of long covid was about the same as for unvaccinated

Or to put it visually:

How bad are these long covid cases?

A thing I wanted to know was how bad these ongoing symptoms are for people. So I asked some more quantitative questions, as well as for open-ended descriptions.

There are only nine cases of long covid from more than two months ago, and I asked people a few different things about them, so it seems easier to tell you their answers than to try to summarize, especially because I think it’s helpful to see all of a person’s answers together:

(8th column heading ends: “.you were on average in 2019”; for the second column, other options that nobody in this group agreed to were ‘I am unable to walk up stairs without resting afterward’, and ‘A doctor has diagnosed me with having specific organ damage’.)

Taking answers at face value, about eight of these sound pretty bad to me, just on column Q grounds. However it seems likely that some amount of miscommunication is going on, so I trust more impressions coming from several questions. On that basis, maybe half sound like very likely a big deal to me, holistically (and most of the others plausibly so).

To be maybe continued later.

(Maybe coming up later: What’s going on with all these people who don’t want to be vaccinated? Are people still taking many precautions? Which ones? What fraction of people’s covid-infected acquaintances seem to be severely ongoingly unwell from it? —I had a lot of prepaid survey-taking m...

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: How I'm thinking about GPT-N, published by delton137 on January 17, 2022 on LessWrong.

There has been a lot of hand-wringing about accelerating AI progress within the AI safety community since OpenAI's publication of their GPT-3 and Scaling Laws papers. OpenAI's clear explication of scaling provides a justification for researchers to invest more in compute and provides a clear path forward for improving AI capabilities. Many in the AI safety community have rightly worried that this will lead to an arms race dynamic and faster timelines to AGI.

At the same time there's also an argument that the resources being directed towards scaling transformers may have counter-factually been put towards other approaches (like reverse engineering the neocortex) that are more likely to lead to existentially dangerous AI. My own personal credence on transformers slowing the time to AGI is low, maybe 20%, but I think it's important to weigh in.

There is also a growing concern within the AI safety community that simply scaling up GPT-3 by adding more data, weights, and training compute could lead to something existentially dangerous once a few other relatively simple components are added.

I have not seen the idea that scaling transformers will lead to existentially dangerous AI (after combining with a few other simple bits) defended in detail anywhere but it seems very much an idea "in the water" based on the few discussions with AI safety researchers I have been privy too. It has been alluded to various places online also:

Connor Leahy has said that a sufficiently large transformer model could serve as a powerful world model for an otherwise dumb and simple reinforcement learning agent, allowing it to rapidly learn how to do dangerous things in the world. For the record, I think this general argument is a super important point and something we should worry about, even though in this post I'll mainly be presenting reasons for skepticism.

Gwern is perhaps the most well-known promoter of scaling being something we should worry about. He says "The scaling hypothesis regards the blessings of scale as the secret of AGI: intelligence is ‘just’ simple neural units & learning algorithms applied to diverse experiences at a (currently) unreachable scale."

Observe the title of Alignment Newsletter #156: "The scaling hypothesis: a plan for building AGI". Note: I'm not sure what Rohin Shah's views are exactly, but from what I read they are pretty nuanced.

Zac Hatfield-Dodds (who later went on to do AI Safety work at Anthropic) commented on LessWrong 16 July 2021: "Now it looks like prosaic alignment might be the only kind we get, and the deadline might be very early indeed."

lennart : "The strong scaling hypothesis is stating that we only need to scale a specific architecture, to achieve transformative or superhuman capabilities — this architecture might already be available."

MIRI is famously secretive about what they are doing, but they've been pretty public that they've made a shift towards transformer alignment as a result of OpenAI's work. Eliezer Yudkowsky told me he thinks GPT-N plus "a few other things" could lead to existentially dangerous AI (personal communication that I believe is consistent with his public views as they were expressed recently in the published MIRI conversations).

I do think a GPT-N model or a close cousin could be a component of an existentially dangerous AI. A vision transformer could serve a role analogous to the visual cortex in humans. A GPT type model trained on language might even make a good "System 1" for language, although I'm little less certain about that. So it definitely makes sense to be focusing a substantial amount of resources to transformer alignment when thinking about how to reduce AI x-risk.

While I've seen a lot of posts making the bullish case o...

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: Launching the EA Groups Resource Centre, published by BrianTan on January 19, 2022 on The Effective Altruism Forum.

CEA has launched the EA Groups Resource Centre - a website with curated resources on how people can start and run a successful effective altruism group.

Whether you’re starting or running a city, university, or other type of EA group, we think you can find resources that can help you in the Resource Centre. Check it out now at resources.eagroups.org!

Marisa Jurczyk, Jessica McCurdy, Catherine Low, and I, with the help of some volunteers, have transferred over the resources for group organisers that were originally on the EA Hub. (The list of all EA groups is still on the EA Hub though.) We’ve also condensed and updated some of the pages, as well as added new pages to the Resource Centre.

The goal is for this website to be an easy-to-search and living database of resources for group organisers, with specific guides on some frequently requested topics. We think it’s an improvement on the website before, but we still plan to add more content and consolidate more resources here. For the time being, we still think it’s the best place to go to search for helpful guides or resources related to topics you’re facing.

Updated resources on the Resource Centre

In particular, we are excited about updated resources on:

Funding for groups

Getting EA books for your members

Creating a strategy

Running fellowships & multi-week programs (with sample curriculums)

Why and how to do one-on-ones

Why did we transfer out of the EA Hub?

The EA Hub is adopting a narrower focus that doesn’t include maintaining group organiser resources. It also wasn’t as easy to edit or add new resources there, so we moved to this new site. We have set up redirects from all the old EA Hub group resource pages to point to the Resource Centre, so all of the original EA Hub links should still take you to similar resources. The navigation of the Resource Centre is slightly different but should still be intuitive.

Feedback is welcome

The site has a couple of minor issues, but there may be others we are not aware of yet, so feel free to point them out. We will be improving the Resource Centre in the coming months, and we’ll announce in the EA Groups Newsletter when there are new or updated pages. (We hope to make it even more easily accessible and less overwhelming.)

If you have any questions, comments, or suggestions about the website, including resources you want us to add or create, feel free to submit them in this form. You can also comment feedback below or email them to me at [email protected].

If group organisers have any work-in-progress resources, or want to write new ones for the Resource Centre, I can take a look and give feedback on those too. I can also anonymize resources if needed before sharing with others. Thanks!

We’d like to thank the following people for their help in launching the Resource Centre: Kirsten Angeles, Jenna Ong, Sanat Singhal, Richard Annilo, Nancy Shi, Helene Kortschak, Alex Berezhnoi, and Sebastian Becker.

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: Research idea: Evaluate the IGM economic experts panel, published by MaxRa on January 18, 2022 on The Effective Altruism Forum.

Surveying scientific experts seems like a promising way to gather informed opinions relevant for important decisions, and to improve common knowledge on important topics.

The IGM economic experts panel has been doing surveys for roughly 10 years, and I wonder how well both the surveyed consensus and the format of the platform hold up, as it might be a good idea to

copy the general IGM platform to other scientific fields,

offer ideas to improve the IGM panel, and more ambitiously to improve the economics consensus (e.g. by spotting biases), and

use the survey results when they interact with topics relevant for EA.

Two ideas how to evaluate the survey's usefulness

Look through IGM's survey questions and evaluate those which have something like a "correct" answer in retrospect. This requires somebody fairly informed about the surveyed issues.

Interview someone from IGM, or a researcher knowledgeable about expert consensuses in economics or generally (suggestions?).

Two articles about the IGM survey

Tyler Cowen evaluating some IGM polls in 2015: "But quite seriously, my opinion of the professional consensus — on topics outside an individual’s research specialty — really has gone down as a result of these polls." (link)

discussion of biases in the economics elite, in the surveyed IGM questions and for which experts were selected (paper from 2016)

IGM asking softball questions apparently aiming at eliciting textbook answers

some schools of economics not being represented, e.g. the Austrian school

clustering of responses according to political partisanship and departmental affiliations

Related questions that might be interesting

what has been the impact of the IGM survey?

did it influence policy makers? did it influence public discourse?

does it have an impact on the future research topics of surveyed researchers?

which fields are most promising candidates for a similar survey platform? are there already similar surveys in other fields?

what influences the buy-in that the IGM experts panel enjoys from academics?

what tradeoffs are involved in different formats of knowledge elicitation?

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