EA Forum dev here. At a high level, I think of the EA Forum* as being a group blog where anyone can post. I would group two themes of the problems as
1) ārelevant content discovery is hardā and
2) āposting can be intimidatingā.
My model is that you think 3) ādiscussion happens on facebookā is downstream of 1 and particularly 2.
I have progress to report on 1, so I can start with that. Finding good, old content is hard on any Forum, and Iām pretty happy with the way the Posts by Timeframe feature (previously mentioned) worked out. You can see it live on the LessWrong version of the page; weāll be deploying it soon. Now you can scroll back through the weeks /ā months /ā years and find popular posts that you missed. Iām also most of the way through adapting the sidebar (seen on the left side of LW here) to make it easier to discover that the All Posts page exists.
Itās possible that sub-forums is the best way to solve this. However, Iām worried about a couple of things: a) that sub-forums would cause the specializations to ossify and remove valuable cross-pollination of ideas, and more mundanely, b) that I really want to be careful with the hard-to-reverse and disruptive change of moving to a sub-forum format. Both point in favor of tagging, which Iām somewhat more optimistic about.
Problem 2 Iām more uncertain about. I want people to have a lower bar for asking and answering questions. The new shortform feature from LessWrong will also help somewhat. But making the Forum easier to post to without lowering the quality level is a hard problem. See above for why Iām less optimistic about too hasty a switch to sub-forums to solve this.
Thanks for being an attentive observer and providing feedback.
* This also applies to LessWrong, whose codebase we forked.
On the topic of āwriting posts is intimidatingā: This can definitely be true! I still remember the first time I submitted a full post to LessWrong, checking back every few hours for comments and wondering whether certain Famous (To Me) Internet People would see it.
On the content side, Iām trying to make writing posts easier by offering review of post drafts (and ideas that havenāt become drafts yet). Iāve only heard from a couple of people so far, and have lots of capacity to provide feedback!
sub-forums would cause the specializations to ossify and remove valuable cross-pollination of ideas
I really donāt think this is a problem, I think there is already so much cross-pollination between people in different unrelated cause-areas within EA at EA Globals, other EA events, friendship groups created by these events, facebook groups, etc. And Iām not sure itās useful, thereās too little overlap between topics like AI safety and animal welfare.
I somewhat agree. When I say āIām worried aboutā, I donāt mean āIām confident but using softening languageā ā Iām actually pretty uncertain. The meta point is that Iām worried about it and predict it would be hard to reverse.
On the object level, Iām less worried about AI safety and animal welfare so much as on the boundaries of related cause areas. For example:
1) Hardening currently fuzzy boundaries between different specialties of long-termism
2) Reducing the flow of context from object level work into the meta-EA space
3) Specialty knowledge sharing between cause areas, like outreach knowledge between farm animal welfare and global poverty
These seem like problems that one could at least largely address, but (back to the meta point) Iād expect doing so well would require at least a monthās worth of work.
RE: 2) Posting on the forum can be intimidating is because a lot of EAs working at EA orgs read and interact on the forum fairly regularly (which is good!) and there may be the fear that by posting you will waste their time or (if you post under your real name) it may affect your reputation in the community.
In facebook groups on the other hand, there tends to be more interactions with the average community member, and the groups are semi-private, so only a small subset of the community will see what you post. (These are my impressions of the two platforms, Iām not sure if others feel that way) Iām optimistic about the shortform feature improving things, but iām not sure by how much.
EA Forum dev here. At a high level, I think of the EA Forum* as being a group blog where anyone can post. I would group two themes of the problems as
1) ārelevant content discovery is hardā and
2) āposting can be intimidatingā.
My model is that you think 3) ādiscussion happens on facebookā is downstream of 1 and particularly 2.
I have progress to report on 1, so I can start with that. Finding good, old content is hard on any Forum, and Iām pretty happy with the way the Posts by Timeframe feature (previously mentioned) worked out. You can see it live on the LessWrong version of the page; weāll be deploying it soon. Now you can scroll back through the weeks /ā months /ā years and find popular posts that you missed. Iām also most of the way through adapting the sidebar (seen on the left side of LW here) to make it easier to discover that the All Posts page exists.
Itās possible that sub-forums is the best way to solve this. However, Iām worried about a couple of things: a) that sub-forums would cause the specializations to ossify and remove valuable cross-pollination of ideas, and more mundanely, b) that I really want to be careful with the hard-to-reverse and disruptive change of moving to a sub-forum format. Both point in favor of tagging, which Iām somewhat more optimistic about.
Problem 2 Iām more uncertain about. I want people to have a lower bar for asking and answering questions. The new shortform feature from LessWrong will also help somewhat. But making the Forum easier to post to without lowering the quality level is a hard problem. See above for why Iām less optimistic about too hasty a switch to sub-forums to solve this.
Thanks for being an attentive observer and providing feedback.
* This also applies to LessWrong, whose codebase we forked.
On the topic of āwriting posts is intimidatingā: This can definitely be true! I still remember the first time I submitted a full post to LessWrong, checking back every few hours for comments and wondering whether certain Famous (To Me) Internet People would see it.
On the content side, Iām trying to make writing posts easier by offering review of post drafts (and ideas that havenāt become drafts yet). Iāve only heard from a couple of people so far, and have lots of capacity to provide feedback!
I really donāt think this is a problem, I think there is already so much cross-pollination between people in different unrelated cause-areas within EA at EA Globals, other EA events, friendship groups created by these events, facebook groups, etc. And Iām not sure itās useful, thereās too little overlap between topics like AI safety and animal welfare.
I somewhat agree. When I say āIām worried aboutā, I donāt mean āIām confident but using softening languageā ā Iām actually pretty uncertain. The meta point is that Iām worried about it and predict it would be hard to reverse.
On the object level, Iām less worried about AI safety and animal welfare so much as on the boundaries of related cause areas. For example:
1) Hardening currently fuzzy boundaries between different specialties of long-termism
2) Reducing the flow of context from object level work into the meta-EA space
3) Specialty knowledge sharing between cause areas, like outreach knowledge between farm animal welfare and global poverty
These seem like problems that one could at least largely address, but (back to the meta point) Iād expect doing so well would require at least a monthās worth of work.
I think a tagging system would be really great.
RE: 2) Posting on the forum can be intimidating is because a lot of EAs working at EA orgs read and interact on the forum fairly regularly (which is good!) and there may be the fear that by posting you will waste their time or (if you post under your real name) it may affect your reputation in the community.
In facebook groups on the other hand, there tends to be more interactions with the average community member, and the groups are semi-private, so only a small subset of the community will see what you post. (These are my impressions of the two platforms, Iām not sure if others feel that way) Iām optimistic about the shortform feature improving things, but iām not sure by how much.