Viliam_Bur comments on Open thread, Jan. 12 - Jan. 18, 2015 - Less Wrong Discussion
You are viewing a comment permalink. View the original post to see all comments and the full post content.
You are viewing a comment permalink. View the original post to see all comments and the full post content.
Comments (155)
I think I have noticed a frequent failure pattern when people try writing about complicated stuff. It goes like this:
Instead, this is what seems like a successful pattern:
Seems to me that Eliezer followed the latter pattern when writing Sequences. There is no part saying "this will make sense to you only after you read the following chapters I haven't written yet". But there are parts heavily linking the previous articles, when they advance the concepts already explained. The outline can be posted after the articles were written, like this.
I understand the temptation of posting the outline first, but that's a huge promise you shouldn't make unless you are really confident you can fulfill it. Before answering this, read about the planning fallacy, etc. On the other hand, with incremental writing you have complete freedom, and you can also stop at any moment without regrets. Even if you know you are going to write about A, B, C, and you feel pretty certain you can do it, I would still recommend starting with A1 instead of introduction.
I'm not sure that Eliezer outlined the posts in order- he did mention at some point wanting to explain X, but realizing that in order to explain X he needed to explain W, and in order to explain W...
Agreed. One of the ways I've worked around this is to not post the start of a sequence until it's mostly done (I have the second post to this sequence fully finished, and the third post ~2/3rds finished). I'm not sure I'd recommend it- if you find the shame of leaving something unfinished motivating, it's probably better to post the early stuff early. (I let that particular sequence sit for months without editing it.)