Draft: the most attractive projects

[John Salvatier]

This is what I have for the post on the most attractive projects so far. I intend to post this after Aaron posts about the LW hosting image. 

I am not very happy with the post so far. 2 seems like it sounds lame and not very specific either. Over all the list feels lame and not very inspiring. I am also not sure how to make my introduction more inspiring.  

Also, for the subsections, I feel like I am injecting too much of my own opinion. Without specifics, however, I feel that the list will be even less inspiring than it is right now. 

Feedback would be very useful to me.

We would like to encourage people to do projects which are beneficial to the community. One thing that might hold people back is not knowing what projects are attractive projects to work on.

The public goods team has come up with a list of the projects we think are the most attractive for LW in terms of both value provided to the community and the effort it would take to complete it. The team came up with this list with the input from people who had given the matter some thought. 

1. Improving the LW website. Thanks to Aaron Tucker's work, hacking on the LW website is now much easier than it has been in the past. There are many ways the LW website could be improved: 

a. Expanded Selenium tests for the site. Selenium tests will make it easier to detect bugs, making extending and maintaining LW easier.

b. The bug tracker has many bugs.

c. Larger, more customizable user pages. These will encourage a stronger community and clearer communication. By allowing users who have particular interests and skills more easily and to understand the background of people they are talking with.

d. Tagging and filtering system. Job postings, requests for help and _____not sure what else to put here_____ are all topics that may be highly valuable to a subset of LW, but not valuable to the rest. A tagging and filtering system would allow people to tune the content they see to align with their interests.  

2. Book summaries and reviews. There are many books relevant to LessWrong that could use good summaries and reviews. For example (selected because they were mentally available to me): 

a. Good and Real

b. Rational Choice in an Uncertain World

c. Selfish reasons to have more kids

3. Guides to producing other valuable material; especially ones with examples.

a. A good guide to producing a solid book review. 

b. Advice and guides for doing a literature reviews. Lukeprog had a good post about this, but there are many residual questions. 

4. Advice and materials for hosting meetups

a. Writeups for games like rejection therapy, repetition and paranoid debating

b. Exercises for practicing fermi calculations and simple numerical bayesian reasoning

[John Salvatier]

Also, I currently don't specify how we came up with this list of projects mostly because I feel like it would be mildly insulting to say "well we did this survey but didn't ask the forum". Maybe I'm wrong about that.

[Aaron Tucker]

Having new features like nice meetup posts in an appropriate section, and karma bubbles that hold all the digits is cool, and makes using the site a lot better. There's also a lot of actionable ideas as to future improvements. However, there's a lack of code that turns said ideas into features. TrikeApps has been nice enough to donate their time to write some of them, but they could use some help.

Setting up the software needed to contribute code to the site is fairly annoying, and a non-negligible barrier to contributing. To alleviate that problem, the I've made a Virtual Machine Image configured to be able to contribute to the site. Basically, you download a computer to be run as software, link it to your github account, and are ready to roll.

Instructions are available at https://github.com/tricycle/lesswrong/wiki/Using-the-Virtual-Machine

[Aaron Tucker]

Having new features like nice meetup posts in an appropriate section, and karma bubbles that hold all the digits is cool, and makes using the site a lot better. There's also a lot of actionable ideas as to future improvements. However, there's a lack of code that turns said ideas into features. TrikeApps has been nice enough to donate their time to write some of them, but they could use some help.

Setting up the software needed to contribute code to the site is fairly annoying, and a non-negligible barrier to contributing. The Public Goods Team decided that making a Virtual Machine Image configured to be able to contribute to the site would be helpful to alleviate that problem. So I made a Virtual Machine Image. Basically, you download a computer to be run as software, link it to your github account, and are ready to roll.

Instructions are available at https://github.com/tricycle/lesswrong/wiki/Using-the-Virtual-Machine