Posts

Comments

Comment by dregntael on You Don't Exist, Duncan · 2023-02-03T20:51:35.365Z · LW · GW

I am very glad that you wrote this post, and just want you to know that You Are Not Alone. Hearing over and over again that you are Not Normal, do not fit within any of the boxes, and eventually you start believing it yourself. So thank you for this reminder that It Is Okay To Be Weird.

Comment by dregntael on Third Time: a better way to work · 2022-01-08T07:56:33.913Z · LW · GW

Non-serious suggestion: Fractional-reserve breaking (in analogy to fractional-reserve banking)

Serious suggestion: Liquid Breaks (since the duration of the next break is like a reservoir you are filling slowly, and you can drain it when you want. It also implies flexibility.)

Comment by dregntael on Shoulder Advisors 101 · 2021-10-10T07:33:56.601Z · LW · GW

Thank you for writing this, it really got me thinking. I'm one of those people who don't really have a firm cast of shoulder advisors. In fact, when I saw this appear in fiction (and in particular in HPMOR), I kind of assumed it was just a convenient narrative device and not something real people actually do. I suppose I should read HPMOR again and try to figure out what other blatantly obvious advice I've missed.

This does seem like a extremely useful skill to have, so I'd like to practice it if possible. I just tried to imagine one of these shoulder advisors and I do get an image, but it is very blurry and seems to be "shifting" between different states. I wonder if this is something other people have also experienced, and whether there are any tricks to get a more stable image? Perhaps a fictional person would be easier to simulate since they tend to be not as complex as real human beings. Either way I will continue to try in the coming week and report back here if I get any results.

Comment by dregntael on Critch on career advice for junior AI-x-risk-concerned researchers · 2018-05-13T09:45:28.364Z · LW · GW

Thank you for posting this here, I mostly agree with the statement that acquiring skills early on is more important than producing anything directly. There's one thing that bugs me, however.

Early in your research career, you need to be in "consume" mode more than "produce" mode [...]

Counterpoint: if you spend most of your early research career in "consume" mode, you won't get any practice at producing valuable research or even know whether producing science is a good fit for you personally. I've personally seen people who are extremely good at processing large amounts of content during their studies, but got completely lost when tasked with finding and studying a novel problem that no-one had written on before. This seems like some kind of trap that many PhD student run into. Sometimes there's just no good way to learn how to do research other than, y'know, by doing research.