37

Announcing PriorityWiki: A Cause Prioritization Wiki

This essay was jointly written by Peter Hurford and Marcus A. Davis.

Locating and keeping up to date on all of the information in all every cause worthy of attention is time-consuming work, and a task we think is likely to be duplicated many times over by individuals and organizations. It’s very useful when deciding what interventions to spend your time or money on to know what work has already been done in the area.

However, much research is done and then lost to the sands of time, buried on some webpage somewhere. It takes significant effort to remember where all the relevant research lives. It would be nice to synthesize all of this research into one place.

This is why we’ve created PriorityWiki, a cause prioritization wiki anyone can edit which categorizes particular interventions within broader causes. For each of these interventions and causes, we link out to high-level summaries on the topic that have been done within (and outside of) the broadly defined effective altruism community. From there, you hopefully have a good launching pad to all the relevant research in any topic.

We’ve seeded this wiki with intervention specific links to work from Open Philanthropy, GiveWell, 80,000 Hours, Animal Charity Evaluators, and others on a wide range of topics. For example, within the the high-level of cause of “Developed World Health”, we’ve further subdivided the page into the type of intervention and link out to specific reviews into anti-malarial interventions, vaccinations, smoking interventions.

 

 

While we have seeded the wiki with initial content, it is not currently comprehensive in any cause area and doesn’t include all plausible interventions. We would like for this wiki will serve as a resource to quickly locate and investigate what work has been done on all causes. We would hope that it will at least significantly reduce some of the time-consuming work of trying to gather what research has been done on intervention ideas.

The wiki is editable by anyone without logging in but also supports accounts, and signing in with Google. All of the changes are logged publicly on GitHub with the site itself hosted here and specific pages here.

If you’re interested in expanding this, you are free to add content. Just get started at the PriorityWiki.

Comments (26)

Comment author: RyanCarey 19 June 2018 12:20:36AM 6 points [-]

Do you have a plan for managing information hazards?

Comment author: Peter_Hurford  (EA Profile) 19 June 2018 12:41:12AM 3 points [-]

We control the site, so we can revert the addition of any information hazards if they come up. I imagine the site has the same risk of spreading infohazards as, say, this forum.

Comment author: RyanCarey 26 June 2018 08:11:43AM *  2 points [-]

Do you have a plan for scanning over posted materials (analogously to moderation on the EA Forum), a code of conduct for posts, or a procedure for discreetly flagging hazardous content?

Comment author: zdgroff 21 June 2018 02:15:55AM 2 points [-]

This is awesome. I suppose this is something anyone could fix, but I'm curious why it seems to deviate from the normal EA division of causes and has animal welfare as a subcause? Animal welfare already has a number of categories under it, and not all policy-related, so seems like maybe it should be its own category.

Comment author: Peter_Hurford  (EA Profile) 21 June 2018 05:05:59AM 0 points [-]

Categorizing causes is very hard. I don't think this is this is the best attempt, and I agree that the section on nonhuman animal welfare in particular may span multiple categories. I'm still deciding how to best categorize things and may write more about this in the future. I'm definitely open to suggestions or edits.

Comment author: Chosh 22 June 2018 10:11:56PM 0 points [-]

The animal welfare section is currently fairly sparse. If it gets fleshed out (no pun intended) significantly more, I could see it belonging as a top level cause area. It already has content about equal to the Specific Research and Improving Research, Decisions and Values sections, though of course those can grow too!

Right now Improving Research, Decisions and Values seems like it could get broken up into different sections, between Policy and X-risk mainly, but Improving Policy is such a heterogeneous group that I'm a bit wary of adding more to it...

Comment author: Telofy  (EA Profile) 19 June 2018 03:33:59PM *  13 points [-]

Sweet! I hope it’ll become a great resource! Are you planning to merge it with https://causeprioritization.org/? If there are too many wikis, we’d just run into the same problem with fragmented bits of information again.

Comment author: Peter_Hurford  (EA Profile) 19 June 2018 07:53:39PM 4 points [-]

Good question. The answer is that I'm not sure yet how to handle that.

I agree I don't want there to be five different prioritization wikis, but I also don't want the fact that one person has done the project at one point to mean that no one can ever do the project again. It's hard to strike the right balance there.

We mainly chose to go with a different wiki software to improve the editing experience (especially editing without creating an account) to remove barriers to contribution. We could potentially start migrating content whichever way makes sense.

Comment author: TimotheeChauvin 30 July 2018 05:15:40PM *  2 points [-]

I am strongly in favor of migrating content into a unique wiki.

As for the editing experience, I prefer the Cause Prioritization Wiki for several reasons, partly because it looks closer to Wikipedia and because it is possible to view the edit history in a convenient way. In short, I feel more in control there and I believe it is easier to spot and revert vandalism.

In the case of an EA wiki, I would argue that the barriers to contribution are not the creation of an account or the use of the wiki syntax. Instead, they are the trouble becoming aware that the wiki exists, and the lack of trust in the stability and notoriousness of the wiki.

Therefore, and as pointed out in other comments, I would be glad if:

  • the priority wiki and the cause prioritization wiki were merged into a single wiki that is as close as possible to Wikipedia in its style;
  • this wiki was linked to from the EA Forum, the EA Facebook group, effectivealtruism.org, etc, so it looks like the official EA wiki;
  • a critical mass of content was attained quickly so that readers trust its stability and notoriousness and consider editing (I am willing to volunteer for this phase);
  • (the EA wiki on wiki.eahub.org was deleted, because it is indeed filled with bot spam and, although most of it is certainly quite entertaining, it makes it look like EA is dead).

I'm convinced that with this early work we can have a functional, regularly edited and notorious wiki, and that would be amazing!

Comment author: riceissa  (EA Profile) 20 June 2018 06:22:12AM *  5 points [-]

(Context: I host the Cause Prioritization Wiki.)

I think there might be a misunderstanding here, so I would like to clarify a couple of things.

I also don't want the fact that one person has done the project at one point to mean that no one can ever do the project again.

I assume this is referring to me, and that Peter is saying the Cause Prioritization Wiki is dead. It's true that the wiki was inactive for about two years, but more recently I've been adding more content to it; there is an edit history graph showing activity for the past year.

But even assuming the wiki is dead, I'm not sure starting essentially from scratch is better than reviving the existing project.

We mainly chose to go with a different wiki software to improve the editing experience (especially editing without creating an account) to remove barriers to contribution.

The choices of wiki software and of allowing anonymous edits are not unchangeable. For the former, I've actually been pondering for a while whether switching to MediaWiki would be a good idea (I haven't looked into Wiki.js, which is what PriorityWiki uses), as I've gotten more experience with editing on MediaWiki wikis since the time when I started the Cause Prioritization Wiki. For the latter, my thinking has been that I don't want to spend a lot of time moderating the wiki, which is why I chose to restrict account creation and disable anonymous edits. But if there is enough energy to moderate the wiki, I would be fine with allowing more open editing.

Some thoughts I had about competition while thinking about this situation (I haven't spent a lot of time thinking about this topic):

  • In general I think competition benefits end users.
  • There are four existing wikis about bitcoin that I know of, which might be an interesting case study:
  • With free software, forking is often difficult (existing codebase too complicated to understand, written for a different OS, written in a language that one is unfamiliar with) so there's a proliferation of similar applications. This seems to be less of a problem for prose.
  • Again with free software, different software projects focus on different (sometimes incompatible) things, like speed, feature-richness, memory use, portability. With a wiki, there is still some of that (one can trade off along formal vs informal language, background knowledge assumed, audience's goals) but I think it's less strong.
  • Again for software, there is also the issue of getting stuck in local optima (think how horrible LaTeX is but people are forced to use it). I think Wikipedia is similarly a local optimum for a generic encyclopedia, but this seems mostly problematic because of its deletionism.
  • For products that are sold there is also competition along price.
  • For textbooks, I think it's good that there are a bunch of them for each (topic, level) combination, because exposition style/difficulty can vary significantly. I think for general reference works there is a lot less of that, and even less for inclusionist electronic wikis.
Comment author: casebash 21 June 2018 12:03:30PM 7 points [-]

I suspect that competition isn't especially healthy at the stage where the struggle is just to obtain critical mass. Multiple projects have failed to generate enough momentum to sustain them, if the community divides its attention between multiple such projects, success is less likely.

Comment author: Telofy  (EA Profile) 20 June 2018 11:02:22AM 2 points [-]

I’m against it. ;-)

Just kidding. I think monopolies and competition are bundles of advantages and disadvantages that we can also combine differently. Competition comes with duplication of effort, sometimes sabotaging the other rather than improving oneself, and some other problems. A monopoly would come with the local optima problem you mentioned. But we can also acknowledge (as we do in many other fields) that we don’t know how to run the best wiki, and have different projects that try out different plausible strategies while not being self-interested by being interested in the value of information from the experiment. So they can work together, automatically synchronize any content that can be synchronized, etc. We’ll first need meaningful differences between the projects that it’ll be worthwhile to test out, e.g., restrictive access vs. open access.

Comment author: John_Maxwell_IV 20 June 2018 07:59:10AM *  1 point [-]

I don't think Bitcoin became popular on the strength of its wikis. I also do think that free software projects cannibalizing one another can be harmful, e.g. there was a period where Python had a bunch of serious web frameworks whereas Ruby just had one or two, and I think that was good for the Ruby side because the ecosystem that built up around those one or two was deeper.

Gwern wrote this essay about why non-Wikipedia wikis have a hard time competing with Wikipedia. He recommends using Wikipedia when possible and only falling back on specialized wikis for things Wikipedia won't allow. So that might be a path forwards.

Comment author: gwern 22 June 2018 02:33:10AM *  6 points [-]

Bitcoin definitely didn't become popular because of its wiki. Early on I wanted to contribute to the wiki (I think as part of my DNM work) and I went to register and... you had to pay bitcoins to register. -_- I never did register or edit it, IIRC. And certainly people didn't use it too much aside from early on use of the FAQ.

An EA wiki would be sensible. In this case, while EAers probably spend too little time adding standard factual material to Wikipedia, material like 'cause prioritization' would be poor fits for Wikipedia articles because they necessarily involve lots of Original Research, a specific EA POV, coverage of non-Notable topics and interventions (because if they were already Notable, then they might not be a good use of resources for EA!), etc.

My preference for special-purpose wikis is to try to adopt a two-tier structure where all the factual standard material gets put into Wikipedia, benefiting from the fully-built-out set of encyclopedia articles & editing community & tools & traffic, and then the more controversial, idiosyncratic stuff building on that foundation appears on a special-purpose wiki. But I admit I have no proof that this strategy works in general or would be suitable for a cause-prioritization wiki. (At least one problem is that people won't read the relevant WP article while reading the individual special-purpose wiki, because of the context switch.)

Comment author: John_Maxwell_IV 13 July 2018 09:46:32AM 0 points [-]

If you're interested, I wrote some thoughts here about why projects of this type seem to have so much trouble getting off the ground.

Comment author: John_Maxwell_IV 20 June 2018 07:51:42AM *  11 points [-]

It looks like there is another abandoned wiki here, which was moved to here (but the latter link appears to be overrun with spam). Less than 2 years ago, angelinahli started what looks like a really solid attempt to do something kinda similar to what Marcus & Peter appear to be doing here.

I wish there was some way to solve the meta-level problem of projects like this getting forgotten. If you zoom out a bit, this category appears quite crowded. It seems that attempts to prevent things from getting lost in the sands of time may also get lost in the sands of time!

Maybe it's not an issue of fragmentation so much as lack of discoverability. Perhaps what's needed is to persuade people running the EA forum/EA Facebook Group to link to the directory resource from the forum intro/FB group description. If it's known that a wiki has readers, maybe it will be easier to attract writers. Article view counts might be helpful? (Chicken-and-egg problem: For readers, you need writers, but for writers, you need readers.)

Zooming out, I suspect part of the issue is it feels scummy to promote your own work, so good work often ends up gathering dust--even if the purpose of that work was to help prevent other work from gathering dust! I think people should get over their aversion to promoting their own work some, and also try to promote the work of others more often. Then maybe we can have a wiki which reaches "meme status".

Comment author: vollmer  (EA Profile) 04 July 2018 01:36:26PM 2 points [-]

+1 for merging/coordinating these various platforms more.

The EA concepts platform is also somewhat similar. It could be important to think about how these projects relate to one another.

Comment author: Telofy  (EA Profile) 20 June 2018 10:52:50AM 0 points [-]

That would be an immensely valuable meta problem to solve!

Then maybe we can have a wiki which reaches "meme status".

On a potentially less serious note, I wonder if one could make sure that a wiki remains popular by adding a closed section to it that documents particular achievements from OMFCT the way Know Your Meme does. xD

Comment author: Khorton 19 June 2018 10:31:24AM 2 points [-]

It seems odd to me that there's only one page for developing world health. I'd expect that page to get long and confusing quickly!

Comment author: Peter_Hurford  (EA Profile) 19 June 2018 01:28:38PM 2 points [-]

We can split it up into multiple pages if it gets unwieldy. Figuring out the best organization is hard.

Comment author: JasperGeh 19 June 2018 09:19:42AM 2 points [-]

This is great! To have a collection of links of interventions, arguments, resources etc for each cause is super valuable for people who are not that well versed in the EA-website-bubble.

Looking forward to people adding their resources; S-risks for example are still missing and a quite important consideration. But I imagine someone from the FRI is already writing.

Comment author: RomeoStevens 19 June 2018 01:55:40AM 5 points [-]

In addition to cataloging sources of data and analysis for current and potential EA causes, it might also be nice if there was a repository of info on why some common cause areas are not generally recommended by EA. I'm unsure how one would incentivize such info being added though.

Comment author: Peter_Hurford  (EA Profile) 19 June 2018 02:46:10AM 3 points [-]

it might also be nice if there was a repository of info on why some common cause areas are not generally recommended by EA

Good idea. I had been experimenting by adding summaries at the top of some articles (for example this one on aging) and was trying to figure out how opinionated the Wiki should be. Right now I was trying to err on the side of being less opinionated. If you have any thoughts on this issue, I'd definitely be curious to hear them.

I'm unsure how one would incentivize such info being added though.

We're hoping to eventually and slowly create a volunteer pool to do this kind of work. This seems like the kind of tasks volunteers have done well on in my past experience. Furthermore, given funding, we'd even be able to pay for the assistance.

Comment author: Yannick_Muehlhaeuser 20 June 2018 05:37:59AM 1 point [-]

was trying to figure out how opinionated the Wiki should be

Certainly an important question. 80k certainly explains why they don't recommend certain careers and it's important for them to continue to do so. In my opinion we should make our reasons for considering a cause effective very clear, so they can be challenged. In practice, of course, how such an entry depends strongly on the wording. I would prefer to word it like "Cause X has traditionally been considered not neglected enough/not tractabe/too small by EA organisations. ... According to that reasoning you'd have to show Y to establish X as an effective cause. ..." instead of "X is not effective, because ...".

Comment author: RomeoStevens 19 June 2018 04:02:08PM *  2 points [-]

Maybe also a prize for best new wiki entry periodically?

How opinionated it is probably comes down to tone more than content. Less 'and this is why everyone who supports education is stupid' and more 'this is the story on education studies so far. We hope this can be of assistance to someone trying to develop new educational interventions so they don't go down the same blind alleys as previously' that could help. It could also harm in the sense that controversy engenders engagement and a more confrontational approach would get people to actually argue.

Also I'd like to note that I'm bullish on this idea overall as I think it might allow for genuine philosophical progress. Part of the lack of progress comes from the fragmentary nature of all the various arguments, making people very hesitant to offer critiques since a likely outcome is 'that has already been addressed in 3 places.' We tend towards a community of correctors, which shuts down generative creative thought.

Comment author: Dunja 18 June 2018 10:51:36PM 2 points [-]

This is great, thanks a lot for doing this work!