Re: Which LessWrong projects are the highest value?

[Luke Muehlhauser]

John, thanks for organizing the LW public goods team!

Here are my comments on the list of public good projects Google doc:

1. The bar of entry to Less Wrong is too high. The answer to every newbie question is "Read the sequences" which means "Read 500 pages before you talk to us." Summarizing the sequences is a high-value project, though it requires a very particular skillset, one that Eliezer and Yvain and myself and a few others have. Eliezer would love to see it done, but doesn't have time. Yvain has considered doing it, but I'm not sure if he will. I've considered doing it, but for sure can't get to it for several months. I made a list of posts for the project once upon a time: http://wiki.lesswrong.com/wiki/The_Sequences_Summarized

2. I do plan to continue adding to 'The Science of Winning at Life', and I'm the most natural person to do so. I have drafts of posts for that sequence that I've already put 30+ hours into, etc.

3. How many people actually use Anki every day? That number will determine how immediately valuable the creation of Anki cards is. Future value may be created in that perhaps more people would use Anki if there were more and better decks available.

4. As for introductory material, I am part way into creating a Khan Academy style video that could become a video series. It's an experiment to see if we can reach the YouTube audience and bring them to Less Wrong - or at least, get them to consume Less Wrong memes in videos posted to YouTube. I expect to finish making this first video sometime in August.

5. Book reviews are quite useful. Kaj and badger have done great book reviews in the past.

6. The most high-value projects that need doing are probably tweaks to the Less Wrong code base. Trike Apps is working on some of these, but Trike Apps is usually pretty slow to implement changes. (But they are donating their time!) Probably nothing on the current 'list of public good projects' would make Eliezer happier than if he had a team of 3-10 programmers who, when informed of a small, high-value coding project, would implement it within a week or two. For the current status of Less Wrong coding projects and where help is needed, you should contact Louie Helm. He is in direct communication with Trike Apps.

7. Singularity Institute researchers could definitely benefit from more people who are willing to (1) download PDFs of papers from university computers, and (2) do literature searches. Since there are already a few people who occasionally do this via the Singularity Institute Volunteer Network, the best way for a volunteer to do this would be to contact Louie Helm and tell him you want to be part of the volunteer network and you especially are available to (1) download papers from behind paywalls and/or (2) do literature searches.

8. Yes to LW image hosting service! This is also another thing that could be improved about the Less Wrong code base. One could click the image button and be able to upload an image, and also choose the horizontal alignment - something that currently must be done by hand-editing the HTML. There are other improvements to the 'compose article' page that could be made.

9. Developing meet up exercises and materials is valuable, though it may be best to piggy-back on the rationality curriculum that Anna and Eliezer are working on. They are building a map of rationality skills along with other things like (1) exercises to develop those skills, (2) what a person with each skill can achieve, and more.

As for the questions about low-cost, high-value projects, I suspect those mostly lie in the 'programmers willing to quickly improve the Less Wrong code base in small ways' space. Willing, capable, and fast programmers can do the most good with the smallest investment, I suspect.

That's all for now.

Luke

On Fri, Aug 5, 2011 at 9:07 AM, John Salvatier <jsal...@u.washington.edu> wrote:

Dear Luke,

I hope I find you well!

Many projects would benefit people interested in living rationally but no individuals are motivated enough to execute them (more). The Public Goods Team is trying to encourage and facilitate systematic and organized work on such projects. One of our first steps is to identify the projects that are the lowest hanging fruit: those projects which are high value, and inexpensive (in terms of time, motivation, money etc.). We have come up with a preliminary list of such projects.

We are especially interested in your judgment of which projects are most attractive, because we think the judgments of smart people who have thought about this sort of thing are especially informative. Thus we are conducting a small survey of the LessWrong top 10 contributors list and others who seem relevant. We will use the results of the survey to help us focus our efforts on the most valuable projects. We will also write up our results and post them to LessWrong.

We have two questions:

• What projects not on the list seem low cost and high value?
• What projects seem like they are especially high value and low cost?

If you would like to reply publicly (very much appreciated), please send your reply to lw-public-...@googlegroups.com

Cheers,
John Salvatier