Dana Callista Lexa 🏳️‍⚧️

Uhhh no, to any of the couple things I've published, but look, the next thing I want to write is a Magical Girl hack for City of Mist, because I actually want to run that for some friends soon, so if I do that and clean up my notes to publish, I'll have actually played something on my itch page.

What is the purpose of automation? I've been working with and researching generative algorithms for almost fifteen years now, and I've used all sorts of techniques in all sorts of ways. I've also gotten to watch as procedural generation changed from a relatively rarely-seen technique to a core part of several game genres and a major point of research and development for some of the biggest game developers in the world. Yesterday, Unity announced several new tools powered by some of the latest trends in AI - you'll have heard lots of different names for them, but let's call them 'generative AI'. I'm skeptical about a lot of it, and I shared some of the reasons why on Twitter. I wanted to expand on one point in particular in this blog post: what is automation actually for?

who the fuck is scraeming WE WANT SHORTER GAMES WITH WORSE GRAPHICS at my house. show yourself, jokester. I will never make a short game with bad graphics

"i'm sorry please don't shoot me i've been doing pixel art for half my life i don't know how to make worse graphics please i have a family"

There's a bunch of discourse on my tl about respect and curiosity towards game design and I agree generally speaking but also I think for me there's an element of optimism to statements like these in that probably 70% of the games I've ever worked on were, to quote the excellent Karla Zimonja, "drunk-walking towards completion" and sometimes you really do end up with a design cobbled together from a bunch of goals you're not sure how to execute, a bunch of decisions that might have been good ideas separately or at the time and now you're stuck with them. Treating that as always intentional and artistic is well meaning but well and truly, sometimes game development is in fact a polite disaster that somehow turns out okay (or doesn't)

joking aside I do feel like it's possible to be able to tell—it's just that you have to come at it from the angle of "how did this further the goal it was intended to do?" sometimes you have good decisions piled up on top of bad ones, or decisions made to offset things that don't work, or more rarely in indie but insanely common in AAA, things that achieve their goals in isolation but don't necessarily interface with other decisions made in unrelated places by a completely different part of the team. nevertheless, I think you can't tell if something's a failure vs personal frustrating vs friction that exists to change the way you interact with the other parts of the game unless you understand what the goal of the decision was. and if goals end up being contradictory, well, then you've probably got an interesting failure worth talking about too, or at least a source of interesting tension.