I recently posted on the EA Facebook group asking if there existed any (1) comprehensive, (2) crowd-sourced / semi-crowd-sourced and (3) well updated databases cataloguing blog posts, articles and general media sources related to different EA issues.
Seeing as no one who saw the post suggested anything that fits the above description, I thought I might take a stab at the idea. This post is potentially the first in a series to explore and chronicle the emergence of this idea.
Why might this be worthwhile?
But first, in the interests of not creating a database for the sake of creating one, and also because it might be good to hold off on proposing solutions, I wanted to brainstorm in what ways such a thing might be useful.
In particular my inner skeptic wants to know how much better this would be to what we have now (e.g. just googling "effective altruism + keyword"). It's not obvious to me prima facie that an imposed database would be significantly better than the informal systems we have now, and I loath to create a useless shell of a project that dies a death of disuse.
Here are the reasons I ended up with:
A. Enhancing knowledge
1. EA 201 intermediate study - There exist several excellent introductions to EA, including the EA handbook and Ben Kuhn's great reading list, but after a brief primer on EA topics there doesn't seem to be a second step for people to take to expand their understanding of EA issues. There are some structures that try to address this (e.g. the EA newsletter?), but none come to mind that allow for easy, focused and intentional study of different subject areas. I'm worried that people who do not have the time or patience to peruse the EA forum / blogosphere themselves will become disengaged with this idea, and stop engaging with it. Helping people bridge the gap between "newbie" and "fanatic" seems like a useful thing.
2. Fill in knowledge gaps within the community - Similar to the first point, without some systematic way to approach studying different EA topics, I think individual EAs will start falling prey to known unknowns and (worse) unknown unknowns. This seems really bad in terms of helping people optimize for donations. If I do not understand a cause area well enough or even known it exists, it's likely to be a blind spot in my giving patterns. For instance, before stumbling across Brian Tomasik's blog it never occurred to be to even think about wild animal suffering as a thing that existed, and I am sure I have other blind spots I haven't interrogated yet. Having a way to fill in community knowledge gaps that doesn't purely rely on serendipity seems useful too.
B. Coordinating conversations
3. Reduce repeat discussions - There are merits to having the same kinds of discussions again several times (perhaps they allow the people having those discussions to reach a deeper understanding of the issues being discussed), but those benefits don't seem to accrue to observers. In particular, it would be nice to point interested people to conversations that have occurred already - I'm reminded of the practice within some feminist blogs to point newbies to FAQs that answer basic questions about feminist issues so that the conversation can proceed to new ground. Hopefully this won't shut down conversations but will make the quality of discourse better.
4. Identify gaps in EA literature - This might be more difficult to achieve, but ideally it'd be really nice to have a way to know when we are missing solid articles on a particular subject. This would require a sufficient comprehensiveness in a database such that the absence of literature in that database was a reliable indicator of a general absence in literature and not poor record-keeping within the database itself. If we could have a database that is pretty well crowd-sourced, maybe we could capture some of these benefits.
5. Make EA debates clearer to follow - Two possible advantages: (1) Allows for more systematic understanding of EA debates, and (2) Puts clashing voices in more direct conversation with each other. (1) After falling down the clickhole of EA blog posts I sometimes find myself just agreeing with whichever side of the debate I read about last. I'd like to stop doing that, and start understanding better why the debate exists, etc. A good database can make informal literature reviews easier. (2) It seems plausible that different voices on the same issue haven't been introduced to each other, and this database could be a way to do that. But I can't think of any examples of where this exists, so maybe this is not a real benefit I can co-opt.
C. Resource for other EA activities
6. Better offline discussions in EA groups - I was talking to some folks at a Boston EA meetup, who suggested that it'd be nice to have more specific in person discussions about EA topics (e.g. population ethics), and that a good way to facilitate this might be to hold "blog reading groups", where people come in and discuss topics after having read a list of relevant resources. A solid EA database could help make these kinds of conversations happen, and I'm sure other groups could find other uses for such an idea.
7. Fodder for targeted links EAs can share with friends - I occasionally find myself talking to a friend and realizing they should really be introduced with one part of the EA blogosphere. It'd be nice if after having those thoughts I could go to one place and find a thing for them to look at. Maybe this is a problem other people face that could be solved.
Some concluding thoughts (before phase 2)
Based on the above reasons I am inclined to think something along the lines of my initial suggestion might potentially be able to offer enough benefits to the community to be worth pursuing. But I'm also aware that now that I've sunk some time into brainstorming reasons in support of such an idea I'm more inclined to agree with myself.
So before sinking more time into this, I'm interested to know whether there are things that my mind hasn't thought of yet. In particular, I'd be very interested to know, in this order: (a) if any of my proposed benefits above seem off, (b) if there already exist ways that create the above benefits, (c) if there are better ways to create the above benefits, and (d) if you can think of any ways other communities have solved for the above benefits / anything else useful for the creation of such a thing.
I know I said I'd hold off on proposing solutions, but some ideas on things that look like this database idea proposed by others on Facebook include: something like an open-source, wiki-style, public annotated bibliography (Jay Quigley), reading lists / databases compiled by EAs who have relevant experience - such as current and former employees of EA orgs (Jacob Funnell), something like the pareto bookshelf - a list of recommended reads by those who applied for the pareto fellowship (James Aung), something that looks like Aaron Gertler's list of articles but more organized/comprehensive/selectively compiled. Other ideas are to replicate something that looks like LW's sequences, maybe delegate to people working in different EA fields the task of compiling reading lists on those specific issues and then consolidate those reading lists to a specific page, etc.
If people tend to think this is a thing worth having, I'm also aware I might not be the best person to try and create such a thing, since the EA community seems to have an ample supply of computer science-y people and I am not one of those people (my comparative advantages right now lie in the domain of having a fair bit of time on my hands and being willing to do the grunt work of reading/compiling good articles). If someone likes this idea and would like to co-opt it, I would be very happy to pass the torch on. But if (a) no one more suitable than me wants to do it and (b) it still seems like this would be a nice thing to have, I'm willing to have a go at it.
Thoughts are very welcome!