Game Follows Play. Software is just Content. Separate the Church of Software & Interface from State and Data
Joined January 2011
- Tweets7.2K
- Following1.4K
- Followers3.3K
- Likes2.7K
One of the force multiplier in my setups is that as part of the setup I always ask my coding agent to build a text book in parallel, in reality I never ever look at them but it turns out its really amazing when you later want to remix an idea, or build on top of how much more solid a prototype gets when you smash a bunch of text book together.
So right now I explore a game genre a day to discover some of the things that made specific genres click.
Game Engine as Artefact is progressing nicely. =)
A naive idea I have is to build my own game engine. And after what feels like 100 iterations and generations of mixing and remixing ideas I think I am getting finally closer. Now obviously the art style is "roblox meets minecraft" light but it all runs on claude vanilla chat as artefact. Which means anyone can prompt games into existence easily and you can leverage things like the free multiplayer etc. you get from anthropic too.
It seems that under the hood claude artefact are just kind of like durable objects. Right now I can prompt games into existence, I have some minimal ML capability, and a "CAD" for assets in there.
Probably something. Actually really worth a following have been a fan of @JungleSilicon for a long time now and his experiments are a great inspiration for myself and somehow he just opensourced a multiplayer shapeshifting computer
the open source repos:
github.com/siliconjungle/ink…
github.com/siliconjungle/ink…
github.com/siliconjungle/ink…
A naive idea I have is to build my own game engine. And after what feels like 100 iterations and generations of mixing and remixing ideas I think I am getting finally closer. Now obviously the art style is "roblox meets minecraft" light but it all runs on claude vanilla chat as artefact. Which means anyone can prompt games into existence easily and you can leverage things like the free multiplayer etc. you get from anthropic too.
It seems that under the hood claude artefact are just kind of like durable objects. Right now I can prompt games into existence, I have some minimal ML capability, and a "CAD" for assets in there.
Love that this patterns become more normalized. One thing is you can do this actually fairly easy without that much effort already today in Claude vanilla chat today.
I have an artefact that is a full-fledged malleable desktop I can turn it into Notion, into a Whiteboard, into a Desktop with apps as much as I want. What matter is that you have a clear minimal protocol to separate the interface from data. Once you have that every app or inside of Claude artefact is just like a map projection on top of that data.
The video is of me brainstorming a zine I am writting right now, and it reads and writes to my local knowledge hub.
So having the ability to point and prompt is amazing its why i run all my software independent from data so its keep being malleable.
more on liquid interfaces
moldandyeast.com/p/bxhy891a/
I'm now building an "improvements" feature into all of my personal vibecoded applications: a freeform text input that's hooked up to an agent that yolo implements whatever is submitted.
It's an interesting experience -- the activation energy barrier to sanding any rough edge or implementing any new feature becomes extremely low. Someone else who uses one of the applications I created has already implemented 150 improvements over the course of a few days. This pattern feels like a new kind of tool -- liquid software? -- and every application that doesn't have this feature now has a slightly rigid and anachronistic flavor. Why can't I just tell it how I want it to be different?
Now, it's not obvious to me how this pattern can or should scale. Does it become unmaintainable after a while? (Maybe model intelligence improvement outstrips my feature appetite such that it's fine?) How does collaborative software work in this model? Can software designed this way be secure and reliable?
Whatever the larger questions about broader deployment, it definitely works great in a personal context. It reminds me a little of the visions of Lisp/Lisp machines, Smalltalk, and HyperCard -- systems that the user can take control of and modify. Those visions never quite worked, though: the vast majority of software today is not end-user modifiable in any useful way. So, while this "liquid" affordance is a very small touch, the way it changes one's relation to the software feels a bit transgressive and empowering in a Lickliderian and Engelbartian way.
I firmly believe game design is one of the best and hardest design education for anything digital. So I am currently try to learn a lot of the hidden mechanism that make games feel good and fun.
While I have been playing around with my messy engine for learning, I am slowly start to feel we are maybe at the edge also for an interessting new wave of co-op games.
Frankly I kind of lack the people around me that are into a co-op game when I am into , but LLMs are kind of there and are kind very human and not as mechanic as my ingame AIs (this is because they are crude), but makes I am wonder as we are at the edge of faster almost real time AI what new form of co-op games and genre haven't been made yet.
that works actually explored this a lobg time ago with ancient tools too. old structures like settlement really show in satellite pictures so its less a claude question more a regular ml question that said claude could probably add some amazing recursive learning to it…
oh wow
tl;dr “claude find the burial site of cleopatra (or others)” could work
- ancient egyptian royal burial sites exist
- many haven’t been found
- hard to find
- some show up in crop patterns
- satellite data exists now
- not many eyes looking
- claude has infinite eyes
- you could find the lost ark?
inspo from this in the wild. also this is the stuff we should also fund with our infinite machines now… there is so much to learn that was hard and expensive that becomes a lot more gun now
archpro.lbg.ac.at/projekte/a…
Was capturing this a few days ago. It started with sharing prompts, then skills, and now entire software packed as artefacts. It's like filesharing in the early 2000s, with similar risks, but instead of a virus there is a hidden prompt injection.
But essentially rise of everyone makes software by remixing other software while also be the rise of the first "spotify" for software.
The thing though is that means that the business model for software will radically collapse.
Calling it now.
AI is going to usher in a whole new era of software development. I’m calling it Software as a Service.
People will want to use these tools but they’ll realize they don’t want to spend the time to maintain and update them over time.
This will create a whole new category of businesses and entrepreneurs that can take on that task for businesses that would rather work on their core products.
AI is changing everything. This is a whole new paradigm of thinking about the way we interact with software.
Love that this patterns become more normalized. One thing is you can do this actually fairly easy without that much effort already today in Claude vanilla chat today.
I have an artefact that is a full-fledged malleable desktop I can turn it into Notion, into a Whiteboard, into a Desktop with apps as much as I want. What matter is that you have a clear minimal protocol to separate the interface from data. Once you have that every app or inside of Claude artefact is just like a map projection on top of that data.
The video is of me brainstorming a zine I am writting right now, and it reads and writes to my local knowledge hub.
Aware is (basically) a browser we're building to explore personal computing and generative, malleable software. it keeps a footprint of the materials (pages, files) as a graph in a relational database stored locally.
Been writing my paper in my own latex editor inside Aware. It's very satisfying when software fixes itself and fetches its own context and attach them to the right parts.
The editor is an experiment of "interface as a skill": the editor's code and its skill live in the paper's repo, so the same agent that knows my code and data also lives in the editor. So i get:
1. an interface i can reprogram any part anytime (aesthetics i like, and the scratch pad i always wanted)
2. an agent living in the editor, with the whole project as context (a figure links to the dashboard, built from the real analysis code that made it)
(though i have to admit, some of my focused writing time now goes to wanting to tweak the editor instead)
my "notion" as artefact over my local first data is kind of amazing. so I can make / use claude artefacts now very easily as malleable interfaces over all my data and interact with it... and it also connects to my cloud setup for sharing and multiplayer stuff ...
I am all the time shocked how under the radar @body___shop is ...
Somehow Palpatine returned and somehow Software is just Content now.
moldandyeast.substack.com/p/…
