Yes, that's correct.

Yes and no -- the only concrete thing I see @WillieG having done was "sign[ing] letters of recommendation for each employee, which I later found out were used to pad visa applications." 

I would find refusing to write a letter of recommendation on "brain drain" concerns to go beyond not funding emigration efforts. I'd view this as akin to a professor refusing to write a recommendation letter for a student because they thought the graduate program to which the student wanted to apply was a poor use of resources (e.g., underwater basketweaving). Providing references for employees and students is an implied part of the role, while vetoing the employee or student's preferences based on the employer's/professor's own views is not.

In contrast, I would agree with your frame of reference if the question were whether the EA employer should help fund emigration and legal fees, or so on.

Who said we should "PaNdEr" to conservatives? That reads like a caricature of the recent post on the subject. If you're claiming that there is a pro-pandering movement afoot, please provide evidence and citations to support your assertion.

I think the significant majority of people here -- including me! -- are somewhere between unhappy to extremely upset over yesterday's events, but that doesn't justify caricaturing good-faith posts. If you have a concrete, actionable idea about how we should respond to those events, that would make for a more helpful post.

Good observation -- most of the drop in the number of new donors was seen in 2022, but little of the drop in the amount of donations from new donors happened then [$43.4MM (2021) vs $41.1MM (2022) vs. $20.5MM (2023]. Because of their size, the bulk of the 2021 --> 2022 drop was almost certainly people giving under $1,000, which is somewhat less concerning to me due to the small percentage of GiveWell's revenue that donations under $1K provide (less than 3%). There are a good number in the $1-$10K range, but they did not show a significant decline overall between 2021 and 2022. 

Presumably, the 2022 --> 2023 drop in revenue involved loss of new higher-dollar donors. My assumption is that higher-dollar donors act somewhat differently than others (e.g., I expect they engage in more due diligence / research than those donating > $1,000 on average). So it's plausible to me that the 2021 -> 2022 numerical decline and the 2022 --> 2023 volume decline have (or do not have) very similar causes. I'd guess FTX might hit higher-dollar new donors more because of the extra due diligence. 

The following chart is for all donors, not new ones:

The other number I found potentially concerning was the 50% drop in year-over-year funds from new non-anon donors (p. 10 of the 2023 metrics report, see paste below). Funds from new/non-anon donors in 2021 were slightly higher than in 2022 per the 2022 metrics report, so the prior year wasn't the anomaly. 

I don't want to over-update on a single year's Y/Y difference, but my concern would grow if 2024 ended up similar to 2023.

I would not have predicted much effect of the FTX affair on GiveWell's new donor acquisition, but it's possible that played a role.

You seem to be assuming that the primary harm of malaria deaths and (conditioned on "fetuses counted as people") of abortion is the suffering that children and fetuses experience when dying of malaria and abortion, respectively. That's an unusual assumption; I think most people would identify the primary harm as the loss of ability to live the rest of the child or fetus' life. 

So I think you're missing a step of either (1) explaining why your implied assumption above is correct, or (2) comparing human loss-of-life to chicken suffering rather than suffering to suffering as your infographic does. (In the world where factory farming ended, these chickens would likely not exist in the first place, so I wouldn't include a loss-of-enjoyable-life factor on the chicken side of the equation).

Reading Evan's comment and Sarah's response -- along with some other comments like @titotal's -- updates me to a mild-to-moderate degree toward the possibility that there may be a felt (and possibly real) need for two or more related spaces that call for mutually inconsistent design criteria. One might be more academic, formal, and rigorous while the other related space would be more flexible, open, and accessible. That feels like a big change from the status quo, and I'm hardly confident my update is directionally correct. But I think it's worth pondering whether different groups of users may be seeking things from the Forum experience that are valid, worthwhile, and yet incompatible.

Over the course of 2024 (and indeed, since early 2023), Forum usage metrics have steadily gone down^[1]. My subjective opinion was that the Forum did not meet my (perhaps too high) expectations in terms of producing valuable discussions that enable collective intellectual progress on the world’s most pressing problems^[2]. 

I would start with the assumption that this had a lot more to do with the larger zeitgeist vs. anything to do with what the Forum team did / didn't do. For instance:

• In the era with fairly accessible and expanding financial & human resources, people might have been more motivated to devote time to proposing novel and exciting stuff because they assessed a higher probability of launch feasibility;
• In the immediate post-FTX era, critical voices might have felt that the kettle was hot and that they had a better chance of getting desired reforms through vs. now;
• And so on.

Some of this is normal, inevitable, and even necessary as a social movement develops. I don't have any clear opinion on whether what you're identifying here fits into the normal/necessary bucket or the something-to-be-addressed bucket. My low-confidence guess is that there is something in both?

All that is to say that I would be cautious about weighing raw quantitative or qualitative data about the quality of Forum discussions too heavily in the Forum team's feedback loops. There is likely to be a lot of noise.

I directionally agree with the second paragraph but there are some relevant differences in my mind. First, to the extent that a large donor chose to have their donation advertised (as opposed to remaining anonymous / confidential), they can be seen as making some implied assertions (which they may or may not be consciously intending to make!):

• I am public-spirited / charitable / deserving of status and praise for what I did; and
• Others who are public-spirited / charitable should consider donating as I did.

This is particularly true if they get the concert hall named after them or something. I think we need to be somewhat gentle, but I think we're entitled to get our viewpoint out on those claims. 

In contrast, I don't think anyone who spends all their money on yachts and mansions can be reasonably seen as making these kinds of assertions. The stronger response to the ineffective donor can be seen as a means of combatting these implied messages; there is little risk of anyone misunderstanding the moral value of Jeff Bezos' non-philanthropic choices.

There's also the practical reality that the tax breaks for charitable donations in the US mean that the taxpayers (including myself) -- as a functional matter -- pay for a meaningful fraction of almost any significant charitable donation to a 501(c)(3).At some point, that gives me somewhat more of an interest in criticizing what the rich donor is claiming the tax writeoff for than in what someone is buying without a subsidy from me.

Policing strong downvotes better may be a relatively low-cost way to mitigate this. The status quo risks disincentivizes making comments to which a few people who are willing to use their strongvote hammers will react negatively.

With the caveat that underlying data are unavailable, I get the sense that some users are too trigger-happy on the strong downvote button for content with which they disagree. I've suggested requiring strong downvoters to check a box or enter text justifying their vote -- which might serve as a "stop and think" moment against reflexive use of the button. The voting norms are relatively restrictive on what rises to the level of justifying a strong downvote, although these are not exclusive:

• It contains many factual errors and bad reasoning
• It’s manipulative or breaks our norms in significant ways (consider reporting it)
• It’s literally spam (consider reporting it)

I wouldn't be opposed to giving mods the power to downgrade strong downvotes to standard ones in certain circumstances. For example, where there is a significant number of upvotes on a post or comment, that discrepancy suggests that the strong reaction of a strong downvote may be outside the range of reasonable responses to the post or comment. Requiring that kind of objective indicator would prevent mods from downgrading strong downvotes willy-nilly.