A Case for Better Feeds

post by Nathan Young (nathan) · 2021-08-24T16:28:37.743Z · EA · GW · 15 comments

Contents

  Introduction
  Databases are important to EA
  Databases are like water tanks
  Sounds dangerous
  What is happening so far
  A proposed framework of some of the best ideas
  Some worse but perhaps interesting suggestions
  Cost
  Impact
  Thanks
None
15 comments

Tl;dr

Introduction

Much governance is database management.


Think about any modern service organisation. It has many databases, often creating a large proportion of its value. Whether Amazon’s stock lists, your government’s driving test passers, Tinder's millions of swipers or the lists of scores that power Google’s searches, many organisations are building scoring and rearranging databases and charging you for the privilege of accessing them or the objects they track.

Databases are important to EA

Much of EAs work is on the ground, doing research or supplying antimalarials. But around and behind these operations are databases, including several of EAs leading lights:

For a subset of these, there would be value created by making the databases easier to access and take feeds from.

Databases are like water tanks

Water isn’t useful if it sits in a tank. It needs to go somewhere eventually. Data is the same. Either it needs to be movable or we shouldn’t be storing it in the first place. And move it does:

But there are other spaces that it could be easier for things to move to notably:

Sounds dangerous

Giving everyone access would be bad, which is why I’ve tried to suggest a range of easy database “pipes” that would be low risk. Largely it’s about allowing more options for things that exist already. I think if this goes well there is a larger discussion about other databases under EA control.

What is happening so far

Well Niko Bosse has built two of the suggested pipelines (forum posts to twitter, twitter jobs to twitter - we worked on this together). But he’s only one man doing it in his spare time. I think this is a pretty tractable way to make a lot of EA’s stuff easier to access.

A proposed framework of some of the best ideas

Some worse but perhaps interesting suggestions

Cost

I guess a single developer could do all of the “best ideas” in a year, it’s over different orgs but you get the picture. I’ve tendered out work but maybe it’s much harder than I understand. Let’s times it by 3. $300k.

Impact

I’m not confident about how to ballpark impact for this. And I'm pretty tired, and I wanted to actually publish this. If I were going to work it out I would use the cost of the 80k jobs board here and try and find some estimate of impact from the jobs board (I haven't been able to find) and assume that this increased it by 5%. Similarly with forum views. Feel free to try a better job in the comments.

However, to finish I'd like to get a sense of the following two questions. Please write your feelings on them:

  1. Would you apply to more jobs if you could have them just the way you wanted them?

    If they fit perfectly into your workflow, would you apply to more stuff?
     
  2. Would you read more EA content if you could have it in the form you wanted it?

    Would you read, or be otherwise more engaged in the community if content came to you exactly how you wanted?

Thanks

Thanks to all who read this whilst I was writing it. Please notify me of mistakes and I'll correct them. Thanks for all you (yes you reading this) do, it means a lot to me. It is my job to know you all and work with you all on trying to make the world better. 💙

15 comments

Comments sorted by top scores.

comment by Pablo (Pablo_Stafforini) · 2021-08-24T16:43:24.749Z · EA(p) · GW(p)

One reason for pessimism about the value of this kind of proposal is that a website I've built, Effective Altruism Blogs, receives extremely little traffic (about five users/day, according to Google Analytics).  The site aggregates content from a long list of curated sources, which I keep regularly updated. Although I know a few people who told me they derive considerable value from visiting the site regularly, this is clearly not the case for the vast majority of EAs, for whom "get[ting] the information from those databases to where people read them" does not appear to be a problem in need of a solution.

Replies from: nathan, alexanderklarge
comment by Nathan Young (nathan) · 2021-08-24T17:15:55.998Z · EA(p) · GW(p)

Thanks for reading.

I think your key case study here is different that what I propose. You built a new site, I'm suggesting making it easier for people to use their current apps (email, RSS readers, this site) to access EA content, not to get them to change. I agree that getting people to use a new thing is hard.

So to your second point, I agree.  I don't know how many would find this valuable. Though my sense is that better 80k emails might get 5-10% take-up. Is this worth it?

I'm sorry if I've been unclear. I don't think this is about changing things people currently like, but about giving more options to those who currently "bounce off" EA content

comment by alexanderklarge · 2021-08-24T18:18:41.739Z · EA(p) · GW(p)

Wow, this website looks like a great resource, will definitely be returning to it!!

One thing to note is that a search bar would be really handy, and maybe an option to filter by date etc. And perhaps a really quick about page.

comment by Aaron Gertler (aarongertler) · 2021-08-25T06:19:18.008Z · EA(p) · GW(p)

Regarding groups and events: The current feature [? · GW] is a prototype being tested by a few groups before we roll it out for everyone to use. But if all goes according to plan, the Forum will eventually be used by hundreds of groups to share their profiles and events, with many new features to help out.

Meanwhile, the current EA Hub list is easy to filter and mostly up-to-date (groups can update information themselves), and it's heavily linked across the mainstream EA internet — hopefully, it's pretty hard to miss, even if the current list is feature-light. This page is an above-the-fold Google result for "effective altruism group" and "effective altruism meetup", so if someone knows that EA groups exist at all, I think they'll find it. (And if they don't, I'm not sure what a new resource would do.)

Also, if you want to get more feedback on questions like this, consider using EA Polls [EA · GW] — the whole comment upvote/downvote thing seems to get less input than the average poll in that group.

Replies from: nathan
comment by Nathan Young (nathan) · 2021-08-25T09:48:31.856Z · EA(p) · GW(p)

So quick note on the forum experience here:
- I spent 3-8 hours writing and redrafting a post
- The post has recieved little karma so clearly people didn't like it
- A lot of people liked a critical comment which I guess implies they think that this isn't a high priority issue
- I have no feedback other than that
- I feel hurt and a bit dumb

This is okay, I'm an adult. I can get over feeling a bit dumb. but sometimes I read that that we'd like more people to post on the forum. For me, this is why I don't.

I don't mind having written an article that people didn't like, but I'd really appreciate some feedback.  Having an article that languishes feels worse than people saying clearly what would have made them upvote.

Some closing thoughts:
- I still think that a norm of allowing karma free comments would allow more feedback ("I don't think this is high enough impact", "I think this is a concern of only a few people" etc)
- Perhaps people could give private feedback while an article is small. Maybe people didn't want to hurt my feelings
- Perhaps articles could be reset to draft. I imagine I'll try and write a better version of this article at some point I still believe it to be true
- It's good that people don't upvote articles they don't like. While I feel a little bad at the moment it's better than that the community gets recommended unsuitable articles.

I don't want a response from you here Aaron, I just thought I'd allow others to see my experience in case they shared it.

Replies from: Pablo_Stafforini
comment by Pablo (Pablo_Stafforini) · 2021-08-25T13:52:48.067Z · EA(p) · GW(p)

A lot of people liked a critical comment which I guess implies they think that this isn't a high priority issue

If you mean the comment I posted, I would like to clarify that it wasn't meant as a critical comment, but rather as a weak consideration for thinking that this kind of project may not be as valuable as it may initially appear. My apologies for not communicating this more clearly.

Moreover, as you note in your reply, which received almost as many upvotes as my comment did, the case I discuss is different in some key respects from what you were proposing, so it's unclear whether my experience should inform estimates about the value of that proposal. 

Replies from: nathan
comment by Nathan Young (nathan) · 2021-08-25T14:56:53.195Z · EA(p) · GW(p)

Hey :) I don't think critical is bad and I want peoples comments both in support and critical. Thanks for clarifying though, that's kind of you.

I would label your comment as disagreeing with the thesis of the post and since it has more upvotes than the post, it suggests others agree with you. Now since we agree that the situations are a bit different I don't quite know what people's specific disagreement is. And that's okay, people can do what they like, but it is disheartening to write an article, receive relatively little takeup, have people like a disagreeing comment more than that article and then not know why. Right?

comment by Nathan Young (nathan) · 2021-08-24T20:07:37.292Z · EA(p) · GW(p)

Upvote this if you'd read more EA content if you could have the right feed go to where you read it.

Downvote the comment attached to this so I end up with neutral karma.

Replies from: nathan
comment by Nathan Young (nathan) · 2021-08-24T20:08:05.509Z · EA(p) · GW(p)

Downvote this so Nathan has neutral karma

comment by Nathan Young (nathan) · 2021-08-24T20:05:57.347Z · EA(p) · GW(p)

Upvote this if you would apply to more EA jobs if you could sort them more easily add they arrived how you wanted them.

Downvote the comment attached to this so I end up with neutral karma.

Replies from: nathan
comment by Nathan Young (nathan) · 2021-08-24T20:06:25.991Z · EA(p) · GW(p)

Downvote this to ensure Nathan gets neutral karma

comment by Nathan Young (nathan) · 2021-08-24T16:33:09.376Z · EA(p) · GW(p)

Also I'd value feedback on things you really liked and really didn't like about this article. I have about 3 half-written ones and knowing what the community likes and dislikes would give me more confidence.

comment by markus_over · 2021-09-04T11:39:24.170Z · EA(p) · GW(p)

Recently I've been thinking a lot about the flow and distribution of information (as in facts/ideas/natural language) as a meta level problem. It seems to me that "ensuring the most valuable information finds its way to the people who need it" could make a huge difference to a lot of things, including productivity, well-being, and problem-solving of any kind, particularly for EAs. (if anybody reading this is knowledgeable in this broad area, please reach out!)

Your post appears to focus on a very related issue, which is how EAs source their EA information and some specific ways to improve it. I definitely agree that this is an issue worth looking into and worth improving (I personally think that either the EA forum or the EA Hub are in the best position to make such improvements, although I'm unsure what these improvements would look like).

The EA Forum Job Hunt idea admittedly doesn't seem very promising to me from how I understood it -- it sounds like by far the most work of all the suggestions, for a problem that, to me, seems as if it's solved to a pretty reasonable degree. 

I don't quite understand the EA Hub suggestion. What would be submitted and upvoted? Just the existence of (local) groups?

The remaining points regarding Twitter bots and feeds sound good to me, simply because they sound like very little work (unless I'm misjudging that), while potentially being helpful to probably many dozens of EAs.

By the way I do wonder what ratio of EAs is actively using twitter. I for once am not at all, and am not aware of many people I know personally doing so, but that might not mean much and not be very representative.

comment by AlexanderSaeri · 2021-08-25T04:30:58.331Z · EA(p) · GW(p)

I use Feedly to follow several RSS feeds, including everything from the EA forum, LessWrong, etc. This lets me read more EA-adjacent/aligned content than if I visited each website infrequently because Feedly has an easy to use app on my phone.  

Here is a screenshot on browser of my Feedly sidebar. (I almost never use a browser)
Here is an example of the Feedly 'firehose' from my mobile phone, previewing several posts from EA forum and elsewhere.

 

I liken it to a 'fire hose' in that I get everything, including all the personal blogs and low-effort content that would otherwise be hidden by the website sorting algorithm. There's also no (displayed) information in Feedly about the number or content of comments - instead I need to open each interesting post to find out if someone has commented on it. 

For some posts, the post content is the most valuable. In other posts, the commentary is the most valuable, and Feedly/RSS does a bad job of exposing this value to me easily. I also find that engagement is highest within the first 1-2 days of a post, but takes several hours to start. 

All of this is to say that I think the 'right' feed is probably still something like one or more RSS feeds - especially given their interoperability and ease of use - but that the user experience is likely to be highly variable depending on their needs and appetite for other- vs self-curation of what is in the feed.

Replies from: nathan
comment by Nathan Young (nathan) · 2021-08-25T09:36:22.731Z · EA(p) · GW(p)

This is exactly my point. Imagine you could customise your RSS feed like you do your front page.