So here's what Elliot said in reply to my blog post:
> https://ramirustom.blogspot.
>
>> I noticed that it took me 30 sessions of Cap kingdom (first kingdom) to go from ~200% of WR time to ~119% of WR time. So I thought that each kingdom after that would need at least that much because those kingdoms are more difficult (assuming all other things are constant, like my skill). This is a vague impression I got a long time ago when I first started speedrunning. And to account for the extra difficulty of the next kingdom, I decided to reduce the goal from 119% to 125%.
>
> I think you should prioritize finishing a whole run. 200% of the WR time is fine. start by aiming there (or even 300%), then get it down to 175% then 150%.
>
> then you figure out which parts you have trouble with and practice those.
>
> don’t optimize the small parts so much b/c you don’t have the whole picture yet, and don’t know where the bottlenecks are, and optimizing the details a lot delays finishing the first big milestone (completed any% speedrun using an actual speedrun route and split timer instead of just playing the game casually).
>
>> I noticed that it took me 30 sessions of Cap kingdom (first kingdom) to go from ~200% of WR time to ~119% of WR time. So I thought that each kingdom after that would need at least that much because those kingdoms are more difficult (assuming all other things are constant, like my skill). This is a vague impression I got a long time ago when I first started speedrunning. And to account for the extra difficulty of the next kingdom, I decided to reduce the goal from 119% to 125%.
>
> I think you should prioritize finishing a whole run. 200% of the WR time is fine. start by aiming there (or even 300%), then get it down to 175% then 150%.
>
> then you figure out which parts you have trouble with and practice those.
>
> don’t optimize the small parts so much b/c you don’t have the whole picture yet, and don’t know where the bottlenecks are, and optimizing the details a lot delays finishing the first big milestone (completed any% speedrun using an actual speedrun route and split timer instead of just playing the game casually).
I want to practice analyzing Elliot's reply. I think this will help me better understand all of Elliot's other replies.
Elliot's reply is a criticism of my methodology. It's also a suggestion to change my methodology.
A criticism only has meaning in the context of a goal to be achieved. A criticism explains why the goal can't be achieved.
So what is the goal that Elliot and I are referencing? Surely Elliot has it in mind, as did/do I. But maybe we don't have the same idea about it. So I want to explore that potential difference in perspective.
So what's my goal for my speedrunning activities? Learn rationality. But that's not detailed enough. I guess Elliot and I both think that there is a time factor. Surely the idea is to learn rationality in a reasonable amount of time. Like I could imagine someone slowly learning rationality such that over 20 years he barely makes any progress. I don't want that. I want something much more than that. I want to optimize that. I want to learn as much as I can learn within the limits of my time spent on philosophy and my current knowledge (while learning ways to increase my time spent on philosophy, and of course improving my knowledge such that I can further learn rationality).
I think Elliot's criticism is designed to save time. It's designed to setup priorities such that more time is put towards the highest priorities and less time towards lower priorities.
I think Elliot's criticism has a lot of reach. I think it applies to many sorts of activities. I think I could use his criticism by applying it to all of my time spent on philosophy (i.e. time spent on my "learn rationality" goal). But I currently don't know how to do that.
What do the kingdoms in SMO represent? Different subjects like grammar, math, logic, reading comprehension, idea trees / outlines, etc. I think Elliot's idea (which comes from Goldratt I think) implies that I should do a little bit of each of these things before trying to further optimize any one of them.
This is connected to the idea of lowest hanging fruit. It makes sense to aim for lowest hanging fruit first before going after the more difficult stuff. Why? Because you get more output per unit effort that way.
Another area that I noticed Elliot's idea works for: writing essays/books. First write an outline, before trying to optimize a portion of the outline (like a paragraph or a sentence).
Hmm, I just noticed that Elliot's idea also applies to my reply to his post (that contains his idea). I'm trying to optimize the whole thing first as opposed to starting by optimizing a part. The whole thing refers to all that I can get out of Elliot's idea -- everything I learn from the entire series of posts I plan to do in response to Elliot's idea (3 posts so far, including this one). A part is how his idea applies to my speedrunning work.
No comments:
Post a Comment