Should pretty much all content that's EA-relevant and/or created by EAs be (link)posted to the Forum?

By MichaelA🔸 @ 2021-01-15T06:56 (+46)

Currently, there's a lot of EA-relevant content created by EAs on the Forum, but also a lot of such content scattered across a wide range of other sites. And many of those sites lack some of the features I like about the EA Forum (e.g., a tagging system, ability to vote, ability to comment, a community that's more thoughtful and epistemically charitable than average).

So perhaps all EA-relevant content, all content created by EAs, or all content meeting both criteria should be posted or linkposted to the EA Forum. 

I think that that might have the following benefits:

I'm interested in whether other people agree with those basic ideas.

---

And if people do agree with those basic ideas, where do you think we should draw the line? 

I think it'd clearly be silly to linkpost every single academic article on animal consciousness or bioengineering (even just those released from now on), or to linkpost every single blogposts EAs write (even those that are just about their gardens). At the other extreme, I think it'd clearly make sense to linkpost some papers, blog posts, podcast episodes, etc. What about the content between those poles?

I think my tentative independent impression is that a good policy would be:

Linkpost any content that meets the following criteria: 

Linkpost some content that doesn't meet all of those criteria, if there are other good reasons for linkposting it. E.g., if it's a good example of non-EAs from a particular field beginning to take interest in EA-relevant issues, or if it sparked an interesting thought that you want to comment on briefly in the linkpost.

But I think that that policy would be a substantial departure from what's currently done, so maybe there are good arguments against it.

This question post was sort-of prompted by an earlier comment from EdoArad, so thanks to him for that. It was also sort-of prompted by me deciding to linkpost the latest episode from the 80,000 Hours Podcast feed, and wondering whether all such episodes should be linkposted.

Aaron Gertler @ 2021-01-15T10:12 (+19)

Eventually, we'd like it to be the case that almost all well-written EA content exists on the Forum somewhere. Maybe in full-text form, because that's more searchable, but well-tagged linkposts with summaries are a good option too.

This quarter, I plan to start doing a bunch of backfill, using the EA Forum Archives account to post a lot of material. Depending on how the material is formatted, this might be full-text or might just include an abstract or summary. (The separate account is meant to stop my karma from getting inflated, and to draw attention to the importance of archiving.)

That said, I also appreciate other people who do this! If something is linkposted before I get to it, I've saved a bit of time.

And of course, we'll need many users working in concert to source, crosspost, and tag all the new content that pops up.

evhub @ 2021-01-15T22:26 (+9)

I'd personally love to get more Alignment Forum content cross-posted to the EA Forum. Maybe some sort of automatic link-posting? Though that could pollute the EA Forum with a lot of link posts that probably should be organized separately somehow. I'd certainly be willing to start cross-posting my research to the EA Forum if that would be helpful.

Aaron Gertler @ 2021-01-18T07:20 (+10)

Instinctively, I wish that discussion on these posts could all happen on the Alignment Forum, but since who can join is limited, having discussion here as well could be nice.

I don't know whether every single post should be posted here, but it would be nice to at least have occasional posts summarizing the best recent AF content. This might look like just crossposting every new issue of the Alignment Newsletter, which is something I may start doing soon.

MichaelA @ 2021-01-16T01:17 (+4)

That sounds good!

almost all well-written EA content

So by this I assume you mean "content that's quite EA-relevant and written by EAs"?

Do you have thoughts on whether the EA Forum should also be home to a bunch of linkposts to content that's just either quite-EA relevant or written by EAs? E.g., an article on nuclear risk from a non-EA academic? Or a well-written blog post by an EA that's about philosophy or politics but not in a way that makes connections to EA focus areas very clear?

Aaron Gertler @ 2021-01-18T07:25 (+6)

I meant "quite EA-relevant and well-written". I don't especially care whether the content is written by community members, though I suppose that's slightly preferable (as community members are much more likely to respond to comments on their work).

an article on nuclear risk from a non-EA academic

Heck yeah.

a well-written blog post by an EA that's about philosophy or politics but not in a way that makes connections to EA focus areas very clear

Depends on the post. Sometimes, connections to EA become clear if you delve deep enough into a topic, or if a bunch of people with EA-related specialties read it and consider how it might apply to their work. But if there really is no clear connection at all, I'd label the post as "Personal Blog" so that people can more easily choose whether to see it on their homepage.

MichaelA @ 2021-01-19T00:45 (+6)

I don't especially care whether the content is written by community members, though I suppose that's slightly preferable (as community members are much more likely to respond to comments on their work).

>an article on nuclear risk from a non-EA academic

Heck yeah.

The main reason I sort-of suggest "written by community members" as a possible criterion for deciding whether to linkpost things here is that it seems like, without that criterion, it might be very hard to decide how much to linkpost here. There are huge numbers of articles on nuclear risk from non-EA academics. If someone decided to linkpost all of them here, or to linkpost all of the peer-reviewed non-EA articles on any one of many other EA-relevant topics, that batch of linkposts might suddenly become a large fraction of all posts that year. 

We could go with something like "linkpost all especially high quality articles on nuclear risk that are especially relevant to the most extreme risk scenarios (not just e.g. the detonation of 1 or a few bombs by terrorists)". But that's a murkier principle, and it seems like it could easily end up "going too far" (or at least seeming weird). And I think worrying that I'm going too far might lead me to hold back more than is warranted.

Maybe this could be phrased as "Making the decision partly based on whether the content was created by an EA could help in establishing a Schelling fence that avoids a slippery slope. And the existence of that fence could help people be more comfortable with beginning to travel down the slope, knowing they won't slip too far."

Aaron Gertler @ 2021-01-19T06:00 (+10)

A few notes on "deciding how much to crosspost":

  • A single crosspost with a bit of context from the author -- e.g. a few sentences each of summary/highlights, commentary, and action items/takeaways -- seems better to me than three or four crossposts with no context at all. In my view, the best Forum content tends to give busy people a quick way to decide whether to read further.
  • "Written by someone connected to EA" is a decent filter, but quality/"special" relevance seem like better filters.
  • In some ways, non-EA academics could be better to crosspost -- they're less likely to post their own work, and they're more likely to be "discovered" by people who hadn't seen their work before because it was outside the community. (That said, the greater likelihood that an EA-involved person participates in discussion still makes that feature seem net-positive to me.)
  • If people are sharing too much interesting information on the Forum, and the site becomes cluttered, that's our team's responsibility to handle -- not a problem caused by the crossposter. 
    • We might eventually try to push for higher standards if crossposts overwhelm the Forum, but I think we're pretty far from that point right now.
    • Absent these "higher standards", we have other ways to mitigate a potential flood of crossposts; for example, we could add a way for people to filter out crossposts from their feed (using a "crosspost" tag is the simple version of this, but linkposts are distinct from regular posts in our code, so there are probably other ways it could be built).
MichaelA @ 2021-01-19T13:28 (+2)

That all makes sense to me - thanks!

G Gordon Worley III @ 2021-01-15T21:34 (+11)

On a related but different note, I wish there was a way to combine conversations on cross-posts between EA Forum and LW. I really like the way AI Alignment Forum works with LW and wish EA Forum worked the same way.

MichaelA @ 2021-01-16T01:14 (+7)

Yeah, I agree with that.

A related point is that at least once I've seen something that was posted to both sites without the author noting that. This means users won't even know to check the other site for more comments, let alone having them automatically visible from the first site they were on. 

Not sure how often this happens. Maybe it's not a big deal.

Also not sure if there's a good technical fix for this. My first thought is an opt-in checkbox that always appears on either site to ask "Do you want this to also appear as a cross-post on [other site]?" (which would fix it because then presumably no one would bother manually making a cross-post, and the automated cross-post would always say at the top that it's a cross-posted). But that might lead to too many cross-posts (I don't know). 

The best fix - if this is even a semi-regular problem in the first place - might just be to occasionally prominently mention the option of cross-posting and that one should label posts as cross-posts when one does so. Or just to comment on the relevant posts when one happens to notice that this has happened.

EdoArad @ 2021-01-15T08:12 (+8)

Daydreaming a little bit. 

Imagine that there was an EA Browser, that acts just like your favorite browser but also has an option of up/downvoting, tagging, and writing comments on any web page. 

Imagine all the people in the EA community are using that browser as they go through their day and casually upvote some webpages or write some comments.

(Imagine there's no spamming.. 🎶)

How would you design a forum feed based on those web annotations then? Probably have some default high bar (or quantity? or perhaps randomly??) on what goes to the main feed, and an option to view all web annotations.

This could be implemented rather easily by building a chrome add-on (or starting with using a private EA group for https://hypothes.is/ and feeding that to the forum).

I imagine that this would be surely useful, if people would use something like this, in that I don't see major drawbacks. That makes me think that this is a solvable design problem.

Ozzie Gooen @ 2021-01-15T17:53 (+4)

I've been thinking about this for a while. 

I've had decent experience with WorldBrain's Memex, while I haven't really enjoyed using hypothes.is as much. There are issues I have with Memex, but I'm more optimistic about it. They're adding collaboration functionality. I've talked to the CEO and they might be a good fit to work with; if it were the case that a few community members were bullish on it, I could see them listening to the community when deciding on features.

https://getmemex.com/

It's all a lot of work though. I'd love for there to be some sort of review site where EAs could review (or just upvote/downvote) everything. 

EdoArad @ 2021-01-15T19:26 (+2)

Cool, I'll check it out

Aaron Gertler @ 2021-01-15T10:07 (+4)

What is the function of a private group within hypothes.is? Does that mean you only see annotations from other members of the group(s) you are in?

Given how sparse the Forum's comments are, I'd be curious to see how many people would actually make annotations. Maybe some people don't like commenting but would enjoy annotating, because it feels more direct or more private or something. 

I suppose you'd want to hook this system to the Forum so that annotations showed up as comments when a post was crossposted here (otherwise, you'd have comments scattered between both places). That sounds like a lot of technical effort, but if it drives engagement, it could be worthwhile.

However, if the system isn't hooked up, you have the classic competing standards problem, where we've now added one more place for people to discuss content, and each lone person trying to start a discussion in their place is a bit less likely to succeed. But annotations might also increase the overall number of people who discuss content, with a net positive effect...

...anyway, if someone creates a concrete plan for this and can get buy-in from a group of people, I'd be interested to see where it goes.

EdoArad @ 2021-01-15T10:58 (+3)

Exactly, a private hypothes.is group is one where you only see annotations from members of the same group, and only annotations that were tagged as annotations for that group. 

Definitely agree that doing something like that should be hooked up to the forum, and that it is a bit of a technical challenge. 

I am not sure if engagement is the right metric to use here, though. Not sure if it isn't. I'm also not sure if that's an important point so I'll just keep this in the back of my head and maybe something will come up in the future. 

MichaelA @ 2021-01-15T08:04 (+6)

Semi-related: I think it might often be good for people who make linkposts and want to add some commentary to add it in the comments section, with one comment for each point/question (or cluster of points/questions) - like I did  here - rather than putting all the commentary in the body of the post. I think that the former approach could be better for encouraging discussion and keeping it organised. 

(I think I've seen other people recommend roughly this, for similar reasons, but I can't remember where.)

(Of course, there might also be other times when it would be more natural/useful to put all the commentary in the body of the post.)

EdoArad @ 2021-01-15T08:29 (+4)

I think it was Aaron that raised a related suggestion - to add points for discussion of a post in the comment section.

MichaelA @ 2021-01-15T08:57 (+2)

I hadn't seen that, but that's indeed relevant, and I think what I'd seen was probably Aaron positively noting that a post that won a Forum Prize had done roughly this.

MichaelA @ 2021-01-29T09:18 (+2)

Update: I've just posted Notes on Schelling's "Strategy of Conflict" (1960). This is an experiment in:

  1. linkposting EA-relevant content that wasn't created by EAs
  2. a relatively low-effort approach to linkposting with something like a summary/commentary/notes, by just sharing the Anki cards I made for myself when reading the book

In a comment on the post, I also solicited people's thoughts on whether posts like that seem useful.