Less Wrong is a community blog devoted to refining the art of human rationality. Please visit our About page for more information.

Comment author: Vaniver 27 November 2016 10:28:58PM 3 points [-]

Having "trouble to find people willing to do the work" usually means you are not paying enough to solve the problem.

I had difficulties finding people without mentioning a price; I'm pretty sure the defect was in where and how I was looking for people.

I also agree that it makes more sense to have a small number of programmers make extensive changes, rather than having a large number of people become familiar with how to deal with LW's code.

I believe in long term it would be better to rewrite the code from scratch, but that's definitely going to take more than one month.

I will point out there's no strong opposition to replacing the current LW codebase with something different, so long as we can transfer over all the old posts without breaking any links. The main reason we haven't been approaching it that way is that it's harder to make small moves and test their results; either you switch over, or you don't, and no potential replacement was obviously superior.

Comment author: ananda 29 November 2016 05:31:53PM 18 points [-]

I'm new and came here from Sarah Constantin's blog. I'd like to build a new infrastructure for LW, from scratch. I'm in a somewhat unique position to do so because I'm (1) currently searching for an open source project to do, and (2) taking a few months off before starting my next job, granting the bandwidth to contribute significantly to this project. As it stands right now, I can commit to working full time on this project for the next three months. At that point, I will continue to work on the project part time and it will be robust enough to be used in an alpha or beta state, and attract devs to contribute to further development.

Here is how I envision the basic architecture of this project:

  1. A server that manages all business logic (i.e. posting, moderation, analytics) and interfaces with the frontend (2) and database (3).
  2. A standalone, modular frontend (probably built with React, maybe reusing components provided by Telescope) that is modern, beautiful, and easily extensible/composable from a dev perspective.
  3. A database, possibly NoSql given the nature of the data that needs to be stored (posts, comments, etc). The first concern is security, all others predicated on that.

I will kickstart all three parts and bring them to a good place. After this threshold, I will need help with the frontend - this is not my forte and will be better executed by someone passionate about it.

I'm not asking for any compensation for my work. My incentive is to create a project that is actually immediately useful to someone; open-sourcing it and extending that usability is also nice. I also sympathize with the LW community and the goals laid out in this post.

I considered another approach: reverse-engineer HackerNews and use that as the foundation to be adapted to LW's unique needs. If this approach would be of greater utility to LW, I'd be happy to take it.