I worked as a software/product engineer at the Centre for Effective Altruism for three years, and recently became the 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.
From past user interviews, I feel like a surprising amount of Forum content gets shared. A common story is that people only really go to the Forum when a link is shared with them. However, it’s possible that that view is outdated and less sharing is happening now.
We have improved the shareability of posts in the past year or two (I forgot exactly when we did some things), like by letting authors edit the social preview image and text for their posts, and showing that image in the post hoverovers, and adding a “share” icon at the top of the post page with a dropdown of options. As far as I know those didn’t move the needle a lot, though I haven’t personally dug much into that data. I would hesitate to add more to the UI than that to encourage sharing content, because I don’t want to add clutter to the post page.
We are considering updating our default social preview image at least, so maybe that will help. :)
I feel like it’s hard to say how that impacts the Forum. Some people also crosspost their blogs onto the Forum, and our team even handles the crossposting for some authors. Some authors get more engagement on the Forum than on Substack. I’m just not sure if it overall helps or harms Forum usage.
Yeah I agree that “friendliness” and “welcomingness” are more important than “fun”, and I’m more confident that we should be prioritizing increasing those qualities.
I’m so glad that you brought this up! This is broadly the headspace I was in for like, most of the time I’ve been working on the Forum. I think these are important points and I agree that external factors can play a huge role in how much usage the Forum gets. This feels like a clear takeaway from looking at how media around FTX affected Forum usage, for example.
In early 2024, our team did a marginal impact analysis that provided some evidence that our work was cost-effective (above an internal bar that I’m not going to explain more here). This made me made me feel like our work was valuable even if the overall usage numbers were going down, because I believed that the primary (basically in my head I rounded it to 100%) driver of usage was external factors (outside of some spikes around drama). As in, it didn’t feel like the tech work we did made any significant improvements (I had less visibility on the content side so I won’t really comment on that).
This was my view going into this new role. However, chatting with @Will Howard🔹 more about Forum strategy actually changed my mind on this quite a bit. My current hypothesis is something like: the actual current bottleneck around increasing Forum usage and the value of discussions is content quality (and vibes), rather than the number of visitors to the site. For example, I believe that traffic to 80k resources is still overall increasing (I’m not 100% sure though), and a lot of their traffic should flow through and impact Forum usage (like, they link directly to Forum posts in various places). The fact that our usage and [in my low-confidence, lightly-held, subjective opinion] overall value of discussions has still gone down is a sign that we cannot lay all the blame on external factors.
Broadly, I think that users’ perception/expectations/feelings about the Forum matter quite a lot for reaching our goals, and I am currently optimistic that we can make significant improvements by focusing attention and intention on community building.
(I also think that things like public awareness and marketing do matter still, which is why our team is putting some capacity towards promoting content externally, and we will likely explore things like paid advertising later this year.)
On a more meta level: in some way I think it would be the “easy way out” to continue believing that external factors are the overwhelmingly largest influence. There is certainly data one could use to support that hypothesis, and maybe it would be easier to believe that I have very little influence and therefore very little culpability. But I also feel like, ignoring the possibility that our team could make a big difference (by focusing on community building) would be neglecting our duty. So you could view this as me taking a bet on a new hypothesis (which may or may not pan out).
I appreciate your feedback, and I agree that there’s room for improvement. I think ultimately the ideas need to come from the community rather than the Forum Team, and hopefully by putting more attention and intention around community building we can improve the rate of collective intellectual progress made on the Forum.
That said, I don’t necessarily think that new ideas is the right goal (though it depends on what one categorizes as “new”). For example, writing good summarizes of existing work (or doing things like meta-analyses) creates useful foundations for future work. Research like this may not be considered a “new idea” but builds on existing ideas and adds important data and nuance. Finding specific ways to practically implement ideas that have previously been floated is important and valuable progress.
There’s also the possibility that a enough low-hanging fruit ideas in this space have been brought up already and so it’s just by default harder for someone to come up with a new idea than it was in the past.
most people will only look at articles on subject matters that they are already familiar or on meta-level conversations
I agree that this is a thing to some extent, but I’m guessing I view this as less of a problem than you do (both that I think it’s not happening to an extreme extent [like I would guess that there are not that many people who have background knowledge on cluster headaches but this post got a fair amount of karma and comments], and that I think some of this is fine/good since people probably should spend relatively more time on posts which match their subject matter knowledge).
Thanks for this! The examples are really helpful, and your overall take resonates with me. (I acknowledge that your comment is basically just reinforcing my own view though, so maybe this is just confirmation bias — I would love to hear if anyone has any counterexamples!)
Thanks both! Yeah I do think ultimately the goals of EA and the Forum are not “fun” per say, but hopefully we can still create a space that has positivity. And I think currently, when people write about good news or progress or celebrating success, they get a relatively large amount of karma.
I also agree that we don’t want the tone to be too grim, both because that’s probably bad for motivation, and because I think that can even skew the topics that get discussed in bad ways (like if discussions were all doomy). I think that the Forum could use more cheerful content on the margin. :)
Good point — yeah the Forum has a few tags in a similar vein (like Opinion), so that’s a possibility. Personally I feel like the question of tone matters more for comments than top-level posts (like, posts are almost always serious, except around April Fools day) so I’m not sure this would make much difference. (Also I think adding more tags would make the UI more cluttered so I’m hesitant to encourage that.)
Thanks! I think the question of how much “fun” is allowed or encouraged is an important one for us to consider. Honestly I still feel unsure about how to balance this. Here are some disorganized thoughts:
I agree that if people felt more like it was fun to be on the Forum, that might increase our overall usage.
I think we can make the Forum compelling even if it doesn’t fulfill all of the reasons one might enjoy a community (for example, discussions can still be thought-provoking and feel valuable without being fun).
I think that moving away from seriousness runs the risk of making communication harder, and I think clearly communicating is quite important because this is a space where we should care about nuance and truth-seeking. I worry that the more people talk past each other and the less we understand what the other is saying, the less productive discussions will be.
I also think it matters how people who are unfamiliar with EA read and interpret what happens on the Forum, more so than other spaces (like EA Facebook groups). So, at least when just considering this reason, the importance of clear communication here is higher than other platforms.
I’m guessing that if we are otherwise successful at increasing the amount of “community” on the Forum, that one side-effect is that [groups of] users will know each other better and be more able to communicate clearly between them, so naturally there will be more writing that is informal and the overall tone will become less serious. (It’s possible that this is a chicken and egg situation though, I just don’t know.)
I generally agree with some points others brought up around how other platforms can fulfill that need (of being a culturally EA space that is more about fun).
It might just be too hard for a single space to succeed at both serious + productive discussion and being fun, in which case I think it’s more important for us to succeed at the former.
I still do want us to be a space where people can do silly and fun things (like April Fools day), partly because like, it makes me happy! :)
I wrote a bit more about seriousness and having high standards in this earlier comment as well.
I appreciate your thoughtful reflections! :)
There’s not as much new content as I’d hope, or at least easily discoverable content on the front page.
I agree with this, and I hope that we can improve the situation with a combination of community building and more actively steering discussions (for example via Forum events).
It’s very serious!
Yeah I agree this is true, but I’m not sure what (if anything) to do about it. At least, if this is a problem, it doesn’t feel like a top priority one to address. My hypothesis is that doing more community building work will address this somewhat, like if users feel more comfortable with each other and feel like they know each other better (than right now).
The standards are very high.
Agree with this one too. :) However I do think that having high standards does have benefits, and I currently think that we should actually have higher standards on the margin. I agree that there are trade-offs, and I am sad to lose out on good content purely due to people feeling intimidated.
If you’re feeling unsure about posting something, please feel free to reach out to Toby or contact the team here. We’re happy to give you feedback, and help you figure out what works on the Forum. Also, quick takes are another great solution, I def recommend using them!
There’s not a lot of disagreement.
This one’s interesting, and I feel like I don’t know how to evaluate whether this is true. I think there’s a related phenomenon where, when someone is first learning about EA, there are a ton of ideas that are new to them, and over time the rate of encountering novel ideas decreases (partly because there are less low-hanging fruit maybe?). I agree that having “more polite, thoughtful, good-faith disagreement” on the Forum would be great though, and I hope that our team can build up a community that encourages that. (If you have any specific people in mind who could bring that energy here, please reach out to Toby and send him suggestions!)
Thanks for flagging that! I want to acknowledge that, independent of the bug I mentioned earlier, there could still be an issue of users downvoting in a way that does not align with Forum norms. One thing we could do is update the wording of the tooltip that explains karma voting, since I think “How much do you like this overall?” is not exactly the right question to ask. I agree that asking whether the content is productive or like, contributing value to the discussion, is closer. I also agree that we don’t want to disincentivize people from productively saying things that they expect others to disagree with.
We have actually considered asking moderators to highlight good comments in a way similar to what you suggest. :) I’ve also considered bringing back the Forum prize but just for comments (I actually currently feel quite optimistic about experimenting with a Forum prize of some kind).
Great comment — this gets at a lot of things that I’ve been thinking about. And I appreciate you sharing your personal perspective.
drive-by professional posts
I like your description that “it breaks the illusion of a community”, that resonates with me. I also think that posts that feel too professional discourage discussion. The flip side is that there are various ways that these kinds of posts create value:
Things like job postings and org calls for donations can pretty straightforwardly contribute to making the world better by encouraging impactful action
Another way the Forum creates value is something like: readers become aware of organizations (such as via org update posts), then they later donate or get involved with the org’s work
In theory, people who do relevant work as their full-time job should produce some of the best research/ideas/work [that is relevant to the Forum audience’s interests]
So I’m thinking about whether we should, for example, move org posts to a separate section, but looking at the Frontpage now, it doesn’t feel to me like there are too many of them. I think this fluctuates so I’ll keep an eye on it.
I feel like a better long-term solution is to offset this feeling by working to increase the number of posts from individuals, and trying harder to build a community (or perhaps clusters of communities) of individuals who feel like they are in conversation with each other. Not that this is easy, but I think it’s worth us trying to do, as it feels more like we are addressing the root cause and building better foundations.
Though I agree that the notification customization UI is not the best — the wording is not super clear, and I think the way notifications are categorized is not ideal. I also agree that too many notifications is a UX problem, and we have a backlog item to explore better ways to batch notifications to improve this experience (though to be honest I’m not sure when we will get around to it).
Thanks for the suggestion! I like the idea of visually distinguishing the most important notifications (i.e. the ones you might want to reply to).
There are a lot of customization options for notifications, so I’d recommend looking through your account settings:
For example, you can set “Comments on posts/events I’m subscribed to” and “Replies to my comments” to notify you both on-site and via email.
You may also prefer to batch email notifications (rather than receive them immediately) if you get too many (for example, you could batch “Comments by users I’m subscribed to” to get an email about it once per day).
Yeah thanks for flagging this, I was going to mention the same thing. The vote count when you hover over the karma includes users who only reacted and didn’t vote on the karma portion. I do think this gives people the overall impression that there are more downvotes than there actually are. :( I put up a PR to try to fix this.
EDIT: Oops sorry I should have watched the video first, it was not the same thing I was thinking of… 😅 But I think the bug I mentioned is worse in any case so hopefully we can deploy a fix.
That’s very fair! I certainly don’t want anyone to use the Forum more than they believe is worthwhile. My guess is that a healthy relationship with the Forum looks different for different people, and I don’t think that every single reader should engage with the Forum more than they currently do. I recommend that people customize their Forum experience and customize their Frontpage to help with them find a good balance.
More on tech vs community building
I think that in all my time working on the Forum, it’s felt more like being on a typical software product team than being on a team of community builders or organizers. For example, when our team grew around late 2022, we primarily hired product and engineering capacity. (Though this feeling is at least partly because I was hired as a software engineer; I could imagine it was less the case for our various content managers.) Our success metrics tend to be things like MAUs (monthly active users), clicks, and time on page / engagement hours.
IMO, focusing on the software made more sense when the community was growing organically, and when it was growing for reasons outside of our control (like via comms/marketing around What We Owe The Future). That’s not our current situation. While I still think there are a lot of valuable improvements we can make on the tech side, right now I believe that the community side is more neglected so we should be focusing our efforts there.
Going down to 1 FTE of engineering on the Forum has been difficult. We’ve let bugs exist for longer on the site, been worse at addressing customer service requests, and have very little capacity for longer-term investments (both in terms of improving the codebase and doing feature work). These things make me personally sad, and I’m sorry to any readers who have been affected by this. As of January, we’re starting to shift capacity back towards Forum engineering, so it will probably settle closer to 1.5 FTE. Even with the focus on community building, I think that the space that holds the community matters quite a lot and we don’t want it to start hindering our work.[1]
For more on this point, see this LW comment from Habryka.
More on where I’m coming from
I joined the CEA Online Team in late 2021 as a software engineer. Most of the time I’ve been on the team, I’ve been focused on building parts of the Forum that are outside of the core reading/writing experience (such as groups and connections). I still believe in my heart that this was worth working on, and that it can be particularly valuable to build out these non-core features on the margin. I have a soft spot for making “doing good” more accessible and helping more people find ways to take concrete steps, and I think the Forum is well-placed to do that.
However, as mentioned above, I think that the value of these non-core features depends on having a strong and healthy Forum community. Faced with being actually responsible for the success of the Forum, I’ve (sadly) decided that our team needs to drop most work on non-core features, and even most of the product/engineering work on core features. We’re down to approximately 1.5 FTE of software engineering on the Online Team, and much of that is taken up by non-Forum work (such as the recent CEA.org redesign). Instead, we are focusing on increasing the amount of high-quality content on the Forum, via a combination of community building and more actively steering the Forum ourselves (such as with more themed events and revisiting our moderation processes).
Since I’m relatively new at this interim role, I feel like I still have a lot to learn, and I’m probably making some mistakes. So I will again emphasize that I am very open to and interested in feedback, especially if you have any disagreements on how/why the Forum is falling short, or disagreements on how the Online Team should be spending our time.
Nice — I like the way you described this, and I broadly agree. It’s possible that it’s more effective for the Forum Team to be doing a lot of the work to share on external sites (like via our Twitter account) so that others can contribute by lower-effort actions like retweeting, liking, and commenting (as opposed to us trying to get individual readers to share content more).