markjgsmith

I'm seeing and hearing a lot of people in early adopter circles getting very interested in AI generated music the past few weeks. It seems to be moving from this is just a silly joke to hold on there might be something interesting here territory. #

Spent several hours today on the social media auto-poster. It’s starting to look really awesome. The project structure is just right and everything is easily repeatable. I showed Gemini the existing setup with a bit of explanation, and basically just said, do the same sort of thing for some other social networks, and it obliged, and synthesized new code based loosely on the existing structure, but with relevant variations to accommodate the different APIs. Gemini is shockingly great when the structure exits and the task is very clear. #

The combination of the Nominative Determinism and Last Mile Synchronicty of this piece about a duo successfully rowing across Pacific Ocean is quite something. #

It occured to me that one of the advantages of having gone down the route of having many plugins for my ssg, even though that became a bit of a headache, was that my github actions made very good use of reusable workflows. I spent a bit of time the past few days seing if I could repurpose the same structure to build an social media auto-poster. I’ve made a ton of progress, and have the core primitives working.

In some ways it feels like I am very close to get it working, yet I know from experience that sometimes getting that last 20% working is deceptively much more work than first appears. All that to say, that I am quite happy that is seems like I might have something working soon, but a bit tentative because I know there are likely a lot of small details that need to be figured out. I’ve been here many times before, and I can feel the trauma of past experience holding me back. #

Yesterday evening I made a start on a small refactor to the backend API handling of scopes. It should ultimately result in fine grained access tokens that make a lot more sense than they currently do. Trying to get my head back into typescript after several weeks of quite intense bash and dockerfiles. #

Mostly cleanup tasks today following on from all the recent progress in the dev environment and on the blog. I need to remove all the old projects that got consolidated and I need to check a few things. I noticed this morning that in Feedly the everything RSS feed for the blog wasn’t showing any posts, so there might be something broken there, I need to check all the RSS feeds. #

Dylan Field (CEO Figma) [09:02]: "Is natural language the interface? Yes right now. I‘ve said this before but I really believe it, I think we will look back at this era as the MSDOS era of AI, and the prompting and natural language that everyone is doing today is just the start of how we are going to create interfaces to explore latent space. So I cannot wait for an explosion of creativity there. I think of these models as an N-dimensional compass that lets you explore this wild, unknown, fog of war in latent space and you can push these models in different directions, through natural language, but if you have a more constrained N there, and you are able to dimensionality reduce a bit, so you can push different ways, there should be other interfaces available than text. These might be more intuitive, but they also might be more fun to explore, and I think sometimes constraints unlock creativity in ways people don't expect. So I'm existed for that, but yes right now natural language is where we are at." #