I have thoughts about "Mediums Overpower Messages". I would spell them out, but I'd be interested in trying to Socrates you about them. That work for you?

I love Socrating.

My reaction to that post was to take it as a massive (and at least slightly unfair) compliment. Can you figure out why?

Well, I know you as the guy who writes D&D.Sci. Is that going in the right direction?

Yes. That's it, but I'd like to see if you can figure out both reasons that made me take it as a compliment. (Pretty sure you already have the easy one but please spell it out.)

This website is dominated by non-fiction dialectic essays. Basically, a claim is started in the title and then argued for in the body.

In my post about mediums, I wrote about how different mediums make me smarter vs make me dumber. It basically came down to how hard they made me think. The harder a media format requires me to think, the smarter it makes me, and D&D.Sci forces the reader to think harder than anything else on this website.

Yes, that's the (easy half of the) answer I had in mind.

(I don't know about it being the most hard-to-consume content on the entire website but sure it's definitely above-median, I'll cop to that.)

Then there's another half of the answer.

At least; I just now realized there's actually like 2.5 pieces to it.

Want a hint?

Hint please.

You said in Mediums Overpower Messages that your list was symmetrical. I just checked and you didn't say it was "perfectly symmetrical", which is good because that would be false. Can you tell me why?

(Or just give me some counterexamples to "perfectly symmetrical"?)

Sometimes there is low-hanging fruit. For example, Dynomight's post My stupid noise journey is both easy to read and actionable. I read that post, immediately bought a pair of noise-cancelling headphones, and it changed my life.

That is actually not a class of counterexample I was considering, but that's a valid answer. Can you think of another class of counterexample?

Fiction. As a general rule, immersive puzzles are better teaching tools than abstract questions.

Is that a counterexample to "perfectly symmetrical" though? Or are you saying that making puzzles immersive is a free action / actively easier?

In theory, making puzzles immersive are a free action that makes them both easier to digest and more educational. However, crafting good narratives takes more skill than nonfiction dialectic essays. It becomes a free action after you've put the hours into learning how to tell a good story.

Which is why I like how your D&D.Sci always has a narrative. :)

I think that matches my experience pretty closely, actually: the narrative for each .Sci is usually the part that takes least time and effort on my part.

But that's also not what I'm fishing for, though it is a valid counterexample.

Hmmm. What are you fishing for?

The other ~half is the fact that D&D.Sci manages to be an awkward janky mess to make as well as an awkward janky mess to play. So if someone works in a medium like that (or better yet, at-least-kind-of-invents and at-least-kind-of-popularizes a medium like that), then per Mediums Overpower Messages . . . that person would be pretty great, no?

What a backhanded self-compliment. 🤣

From my POV, you're the one who gave it to me. :P

Your post actually came across as more congratulatory than that, tbh.

I started making D&D.Scis after trying and ~failing to make educational/rational videogames.

So it wasn't just "You're great!" but "You're great, and being bad at that thing you were bad at is actually Morally Good from at least one standpoint!"

The Darwin Games were also an awkward janky mess to make, and the first one was a buggy awkward janky mess to play too. I had thought of that as a failing, but, now that you point it out, I learned a ton from them.

"Unrelated" tangent: how do you feel about the SV concept/practice of "dogfooding"?

I like dogfooding because it guarantees authenticity. The catch with dogfooding is it often leads me to create a complicated, cryptic, badly-marketed product. My preference is to start with dogfooding to build the heart of a thing, and then iterate in the direction of easy-to-use. I feel this guarantees an authentic product.

You say "catch", but to a lsusr clone, "complicated" and "cryptic" would be selling points. (Right?)

And "badly-marketed" would get you indie cred.

Depends on the lsusr. My heuristics have evolved over time. The older I get, the more I prefer "elegant" to "complicated", and "precise language" to "cryptic language".

Okay, this Socrating thing is way harder than I thought; declaring bankruptcy here. The tension I wanted to highlight is that you have a lot of posts about the value/valor of Effort and Suffering almost as ends to themselves, and then make Youtube videos.

(Which doesn't make you a hypocrite, it means you've changed your mind over time and/or are in a symbiotic relationship with your viewers, just rules out them being lsusr-clones-circa-whenever.)

((And meanwhile I spend a lot of time wishing good University-level educational media which required no/negligible effort existed, and then make challenges players interface with through spreadsheets and .py scripts.))

We do what we can, with what we have.

For what it's worth, I am awful at predicting what will add value to people and what won't. My most popular stories lay around as unpublished drafts on my laptop for months.

So I just throw everything at the wall. The stuff people like gets passed around and the rest fades into quiet obscurity.

Aside regarding Socrating: as you're noticing, there are two situations. Either you can predict what your interlocutor is going to say or you can't.

• If you can predict them, then you can tie their mind in all sorts of knots and paradoxes which breaks their habitual logic and force a stacktrace.
• If you can't predict them, then you need to be ready for all the different possible answers. I like to ask a question with many different answers, not knowing which one to expect, and then pretend that the answer I got was the one I had predicted all along.

I also have thoughts about "predicting what will add value to people and what won't", but they should probably be in a different dialogue.