# 74

tl;dr: The EA Wiki seems probably worth funding, but it is not the most ambitious project that the main person behind it could be doing.

Epistemic status: Although we have data and forecasts for some EA Wiki metrics, the most important variables—such as future growth or impact—require speculation. Readers should care or listen to that speculation insofar as they think that I have good judgment. But although we have significant uncertainties, we can still meaningfully evaluate the EA Wiki. In the future, we might be keen to apply this level of evaluation to other interventions.

This evaluation is a project by the Quantified Uncertainty Research Institute, commissioned by Pablo Stafforini and Max Daniel, financed by the EA Infrastructure Fund, and conducted by Nuño Sempere. Pablo and Max reviewed this document prior to publication.

## Summary

The EA Wiki is a project which attempts to produce and organize EA content. It is hosted within the EA Forum on top of its tag system, and it is the latest attempt on a long line of EA Wiki projects which previously failed.

On the positive side, metrics like "words per dollar" and "views per dollar" are high. Also, Pablo Stafforini—the EA Wiki’s current lead—is a good fit for the project and is very prolific.

On the negative side, Pablo alone spends comparable amounts working on the wiki as users cumulatively spend reading it. The EA Wiki—and in particular the editing festival—doesn't seem to have resulted in junior EAs doing many edits to the Wiki, suggesting that its movement-building-related benefits are on the low side. Further, the wiki doesn’t seem that useful right now in terms of helping current research efforts, such as those of EA research organizations or independent researchers.

On the neutral side, it seems like it will take more time—perhaps on the order of a few years—to determine whether the EA Wiki is successful or not. The case for the EA Wiki right now relies on its increased future usefulness. The wiki is early, so future software functionality and content might make existing content more valuable.

The fact that we need more time to know the influence of the wiki shouldn’t be particularly surprising. Given that similar projects previously failed, it would be surprising if this iteration immediately succeeded.

Some projects could be built on top of the EA Wiki, like more detailed and granular hierarchies of concepts, an accessible curated repository of EA forum posts with links and collected references, or more in-depth articles rather than stubs. Some more examples of projects are given in their own section below. I would feel more excited about the EA Wiki exploring some of these adjacent projects, rather than just continuing to host very short articles (stubs) as it does now.

Overall, I conclude that the EA Wiki seems worth funding for CEA. For Pablo himself, I think he might be able to find more ambitious work to do sometime in the future, but I recognize that this could be difficult.

## Introduction

### Background

The purpose of this evaluation was.

1. Provide evidence and opinion to inform
1. whether Pablo Stafforini should decide to continue working on this project
2. whether the EA Infrastructure Fund should renew Pablo Stafforini’s grant
2. If Pablo does continue to work on the EA wiki, make suggestions on how this could create more value.

As a frequent participant in the EA Forum, I (Nuño) began this evaluation with a skeptical to a negative impression of the EA Wiki. I found the tagging system useful, but I had otherwise not used the Wiki much myself.

### Stated goals of the EA Wiki

Per the two grant rationales evaluating the EA Wiki’s expected impact in April 2020 and June 2021, the two main hopes for it were:

1. That it might help onboard new EAs.

2. That it might help organize existing knowledge

## Evaluating inputs/outputs

### Does the path to impact through better organizing research check out?

The traditional way to organize research is through periodic literature reviews or reference works that occasionally summarize the state of the research. In contrast, a Wiki continuously summarizes the state of work it refers to. Pablo opines that:

Subjectively, it doesn't feel like EA is progressing at a faster pace than I can keep up with. But this impression has limited evidential weight because only a small fraction of entries are substantive enough to require updating. It could be that once most articles cease to be stubs, I won't have enough time to keep them all updated.

Because so many people read Wikipedia, and because journal articles are inaccessible both in terms of paywalls and writing style, Wikipedia does seem like it warrants much of the effort put into it.

However, for a smaller wiki, this is less clear. One could imagine that more effort could go into it than the effort it saves, as illustrated by the following xkcd comic:

In particular, because the EA Forum Wiki doesn’t directly target its summaries and literature reviews to specific EA Forum users, one could imagine that the EA Wiki puts more effort into creating wiki articles than the effort or time it helps save. For instance, one could imagine that the EA Wiki produces good literature reviews on the state of a specific topic, but that no EA researchers need a literature review on that particular topic.

Does the data above check that out? I’d say that this is unclear. Pablo put between 20 and 35 hours of work per week, and the EA Wiki received an average of ~34 hours of engagement per week. But this by itself doesn’t reveal how much researchers’ time the EA Wiki saved. In principle, this could have been more time, in the same way that Google saves people a lot of time even if they spend very little time searching.

Conversely, it could have been less, because not all users might have been looking for something specific rather than browsing the EA Wiki casually or unfocusedly. Similarly, maybe most users were more inexperienced researchers, and trading off Pablo’s time for their time at a close to 1:1 ratio might not have been worth it.

Overall, I think that the path to impact through better organizing research probably does not check out, or does not check out yet. See the corresponding section below on how to consider future impact.

### Does the path to impact through movement-building check out?

As mentioned in the original grant rationale, one hope for the EA Wiki was that it would help engage less-experienced EAs in worthwhile projects.

Readers can find a broader analysis of the editing festival here. The distribution of karma and number of edits of EA forum users who participated in the EA Wiki looks as follows:

Excluding Pablo Stafforini and Michael Aird, the distribution looks as follows:

Readers can find the entire distribution here. I created it by running this script over the users mentioned in the editing festival post. Ideally, I’d analyze the data for all EA Forum users, but I don’t expect it to change the picture significantly.

Overall, I’d say that most participants were fairly experienced (as determined by their karma score), and the editing festival mostly failed at eliciting long-term wiki (rather than tagging) activity. By long-term wiki activity, I mean editing and writing wiki articles after the festival ended, rather than tagging EA Forum articles.

It also seems that, for less experienced participants, writing comments on regular forum posts, or creating their own posts, is fungible with writing wiki articles. If the EA Wiki was not available, users could substitute their wiki editing activities with writing comments and posts, which would probably have similar benefits.

Overall, I’d say that the two pathways to impact proposed in the initial grant rationales mainly haven’t materialized yet.

## Considerations after seeing the data

### The EA Wiki probably needs more time until it can be determined whether it is a success or a failure

Thanks to James Aung for a similar point on community building projects which inspired this section.

When I think about what paths to success the EA Wiki could go through, only a very small number turn out to contain overwhelming successes after its first year. So reaching a point where the EA Wiki is valuable requires enough grit to keep at it for several years.

My impression is that given that the previous wiki projects failed, there probably wasn’t/isn’t a burning need for an EA Wiki, and the Wiki’s case depends instead on cumulatively building value and increasing usage over several years. The EA Wiki’s progress in the last year seems on track for that.

I think that there are two ways one could react to this consideration:

1. A neutral update: Given that the evaluation six and twelve months ago determined that the grant was worth it, I don’t think that all that much information has surfaced since, because the EA Wiki would have taken longer to succeed in most worlds. So if the wiki was promising then, it is still promising now.
2. A negative update: The relative lack of participation and the smallish ratios between Pablo’s time and engagement suggest there isn’t a burning need for an EA Wiki. Sure, if we are still confident that EAs will eventually use the EA Wiki a lot, we might continue to invest in it, but initial results might make us think that this is fairly unlikely.

In particular, considerations around movement building like

lends itself more to negative updates, where previous reasons for excitement were at least somewhat falsified.

On the other hand, considerations about investing in plumbing and infrastructure for future research seem like they lend themselves to more neutral updates. In particular, it would be weird to update negatively much after trying out something one knows is difficult and might take a while to see returns, and then starting out and confirming that the thing is indeed difficult.

### The case for the EA Wiki relies on increased usefulness in future years.

The value that the EA Wiki has produced in the past doesn’t seem large enough yet to justify its monetary cost. The case for its existence depends almost solely on its potential future value.

Note that the EA Forum has been growing at 100% per year, and it’s unclear whether it can keep up that growth. I mention this because the EA Forum hosts the EA Wiki, and increased usage of the forum might produce increased use of the wiki.

Note that the wiki software right now seems to be fairly basic—and Pablo seems like he has mainly focused on content, not software. If the software improves, that could also increase readership further. However, Pablo adds:

A mild reason for pessimism is that the software developments that have occurred so far do not seem to have increased engagement. For example, I don't think the ability to vote on wiki edits has made a difference (though it may be worth verifying if this is in fact the case). FWIW, my prior expectation was that it would make a difference, so I was somewhat surprised by this.

Metaculus predictions

Earlier this year, Pablo posted a few Wiki-related questions on Metaculus, which can be found here. They've so far gotten a few answers. Some of the more relevant predictions are:

1. EA Wiki will probably continue to exist (70%, estimate made by ten forecasters)
2. receive between 3 and 30% of the EA Forum’s traffic (90% confidence interval, 16 forecasters)
3. The EA Wiki will primarily not be written by volunteers (16%, 6 forecasters)

Overall, as evidenced by the very large confidence intervals on the percentage of EA Forum traffic, I don’t think that the Metaculus crowd is very confident.

Some types of activities seem like they could be carried out naturally as part of an EA wiki. Some that Pablo previously proposed are:

• Commissioning articles from experts
• Repurposing existing EA content from other scattered sources
• Publishing the Wiki as a book or self-contained online guide
• Integrating the Wiki with spaced-repetition
• Extracting a glossary of key EA terms

Some other directions suggested by EA Wiki survey respondents (quoted or paraphrased):

• “forum authors should be able to include links to relevant Forum Wiki articles for terms that the reader may not be familiar with and the reader, by hovering with the mouse pointer over the linked term, would see a 1-2 sentence glossary definition of that term (i.e., without having to click on the link)”
• Concept mapping
• More detailed and granular hierarchies of concepts
• Better ways of organizing editors and directing their attention, like how Wikipedia has various "projects" and has at the top of Talk pages the box showing how high priority and how developed an article is
• More in-depth articles rather than stubs
• Original research
• SEO work on the wiki. “I think it would be really cool if people searching for "diminishing returns" or "consequentialism" or "gene drives" or whatever get sent to the Forum, which might draw them into EA ideas.”
• Coordinated with the other existing wikis, like causeprioritization.org, to merge old content

Some other potential ideas for projects to build on top of the EA Wiki might be:

• suggesting bibliographies to EA forum posts,
• trying to criticize EA from the inside,
• trying to make EA Forum content less ephemeral,
• using the EA Forum to send out surveys,
• an accessible curated repository of EA forum posts with links and collected references,
• etc.

Right now, content on the EA Forum feels ephemeral, meaning that articles in the EA Forum mostly just get engagement when they are published. Yearly reviews or similar activities, or a rotating bar of high-quality past articles, like on LessWrong, could also fit within the EA Wiki’s purview.

If asked to brainstorm, EA community members might also have some ideas about how to increase or x10 the impact and reach of the wiki.

I would feel more excited about Pablo trying out some of those projects alongside or in combination with working on the Wiki, rather than working on the EA Wiki full-time.

### More details could be gathered about the EA Wiki’s usefulness

I was going to suggest using a few surveys to gather information about researchers' and community builders’ levels of excitement about the EA Wiki, but Pablo has sent some surveys already. If their answers are very positive or very negative, I imagine that this could more directly inform whether to continue with the EA Wiki or not. Analyzing the survey results in depth seems outside the scope of this evaluation, but so far, they seem ok.

One possible survey question might be about willingness to pay. My impression is that some organizations, chiefly Rethink Priorities, might be willing to pay somewhere between $100 and$500 per month for their posts to be tagged, categorized, and incorporated into a Wiki.

In my experience as an engaged EA Forum user, I haven’t used the EA Wiki much, though I have found the tagging system useful occasionally. I also think the local groups I’m involved with (EA Austria and EA Spain) don’t use it that much.

In contrast, Michael Aird (of both Rethink Priorities and FHI) reported a highly positive experience. He edits the EA Wiki a fair amount and finds himself often sharing articles with newcomers who could use them.

I would expect the tags dashboard to pick up editing activity if the EA Wiki’s use was more widespread. And there is some activity besides Pablo and Michael Aird’s edits, but I’d call it a very small amount.

### The EA Wiki is probably not the very best use of the corresponding researcher time

The EA Wiki seems like it has tractability and neglectedness going for it, but not scale or urgency. In particular, other research opportunities, such as joining Rethink Priorities’s rapidly expanding longtermist team, helping out Charity Entrepreneurship or the Forethought Foundation with their research, or doing some high-value original research, might be more promising.

That is, given an additional full-time equivalent (FTE) researcher in the abstract, I would probably allocate it to other opportunities. That said, producing research and producing content for a wiki are not completely interchangeable skills or activities.

### Pablo’s personal fit is solid

In contrast with the above point, previous grant rationales pointed out that Pablo has a particular affinity to wikis, meaning he is better than average and feels more at home with them.

### Pablo himself might have the best intuitions about the EA Wiki’s value, and he still views it as worth doing

Although I’ve spent a fair bit working on this evaluation, Pablo has been working for upwards of a year. He is an excellent forecaster, and I imagine that his models of the world are more developed than my own around wikis.

Per personal communication, Pablo still feels fairly excited about the EA Wiki. However, also per personal communication, he has experienced some burnout. It seems plausible to me that mental health considerations might dominate.

## Summary of considerations

Positive:

• Words per dollar are similar or cheap compared to adjacent industries.
• Pablo could use the EA Wiki as a launchpad for related projects.
• Pablo’s personal fit is solid.
• Pablo himself might have the best intuitions about the EA Wiki’s value, and he still views it as worth doing.

Negative:

• Pablo is spending comparable amounts of time working on the wiki as users (who are presumably less senior) are spending browsing it.
• The EA Wiki—and in particular the editing festival—doesn't seem to have resulted in great movement-building efforts.
• The EA Wiki doesn't seem that useful right now in terms of organizing research efforts.
• The EA Wiki is probably not the very best use of the corresponding researcher time.

Neutral:

• The EA Wiki probably needs more time until it can be determined whether it is a success or a failure.
• The case for the EA Wiki relies on increased usefulness in future years.
• More details could be gathered about the EA Wiki’s usefulness.

## Conclusion about the EA Wiki for grantmakers

Overall, I tend to think that considerations in its favour outweigh considerations against it. With CEA looking like it will invest in the EA Forum and movement-building for the long-term, the EA Wiki’s usefulness will probably rise as there are more EAs and the EA body of knowledge grows wider.

However, I would still pay attention to metrics like user engagement, feedback in surveys, and a general sense of whether the wiki is being useful. I’d update against there being a burning need in the EA community to edit wiki articles.

## Conclusion about the EA Wiki for Pablo Stafforini

As in the previous section, I think the EA Wiki is worth it, but I don’t think it’s the most valuable or ambitious thing Pablo could be doing. In particular, I dislike that users—whom I believe are on average more junior than Pablo—use the EA wiki for roughly as many hours as Pablo put into it.

Doing projects on top of the EA Wiki and writing more in-depth articles rather than stubs seems like it would be robustly good. Pablo could also take value-of-information options, like trying out original research.

Also, Pablo could consider posting particularly in-depth articles to the EA Forum as posts first, purely as an intervention to improve motivation.

## Acknowledgments

This evaluation is a project by the Quantified Uncertainty Research Institute, commissioned by Pablo Stafforini and Max Daniel, financed by the EA Infrastructure Fund, and conducted by Nuño Sempere. Pablo and Max reviewed this document prior to publication.

Thanks to Misha Yagudin, Ozzie Gooen, Michael Aird, and Vivian Belenky for their comments and suggestions. Thanks also to the EA Forum Team for access to their Google Analytics data. Thanks lastly to Pablo Stafforini for reaching out to suggest this project.

## Appendix: A Guesstimate for the Impact of the EA Forum

This guesstimate estimates that there are about 7 hours (3 to 13 hours) of engagement per hour that the EA Forum team puts into it (I imagine that CEA might be able to get more granular estimates).

When taking into account the time other agents put in (the LessWrong team, Rethink Priorities, other authors, readers), the Shapley value for the Forum team might be around 2 hours (0.86 to 3.5 hours) of engagement per hour put into it.

I’m not yet sure how to interpret that, but it seemed like an interesting fact to note. Less sloppy accounting might also consider that some contributions to the EA Forum codebase also go to LessWrong and the Alignment Forum.

## Appendix: Thoughts and Reflections

### Background

In 2020-2021, Pablo Stafforini got a grant from CEA. As part of the grant, they asked him to get the project evaluated. He reached out to me to do this work. This evaluation was meant to take 10 to 30 hours, with funding coming from CEA. I’m doing this evaluation on behalf of the Quantified Uncertainty Research Institute.

### Reflections

This evaluation took me approximately 50-80 hours to write. I got feedback from Misha Yagudin, Michael Aird, and Vivian Belenky, which likely cost between 4 to 10 hours of their time. Ozzie Gooen helped oversee and edit the evaluation; this took approximately 12 hours of his time.

This is the first paid evaluation that QURI has done. As such, our team spent more time with it than the project was budgeted for, and our team made some mistakes that we’ll learn from for future evaluations. In particular, I was intending to spend 10 to 30 hours, but that proved insufficient, so I ended up spending 50 to 80 hours instead. The difference in hours won’t be charged to CEA.

Overall, the things I would do better are:

• Don’t start the evaluation until I get the Google Analytics data
• Talking to Pablo sooner would have been beneficial

Comments and my own perfectionism made this evaluation significantly better, but also resulted in me spending significantly much more time. But I don’t exactly know how much time I’ve spent on this, because I didn’t really have a good time-tracking software. I’ve installed a simple time-tracker in order to solve this in the future.

Overall, there seems to be some missing theory and practice for evaluations about speculative/longtermist interventions, which I’m hoping to develop in the coming months. For a specific example, there really weren’t any evaluation templates I could use.

It also seems like this evaluation was unusually hardcore, spending $4k worth of time in an evaluation for a$50k project seems somewhat unusual as far as I can tell. But on the other hand, given significant uncertainty about how valuable speculative/longtermist projects can be, it’s not clear that this would be unreasonable, and I’d be curious to repeat this for other individuals, organizations, or projects.

# 74

New Comment

I'm so impressed that Pablo asked for an external review when he was feeling potentially burnt out and not sure about the impact of the wiki. That takes some incredible epistemic (and emotional!) chops. This is an example of EA at its finest.

I also want to flag that I believe Pablo explicitly agreed to have this review made public. I think it would have been very easy for this to be kept private, and I think the value is much greater of it being public, so I'm very thankful he did this.

Thanks for this!

As an infrequent long term user of the forum, I just found the link to the wiki section (for the first time that I remember). I had seen the tags at the top of posts and occasionally checked or added them, but that was it.

It looks excellent based on some quick browsing, and I can imagine it being very useful to me and others in the community as it gets fleshed out.

Perhaps my user experience helps explain the lack of current usage. I have two user journeys on the EA forum: I come to read or post. When I come to read I, I see many good posts, many more than I have time to engage with, look at one or two, (attempt to) make a quick comment then leave. When I come to post, I post quickly then get out. At no point does the wiki really fit into either journey or tempt me to go on a different journey (e.g., to browse new additions).

It has been a while since I was actively seeking to learn about EA (in a broad an unfocused sense), so I am not sure if I would come to this wiki for that. Possibly.  If that is one of its intended uses, the fact that I have never really considered that as an option, despite being on the forum so much might be a problem.

My update from reflecting on all that:

• Perhaps there should be more triggers for using and engaging with the wiki?
• Possibly once a period, one wiki article could get featured in the front pages posts, with some accompanying blurb about the wiki and the value of adding updates? And/or an update about the new content added this month?
• Possibly there is a need for spreading more awareness about the depth of information available on the wiki via a social media campaign across our various groups? I have never seen a post about it outside the forum, I think.
• Maybe we nudge more social media groups, websites and services in the EA ecosystem  to link to the wiki when referencing key concepts?
• Possibly the forum could have a new posting flow where people are asked to submit their post to a relevant wiki article before they add a post?
• Perhaps there should be more incentives to update the wiki?
• Right now people see to get more rewards and status for posting in the forum than working on the wiki and I don't think that that is actually reflective of the respective impacts of the work. A good wiki probably build knowledge better than a lot of unlinked posts on different topics.
• Possibly EA granting organisation could require the final results of funded research/reviews to be integrated into the EA wiki?
• Perhaps it is worth doing some user/conversion testing on the wiki? How do people arrive at it, what do they do, what do we want them to do etc? What percentage of EAs/EA forum users even know about it? What percentage of those who use it find it useful? Etc.
• Perhaps it needs a user growth/marketing person to help Pablo to promote it to key groups who might not know how and why to use it?

Here's one idea: Automatic or low-effort linking to wiki-tags when writing posts or comments. A few different versions of this:

• When you write a comment or post that has contains the exact name of a tag/wiki article, those words automatically link to that tag. (This could potentially be turned on/off in the editor or in your personal prefs.)
• The same as the above except it only happens if you do something special to the words, e.g. enclose them in [[double brackets]], surround them by [tag] [/tag], or capitalise correctly. (Magic the gathering forums often have something like this for linking to cards.)
• The same as the above, except there's some helpful search function that helps you find relevant wiki articles. E.g. you type [[ or you click some particular button in the editor, and then a box for searching for tags pops up. (Similar to linking to another page in Roam. This could also be implemented for linking to posts.)

Strong upvoted! I think something like this would introduce exactly the kinds of people whom we would like to use the wiki, to the wiki. I like the first version best, as many writers might not be aware of the ways to link to tags, and not be aware of what tags exist. Also, this nudges writers to use the same concepts for their words (because it is embarrassing to use a word linked to a tag in another sense then is explained in that tag).

There are some great ideas here! I've added them to the transition doc I wrote for my successor, in case they want to take on improving the Wiki after they've been hired.

These are all good points, thanks!

Do QURI or you (Nuno) generally accept commissions for evaluations like this? I'm potentially interested in commissioning an evaluation of BERI. Of course I totally understand if you don't have time for this, or if you're refusing such commissions for some other reason.

Hey, thanks for the question. We're considering it, but don't really know yet.

I think the EA Wiki pages are really valuable as a list of posts relevant to a topic. It's great to see the most important posts to get up to speed with a topic, such as climate change or suffering-focused ethics. But I'm unsure that it's worthwhile to have very detailed pages on the EA Wiki to introduce a topic. It's hard for random wiki editors to write a quality article summarizing all the relevant points on a topic such as longtermism, with due balance given to different considerations, and I'm not sure that it generally adds enough value compared to the list of tagged articles. I think high-quality articles introducing an topic would come from people who want to write their own article from a certain perspective, rather than through a collaborative wiki editing process. Even on Wikipedia, high-quality articles are often due to one or two editors writing the vast majority of the article.

That said, I think EA Wiki tags should all have at least a one-line description of what the topic is.

I'm not sure how much of Pablo's time is spent on writing articles vs tagging articles. But if it were largely for tagging articles, then time spent on the page (which I believe "engagement" is) would be a poor measure of how useful readers find the wiki. I might spend 20 seconds reading an EA Wiki page but 10 minutes reading the articles linked. That would be pretty valuable if I read better articles thanks to the EA Wiki or if I learn more about the topic, than if the EA Wiki were lower quality.

EA Wiki engagement might be boosted if the home page of the EA Forum had a section with recent edits to the wiki, just like how it showcases recent posts. If the recent edit section shows a count of the number of characters changed, that could help people who make a major edit to the wiki get recognition. However, I'm not confident that recent edits are interesting/important enough to be on the main page. There is a Wiki tab on the main page, but to be honest, I had never clicked on it before writing this comment.

Thanks for the comment. Your observation that EA Wiki pages are valuable mostly by virtue of providing a list of posts relevant to a topic is something other people have noted, including recently in the short exploratory survey we conducted. This is a question that is obviously quite important for us to answer, since it has the potential to influence significantly how we prioritize our efforts or even whether it is worth having paid contributors at all. Answering that question is also relevant for assessing the value of some possible directions the Wiki may take, such as asking experts on a given topic to write an in-depth overview of it.

So far, I have prioritized breadth over depth, since I think at the current margin having e.g. five one-paragraph articles generates more value than having five one-paragraph articles. However, there are still one or two dozen articles of medium length, which may allow readers to determine how much value, if any, they derive from a Wiki entry that goes beyond listing the key works in the literature or offering a one-line description of the topic. I list some below in case anyone wants to take a look and share their thoughts. You can also submit your impressions anonymously and I'll post them myself as a comment.

Thanks for doing this Nuno! I'm a little confused about the Google analytics data: it's showing that there were 10,000 users/week for the wiki during Aug 2021-Sept 2021, but there are also 10,000 users/week for the forum total in that same time window. So ~100% of our users were on the wiki? That seems implausible to me.

Edit: I think I see the view you are using (I work at CEA). What you are doing seems like it should work, but I definitely see page views on things that are posts, not wiki articles. Possibly this is looking at all users who saw a wiki article at some point in their journey?

Hey, thanks for kindly pointing out that mistake. I think I've found a google analytics view that mostly corresponds to what I want, and I've updated the post accordingly. But do let me know if it still looks wrong.

Is it accurate to say that when Pablo was working on this, there was 1 non-Pablo hour for every 1 Pablo hour. And now there are a similar number of non-Pablo hours happening as before, for no Pablo hours, on an ongoing basis?

I guess one of my key questions would be why the old Less Wrong Wiki seemed - while not fantastic - seemed kind of useful.

The EA Wiki seems probably worth funding, but it is not the most ambitious project that the main person behind it could be doing.[emphasis mine]

This is a really minor point, but I think your phrasing here is overly euphemistic. "most ambitious project" taken very literally is a) a very high bar and b) usually not a bar we want people to go for [1]. To the extent I understand your all-things-considered views correctly, I would prefer phrasings like "While I think on the margin this project is worth spending EA dollars on, I do not believe that this project is  higher EV than other likely candidate options for Pablo to work on" or stronger wordings like "I am reasonably confident that other  likely career options for Pablo have significantly higher EV."

[1] A caricatured example of "most ambitious project" might look more like "become God-Emperor of the world" or "solve AI alignment in one week."

I thought it was clear enough, fwiw.