Sarah Cheng

Karma: 1,845

I worked as a software/product engineer at the Centre for Effective Altruism for three years, and recently became the Interim EA Forum Project Lead. If you’d like to support our work, sign up for a 30 min user interview with someone on our team. Hearing about your experience with the Forum helps us improve the site for everyone.

In general, we’d be happy to hear any feedback you have! :) Feel free to contact us or post in this suggestion thread. You can also give us anonymous feedback via this form.

Sarah Cheng 17 Dec 2024 19:39 UTC
5 points
1 ∶ 0
on: The Most Valuable Dollars Aren’t Owned By Us
I think this is a great point (though I also feel that reasonable people can disagree and it’s hard to know for sure), and I really appreciate the call to action and tips! :) I definitely agree that personal outreach to people that know you can be quite valuable.
A few related points:
1. I think overall, EAs donate a surprisingly small amount (this may be outdated info though), so probably by making effective giving more salient in EA spaces you can influence a fair amount of giving.
2. People (including myself) are lazy (or busy) and I’m sure many EAs don’t dedicate a significant amount of time optimizing their donations. I’ve personally been moved to give significant amounts by reading Marginal Funding Week posts — I think it can be really helpful to meet people where they are at (i.e. they care but don’t have a lot of time to spend on research).
3. I’ve been pushing the Forum team to do more Giving Season-related outreach on other platforms, because I also feel more optimistic about the value of influencing donations in spaces that are more “EA-sympathetic” than core (for example, posting in the EA subreddit, on our Forum Twitter account, and our new Forum Instagram account).
  1. Perhaps I would guess that the most valuable dollars are owned by people in these “EA-sympathetic” spaces.

Sarah Cheng 17 Dec 2024 15:30 UTC
12 points
0 ∶ 0
on: Sarah Cheng’s Quick takes
@Toby Tremlett🔹 and I will host an EA Forum Team office hours / co-working session in EA Gather this Thursday, Dec 19! We’ll be in this pod between 2-3 PM UK time (9-10 AM Eastern Time). Feel free to stop by to ask us questions, work on a Forum post, or just co-work alongside us. :)
We may continue hosting these, perhaps for longer slots than just one hour. Let us know in the comments or over DM if you’d like to attend one in the future.

Sarah Cheng 13 Dec 2024 16:54 UTC
2 points
0 ∶ 0
in reply to: Jan_Kulveit’s comment on: Jan_Kulveit’s Quick takes
Quick update: I read your post more closely and I actually still feel undecided, so I let my colleague break the tie and they voted not community, so I’ve removed the tag. :)
To clarify: I tried hard to apply the current definition. I expect we will revisit this definition and system in the near future.

Sarah Cheng 13 Dec 2024 16:49 UTC
4 points
0 ∶ 0
in reply to: Toby Tremlett🔹’s comment on: Jan_Kulveit’s Quick takes
I agree that I don’t think it’s in-practice relegating content. Many of the highest karma posts are still “Community” posts. I’ve heard a user tell me that they just scroll down to the “Community” section when they come to the Forum (possibly a joke but this feels like a joke that has some truth to it). I think it does mean that newcomers to the site will be less likely to see those posts, but that is intentional (we think those posts are less relevant to newcomers).
I also agree that the intention was to essentially “correct” how much attention these kinds of posts were getting, because using only karma (plus recency) for sorting meant that they were overweighted. I believe the “Community” section is still valuable for essentially the same reasons as before (for example, I think it would be a mistake to assume that there will not be major drama/scandals discussed here in the future, and I think that would be worse without a “Community” section).
However, I don’t have a strong opinion about the definition of “Community”, nor even if this section should remain “Community” or if it should have some totally different criteria. I’m quite supportive of criticism so I could see the case for updating this system to put more criticism of EA in the Frontpage. It’s likely our team will revisit this in the next few months.

Sarah Cheng 13 Dec 2024 1:37 UTC
2 points
0 ∶ 0
on: GiveWell’s Re-evaluation of the impact of unconditional cash transfers
Note that there was some discussion of this in a previous post

Sarah Cheng 13 Dec 2024 0:26 UTC
30 points
3 ∶ 0
in reply to: Jan_Kulveit’s comment on: Jan_Kulveit’s Quick takes
Hi Jan, my apologies for the frustrating experience. The Forum team has reduced both our FTEs and moderation/facilitator capacity over the past year — in particular, currently the categorization of “Community” posts is done mostly by LLM judgement with a bit of human oversight. I personally think that this system makes too many mistakes, but I have not found time to prioritize fixing it.
In the meantime, if you ever encounter any issues (such as miscategorized posts) or if you have any questions for the Forum team, I encourage you to contact us, or you can message myself or @Toby Tremlett🔹 directly via the Forum. We’re happy to work with you to resolve any issues.
For what it’s worth, here is my (lightly-held) opinion based on the current definition^[1] of “Community” posts:
The community topic covers posts about the effective altruism community, as well as applying EA in one’s personal life. The tag also applies to posts about the Forum itself, since this is a community space. You should use the community tag if one of the following things is true:
- The post is about EA as a cultural phenomenon (as opposed to EA as a project of doing good)
- The post is about norms, attitudes or practices you’d like to see more or less of within the EA community
- The post would be irrelevant to someone who was interested in doing good effectively, but NOT interested in the effective altruism community
- The post concerns an ongoing conversation, scandal or discourse that would not be relevant to someone who doesn’t care about the EA community.
I agree that the two posts about uni group funding are “Community” posts because they are “irrelevant to someone who was interested in doing good effectively, but NOT interested in the effective altruism community”. I’ve tagged them as such.
I would say that the EAG application bar post is a borderline case^[2], but I lean towards agreeing that it’s “Community” because it’s mostly addressed towards people in the community. I’ve tagged it as such.
I skimmed your post on LW and I think it was categorized as “Community” because it arguably “concerns an ongoing conversation, scandal or discourse that would not be relevant to someone who doesn’t care about the EA community” (as the post references past criticisms of EA, which someone who wasn’t involved in the community wouldn’t have context on). I think this is not a clear cut case. Often the “Community” tag requires some judgement calls. If you wanted to post it on the Forum again, I could read it more carefully and make a decision on it myself — let me know if so.
1. ^
  To be clear, I haven’t put enough thought into this definition to feel confident agreeing or disagreeing with it. I’m just going to apply it as written for now. I expect that our team will revisit this within the next few months.
2. ^
  Partly because I believe the intended audience is people who are not really involved with the EA community but would be valuable additions to an EA Global conference (and also I think you don’t need to know anything about the EA community to find that post valuable), and so the post doesn’t 100% fit any of the four criteria.

Sarah Cheng 5 Dec 2024 2:39 UTC
6 points
0 ∶ 0
in reply to: Sarah Cheng’s comment on: Sarah Cheng’s Quick takes
I’ve just updated the doc with our final OKRs of the year, was somewhat delayed due to a team retreat and Thanksgiving.

Sarah Cheng 5 Dec 2024 0:51 UTC
5 points
1 ∶ 0
on: Is the AI Doomsday Narrative the Product of a Big Tech Conspiracy?
Thank you for writing this! I’m surprised at how often this take comes up in lefty spaces, and I appreciate you addressing it.

Sarah Cheng 5 Dec 2024 0:03 UTC
2 points
0 ∶ 0
in reply to: Ben Millwood🔸’s comment on: Sarah Cheng’s Quick takes
Ah sorry, I don’t think there’s any additions on the markdown side.
As-is it often feels like the markdown editor is a second-class citizen.
Yeah I think this is true in some sense. CKEditor does get more attention in our dev work. Partly that’s because it’s the default, and is used much more often (I ran a quick query to sense check this, and it looks like about ¹⁄₉ of things written on the Forum in 2024 are in markdown, ⁷⁄₉ are using CKEditor, and ¹⁄₉ are other^[1]). I don’t think this tells the whole story though, so I’m open to pitches for us to enhance the markdown editor. :) Though note that we have significantly less engineering capacity now.
1. ^
  Mostly admin-specific or legacy things

Sarah Cheng 3 Dec 2024 18:17 UTC
24 points
0 ∶ 0
on: Sarah Cheng’s Quick takes
Mini EA Forum Update
We now have a unified @mention feature in our editor! You can use it to add links to posts, tags, and users. Thanks so much to @Vlad Sitalo — both for the GitHub PR introducing this feature, and for time and again making useful improvements to our open source codebase. 💜

Sarah Cheng 3 Dec 2024 4:04 UTC
3 points
1 ∶ 0
in reply to: ethai’s comment on: Donation Election Discussion Thread
I think those are both good points. In my experience, different subsets of EA can vary a lot in terms of cause prioritization. Though I’m guessing that Aaron means something slightly different than you do when he says “consensus EA”.

Sarah Cheng 3 Dec 2024 3:54 UTC
4 points
0 ∶ 0
in reply to: Jason’s comment on: Donation Election Discussion Thread
the risk that the results have been affected by off-Forum advocacy and organizing
I’m not sure if this addresses your concern, but just want to clarify that accounts can only vote if they were created before Oct 22, 2024. I think that having to have created an account prior to the announcement of the donation election is a medium bar (at least I think it’s higher than Manifund’s event was) — it’s quite easy to use the Forum without creating an account, so people who create an account tend not to be casual readers.
I do think it would be interesting to compare the overall results with those of the subset of users who have earned at least some karma.

Sarah Cheng 3 Dec 2024 3:40 UTC
12 points
0 ∶ 0
in reply to: Toby Tremlett🔹’s comment on: Donation Election Discussion Thread
Personally I found it really helpful to listen through the Spotify playlist of Marginal Funding Week posts during a recent flight. (Note that unfortunately many of the candidates aren’t in that playlist, apologies!)
In particular, I found @Holly Elmore ⏸️ 🔸’s PauseAI post surprisingly convincing, given that I am generally a bit skeptical of pausing AI as a strategy for reducing risk. I’m also sympathetic to the argument that the EA Forum should allocate our money to charities that are harder to fundraise for elsewhere, and PauseAI scores well there. So I’ve made PauseAI my top choice, which I didn’t spend too much time debating since my one vote isn’t currently the winning one. I may revisit my vote if that changes.
In general I recommend listening to the Spotify playlist (or even reading through all the Marginal Funding Week posts/comments if you’re extra virtuous!) — I found the experience both inspiring (hearing about all the work people are doing to improve the world) and sad (realizing how much more money we need to do all these things).

Sarah Cheng 3 Dec 2024 3:01 UTC
6 points
1 ∶ 0
on: A case for donating to AI risk reduction (including if you work in AI)
I think this post makes some great points, thanks for sharing! :) And I think it’s especially helpful to hear from your perspective as someone who does grantmaking at OP.
I really appreciate the addition of concrete examples. In fact, I would love to hear more examples if you have time — since you do this kind of research as your job I’m sure you have valuable insights to share, and I expect that you can shift the donations of readers. I’d also be curious to hear where you personally donate, but no pressure, I totally understand if you’d prefer to keep that private.
Work in sub-areas that major funders have decided not to fund
I feel like this is an important point. Do you have any specific AI risk reduction sub-areas in mind?

Sarah Cheng 20 Nov 2024 19:53 UTC
5 points
0 ∶ 0
in reply to: Kenneth_Diao’s comment on: Donation Election Discussion Thread
Thanks so much for flagging this, and really sorry for the mistakes. I’ve gone through and updated both, hopefully they are now both up-to-date. Please let me know if you see any other issues.

Sarah Cheng 8 Nov 2024 20:54 UTC
30 points
11 ∶ 1
on: A Critique of Animal Charity Evaluators (ACE)
Thank you for doing this work. I’m very supportive of productive criticism on the Forum. As a moderator, I’d like to recommend this post for tips on how to make criticism more productive. EA is a collective project, and I think that steps such as sharing this feedback with ACE directly and writing a less aggressive title for your post would improve the outcomes of this work.

Sarah Cheng 7 Nov 2024 4:41 UTC
2 points
0 ∶ 0
in reply to: Anthony DiGiovanni’s comment on: Winning isn’t enough
Does Jesse have an EA Forum account? We can manually add them as a co-author — feel free to contact us with Jesse’s account info. :)

Sarah Cheng 28 Oct 2024 20:07 UTC
2 points
1 ∶ 0
in reply to: Gemma 🔸’s comment on: Gemma Paterson’s Quick takes
Appreciate the response! To be clear, I am genuinely glad that you wrote the quick take, so I don’t want to discourage you from doing more off-the-cuff quick takes in the future. Hopefully hearing my perspective was helpful as well. I’m glad to hear that you don’t think we are actually fighting. :)
On collaborating/cutting costs: my outside impression is that AIM is already quite good about keeping their costs low and is not shy about being proactive. So my view is something like, if they thought it would be good for the world to collaborate more closely with CEA, I trust that they would have acted upon that belief. This is something that I respect about AIM (at least the version that’s in my head, since I don’t know them).

Sarah Cheng 28 Oct 2024 18:25 UTC
12 points
0 ∶ 0
on: EA: Renaissance or Diaspora?
Thank you for sharing this Timothy. I left a long comment on the LW version of the post. I’m happy to talk about this more with you or Elizabeth — if you’re interested, you’re welcome to reach out to me directly.

Sarah Cheng 28 Oct 2024 18:20 UTC
8 points
0 ∶ 0
in reply to: JWS 🔸’s comment on: JWS’s Shortform
I agree that it’s useful to make this more visible here, thank you!
[Flagging that I only know my own very limited perspective, and I don’t expect that Lincoln has done anything wrong, it’s just that I don’t see much from where I’m sitting] I found that statement a bit confusing, because as a person who works at CEA, I barely knew that Lincoln was on the EV board, and I have never heard about him interacting with anyone at CEA (though it seems likely that he has done this and I just don’t know about it). I feel like this is complicated because I don’t know what the best practices are wrt how involved a board member should be in the details of the organization; I could imagine there might be good reason for board members not to be reaching out to individual employees and pushing their pet causes. But he specifically said he was “worried about the EA Forum” in his comment, and yet I do not know what any of his views on the Forum are, and I have never heard of him sharing any of his views on the Forum with anyone on the CEA Online Team. So I am left feeling pretty confused. I’ll just say that I have directly reached out and offered to talk with him, because as you can imagine, I am very interested in understanding what his concerns are.

Sarah Cheng

Mini EA Forum Update