All of david_reinstein's Comments + Replies

Amanda Metskas and Clare Harris will present on this today by the way -- https://www.unjournal.org/news/transparent-replications-and-open-science

Great video, will share! 

One question – in the interviews you incorporated, people stated that they thought it would be very expensive to save a life (£100k+) and even more (!) internationally. 

Was this the norm? Because in the academic research I've seen people tend to state very low amounts, vastly underestimating the true (~£5k) cost.   (This also seems to be happening iirc in my own ongoing work with Janek Kretschmer and Paul Smeets. Also why I was interested in seeing someone develop a "how much does it cost to save a life quiz and calc... (read more)

Thanks for this detailed report. It's likely to be helpful to other organizations to understand the reasoning and evidence base behind this in considering whether to start or fund adjacent projects. Il

Some things that woukd also be nice com apologise if you already did this.

Can you share your data and code or spread sheets in case other researchers or founders want to revisit this? potentially this is something students and academic researchers would want to help you with.

You often report pre-values and a “lack of significant difference". Of course, this c... (read more)

Thanks for this. We are trying to prioritize this work for evaluation, feedback and rating at Unjournal.org. Aiming to incorporate your suggestions soon.

We now have a good team in this area (still looking for more).

We're now particular interested in people submitting and suggesting research in this area for The Unjournal to evaluate.

I'm going through the hosted paper ("Forecasting Existential Risks") and making some comments in hypothes.is (see here). 

I first thought I saw something off, but now see that it's because of the difference between   total extinction risk vs catastrophic risk. For the latter, the superforecasters are not so different from the domain experts (about 2:1). Perhaps this could be emphasized more. 

Putting this in a 'data notebook/dashboard' presentation could be helpful in seeing these distinctions.

Could this could be made even close to cost-effective and scaleable? If so, I think it has strong potential appeal. Perhaps not so much to hard-core rationalists and EAs, but as a bridge to making mainstream donation more effective. From this perspective, I'm more optimistic about your 'hands on charity' proposal. 

I discussed this concept in a 2021 post a while back (see especially 'my proposal sketch'). Wonder what you think.

Curious about the "rescue meat" thing.

My take is that buying "about-to-expire meat on discount when in the grocery store" incentivizes meat production less than buying expensive super-fresh premium meat.

On the other hand, stores that visibly see meat rotting on the shelves may be (emotionally?) inclined to reduce their meat orders in the future.

What does raisins have to do with "building muscle in my core"?

Yes to that footnote but the original abbreviation is confusing. It should be something like “disease adjustments to life years” .. not “disease adjusted life years”. Bc life years are good in general.

Ideally, someone who is an expert both in economic growth theory and existential risk would do a really deep analysis of the model presented in the paper, but in the absence of this we feel that giving our thoughts on this is useful.

Do you still believe this is the case? Any updates on the relevance etc?

A pretty good AI generated summary of my largely AI generated post.

I mostly agree with you on the 2nd order consequences. But also, I think a bit of feedback is usually justified even considering the first-order consequences, as I mainly argued in the comment here to Linch's post, and others had similar comments.

Another perspective: many grant applicants and potentially impactful entrepreneurial EAs may waste a lot of time exploring a very dark space. They may spend a lot of time writing and rewriting proposals. 

They do not know whether they are 'close to being fundable' or very far from it, so they don't know:

- When... (read more)

Thanks, this is relevant for researchers and people funding research and prioritizing/evaluating it. This includes Unjournal.org; we are looking to prioritize the evaluation of research relevant to animal welfare, and we have built/are building a 'field specialist' team focusing on this. 

 Some expansion on the theory of change/paths to impact/logic model for some of the leading cases could be particularly helpful. (You mention we should reach out to Martin Gould on this -- I plan to do so.)

While some of these might be amenable to simpler 'desk re... (read more)

Any chance for a Slack plugin?

Some ~first impressions on the writeup and implementation here.    I think you have recognized these issue to an extent, but I hope another impression is useful. I hope to dig in more. 

(Note, I'm particularly interested in this because I'm thinking about how to prioritize research for Unjournal.org to commission for evaluation.)

I generally agree with this approach and it seems to be really going in the right direction. The calculations here seem great as a start, mostly following what I imagine is best practive, and they seem very well docum... (read more)

Maybe 'soon' is slightly too strong, it may take a few years for the tech and the culture to adapt. But voice recognition, translation and language/grammar tools are advancing very quickly.

I imagine in-ear devices that give you quick summaries and translations, and suggest responses or real-time adjustments to what you are saying. And people will become more OK with using these tools in conversation.

It will be able to do this soon I think

Does anyone have any data or anecdotal evidence on how often people have actually stayed in EA houses that sprung from this list?

Anyone know if there is a more web-based version of this paper/research? The 754 page pdf seems like possibly not the best format.

Audio recordings would be good, thanks.

Not sure about the benefit/cost. Am I naive to think something like:

• Tripod (or a small stabilizer on a desk)
• Volunteer (or paid person) in each room, sits at front or operates tripod
• Uses own camera phone
• Uploads to YouTube directly from phone

Time cost: Maybe 1-2 hours of 'equivalent extra person work' per 1-hour session (say 90 minutes).

Benefit: If even 5-10 people watch the videos, I suspect the value outweighs the cost.

• Enabling them to shift time; e.g., do 1-on-1's if attending ...

• Encouraging some people

These are useful, thanks. I would suggest we also enable/permit a lower-quality recording to be posted or shared of the other talks. It should be fairly costless to have a few people record and post these with camera phones, etc., and I believe it would add substantial value.

I think this was downvoted because of a lack of reasoning transparency.

it makes the most sense to donate to them only when someone is going to 3x or more match your donation

Maybe  it makes the 'most' sense relative to donating at another time, but is 
their impactful per-dollar even close to comparable with other top charities?

Also, when does that match occur and is it really a counterfactual match?

, I recommend investing the money you plan to donate so that you donate more.

On what basis? How do you know the investment will outperform the relative... (read more)

I didn't end up getting around to do a more formal survey. Obviously the one in the approval voting above is deeply unscientific and doesn't represent any particular defined sample. 
 

Just a quick note as The Unjournal was mentioned. We commission expert peer review and rating (and pay the evaluators) and all evaluation is made public. We focus on potentially-impactful work in economics, social science, and policy. We are aiming at a standard and metrics that will be comparable and can be benchmarked against the traditional journal tiers, as well as ratings and adding value on other dimensions.

Submitting your work to The Unjournal basically does not preclude you from also submitting it to anywhere else. We don't 'publish' your work or c... (read more)

Thanks, this is helpful. I'll reach out. Fwiw I've added some public comments on the pdf using hypothes.is (although some of these comments are Unjournal-specific). 

All hypothes.is comments on that hosted page are visible with this search query 

or here, in context if you have the hypothes.is plugin.

There'svery little actual substance in this blog. Mainly allusions and shorthand for things in the author's brain, combined with a disdain for government intervention.

This sounds very welcome. I wanted to mention that The Unjournal is looking to expand our agenda and our team of field specialists into this area in particular. We have a few economists with an interest in animal welfare involved, but we are looking for a few more to have a sort of quorum. Please reach out if you are interested. 

I really like this idea. Fwiw, there are some connections to the "Corporate skills charity bake sale" I proposed a while back. Those ideas might synergize with this.

There's a third option to get your research in Google Scholar etc. and get feedback and credible ratings and evaluation without all of the "tiresome process" and "extractive and time consuming" (and format-limiting) features of traditional journals.

Post your work on the web, link it/add i9t to an archive, get a DOI (through Zenodo etc.)

Submit it to be publicly evaluated by The Unjournal or another journal-independent evaluation system (Peer Communities In also have some good options).  

Relevant Forum Wiki Link

Preview the slides here if you like, and add comments and questions on them if you feel inclined.

(Note: there will also be a "Slido" for queuing up any questions).

I think you meant Michael not Jason Kremer

Did you use or are you aware of any good quantifiable conceptualizations and breakdowns of the value of research that could be applied to empirical and applied work?

E.g., (probability of being true)*(value if true) seems inadequate, as what is important is the VOI gain the research yielded. And good research generally doesn't state "we proved X is true with 100% probability" but reports parameters, confidence/credible intervals, etc.

One might consider a VOI model in terms of the 'increase in value of the optimal funding and policy decisions as informed by ... (read more)

2. Shouldn't we be concerned that consumers who think they are 'donating' through these product purchases will reduce their other charitable giving?

[I think there's a good response to this, but it seems worth addressing explicitly.]

I think some of my other critiques and suggestions (note I am fairly positive on this idea, just trying to red team in a helpful way) were not addressed. Will list in separate comments to enable discussion.

1. Low-lying-er fruit: Why not (first) target billionaires who are already philanthropists and who own consumer facing companies, and encourage them to market their products in such a way? Seems easier, higher value, less risky, quicker than trying to start new companies?

The response partially addresses what was intended by my comment. Perhaps I was not complete enough in the original comment.

Note that If the big companies are differentiated in some way (like 'monopolistic competition' suggests, there could be a substantial cost to consumers (and to efficiency) to choosing the 'charity supporting brand'

With monopolistic competition, consumers have different tastes or 'locations' and firms enter with differentiated products (or in 'different locations'). Each firm ends up with some market power and thus charges a price ... (read more)

And some speakers/orgs might be willing to subsidise av and av editing production costs for their own sessions. It’s something they can link to on their website for promo, onboarding etc.

Encourage you to revisit the idea of venues in smaller/less expensive cities. In addition to lower venue costs this imposes much lower accomodation costs on participants or their organizations.

May also have good signaling benefits and attract people who may be underrepresented in the usual EA “bubble”. See all the posts discussing why Oxford, Boston, and SF bay should not be the main/only hubs for further arguments.

Also may allow larger quieter more relaxed spaces enabling calmer conversations and impromptu meetings and work sessions on and off site.

AV cost surprised me. I think I’m on board with using less professional AB but I would hate to see fewer sessions recorded. This seems like a false economy. If few people are watching or listening to them, it might be because they are not promoted well … If they could be made prominent and easily accessible, and even offer comment spaces that the speakers respond to, I think this Could add a lot of value and substitute for conference attendance for some people

Going through this list and the annotated contents, I notice some important gaps in the content that is available (or at least the content that is tracked):

• Almost no economics-based syllabi (such as Phil Trammel's Topics in Economic Theory & Global Prioritization)

• Limited syllabi/courses focusing on empirics

• Few syllabi outside philosophy that engage academic literature.

(David Rhys-Bernard's syllabus basically meets all of the above)

Could you explain this a bit more? It's very shorthand, and hard to know what you are doing and what you are asking us. I think I have a lot of comments, but for most of them I worry I might be missing the point.

E.g., a math proof is something different than scientific evidence, and it generally applies in different domains. If I have confidence in the proof itself (i.e., proof not in error), that would make any other evidence moot. However, in most relevant cases the 'math proof' is a proof of something that is only a very simplified model of the question at hand.

Some questions/comments from a quick reading of this and the link, hope it's OK.

1. Does this analysis assume that Smoke Free Israel's campaign was pivotal in the tax reform? How can that be justified? Did you consider that it might have just led the reform to occur slightly earlier?

2. How confident are you that there is really room-for-more-funding for the next campaign, or will SFI likely get enough funds to run it even without EA involvement?

Given what (I think) is limited room for more funding, perhaps you should be highlighting and boosting the othe... (read more)

Phew. Please fix when you have a moment thanks. (Otherwise people may start to think they are not understanding things and give up reading.)

There’s a 1 in a googol chance that he’ll blackmail someone who would give in to the blackmail and a googol-1/googol chance that he’ll blackmail someone who won’t give in to the blackmail.

Did you mean the opposite of this? Sounds like you are saying he would almost never blackmail someone who WOULD give in and almost always blackmail someone who WOULDNT give in.

Could we start a Role-Economics channel on the "EA Anywhere" Slack please? (That's the 'one slack to rule them all')

Is this being maintained? Any other projects in this space to keep track of? Thanks.