Ode to Objects

My recent exposure to the EA/Rationality memeplex has given me a lot of new great words. One has been "agentic", which I used as a prompt for my first ever post, Dropping Responsibilities. But another one that I've run into a lot recently has been the Meta/Object level distinction.

It's the kind of thing where you always had a vague sense that this dichotomy exists, but once you hear a word for it, it all crystalizes inside your brain.

I think the words match the concepts pretty well, but I'll give a brief explanation.

I'll just steal the definition from LessWrong:

Object level vs Meta level is a distinction between levels of abstraction, with the object level being directly about the specifics of the issue at hand, while meta level is concerned with the general principles which apply to a wider range of examples. It is often useful to move up and down the ladder of abstraction to get points across clearly, with the object level providing firm grounding and concrete examples for a person to grasp, while the meta layer allows them to generalize a new concept to a broad domain.

I think I spend too much time in the meta level domain. If you look through my past blogposts, I think you'll find most of them falling into a fairly meta-category. This one included. They're all huge, life reflections, that fall primarily under the blanket of philosophy. I think that meta-level posts are important and cool. But like many things in life, meta level posts should rely upon a broad base of object level posts. If all you do is meta-level things, you'll spend your whole life talking about doing things and never actually do them.

If you aren't doing anything in your life, what do you have to reflect upon? I think part of this ties into my general alignment with Nassim Taleb's views on the importance of Empiricism. Talk is Cheap. Theories are fake. What you say doesn't matter, all that matters is what you do. I think that the most interesting and valid philosophical opinions come from those who have lived the most interesting object-level lives.

I think that this is one of my biggest issues with EA. It's all meta. EA's would defend this by saying that it's worth spending a lot of time planning out what you're going to do for the rest of your life. Which I agree with. But if you spend sooo much time planning and reflecting on things that you're never able to stay your course and finish anything, you're never going to make an impact.

So, my goal for this semester (a slight readjustment, but not entirely so), is to get more interesting things done. I want to have projects, and research, and make cool plans for the summer. I want to learn all the math and computer science. I want to be competent at the object-level.

The irony of talking about how I "want" to do more object-level things in a post is not lost on me. I have a few plans to do this. One is to keep doing stuff with firelight. The other is to start doing research again. The last is to start a cool CS project.