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

Viliam comments on On the importance of Less Wrong, or another single conversational locus - Less Wrong

84 Post author: AnnaSalamon 27 November 2016 05:13PM

You are viewing a comment permalink. View the original post to see all comments and the full post content.

Comments (362)

You are viewing a single comment's thread. Show more comments above.

Comment author: Viliam 27 November 2016 10:57:09PM *  9 points [-]

Well, if someone would be willing me to pay for one year of full-time work, I would be happy to rewrite the LW code from scratch. Maybe one year is an overestimate, but maybe not -- there is this thing known as planning fallacy. That would cost somewhat less than $100k. Let's say $100k, and that included a reserves for occassionally paying someone else to help me with some specific thing, if needed.

I am not saying that paying me for this job is a rational thing to do; let's just take this as an approximate estimate of the upper bound. (The lower bound is hoping that one day someone will appear and do it for free. Probably also not a rational thing to do.)

Maybe it was a mistake that I didn't mention this option sooner... but hearing all the talk about "some volunteers doing it for free in their free time" made me believe that this offer would be seen as exaggerated. (Maybe I was wrong. Sorry, can't change the past.)

I certainly couldn't do this in my free time. And trying to fix the existing code would probably take just as much time, the difference being that at the end, instead of new easily maintainable and extensible code, we would have the same old code with a few patches.

And there is also a risk that I am overestimating my abilities here. I never did a project of this scale alone. I mean, I feel quite confident that I could do it in a given time frame, but maybe there would be problems with performance, or some kind of black swan.

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.

I would probably try to solve it as a separate step. First, make the new website, as good as possible. Second, import the old content, and redirect the links. Only worry about the import when the new site works as expected.

Or maybe don't even import the old stuff, and keep the old website frozen. Just static pages, without ability to edit anything. All we lose is the ability to vote or comment on a years-old content. At the moment of transition, open officially the new website, block the ability to post new articles on the old one, but still allow people to post comments on the old one for the following three months. At the end, all old links will work, read-only.