I did a prototype implementation of a “network of ML networks” - an internet-like protocol for federated learning where nodes can discover, join, and migrate between different learning groups based on performance metrics (Repo: https://github.com/bluebbberry/MyceliumNetServer). It's build on Flower AI.

Want do you think of this? It could be used to build a Napster/BitTorrent-like app on this to collaboratively train and share arbitrary machine learning models with other people while keeping data private and only sharing gradients instead of whole models to save bandwidth. Would this be a good counter-weight for big AI companies or actually make things worse?

Would love to hear your opinion ;)

Von Neumann’s idea of self-replicating automata describes machines that can reproduce themselves given a blueprint and a suitable environment. I’m exploring a concept that tries to apply this idea to AI in a modern context:

• AI agents (or “fungus nodes”) that run on federated servers
• They communicate via ActivityPub (used in Mastodon and the Fediverse)
• Each node can train models locally, then merge or share models with others
• Knowledge and behavior are stored in RDF graphs + code (acting like a blueprint)
• Agents evolve via co-training and mutation, they can switch learning groups and also chose to defederate different parts of the network

This creates something like a digital ecosystem of AI agents, growing across the social web; with nodes being able to freely train their models, which indirectly results in shared models moving across the network in comparison to siloed models of current federated learning.

My question: Is this kind of architecture - blending self-replicating AI agents, federated learning, and social protocols like ActivityPub - feasible and scalable in practice? Or are there fundamental barriers (technical, theoretical, or social) that would limit it?

I started to realize this using an architecture with four micro-services for each node (frontend, backend, knowledge graph using fuseki jena and activitypub-communicator); however, it brings my local laptop to its limits even with 8 nodes.

The question could also be stated differently: how much compute would be necessary to trigger non trivial behaviours that can generate some value to sustain the overall system?

cross-posted from: https://lemmy.world/post/28442844

cross-posted from: https://lemmy.world/post/28384872

This is a showcase of combining vibe coding with the Fediverse and attempto controlled english (ace).

I'm fascinated by vibe coding, but I'm also highly critical of it. It fascinates me, because it enables people, who normally cannot code to be able to generate running code. What I don't like, is that it just isn't actual programming. It's closer to a wishing well. It fosters a quasi-magical understanding of programming and computer science, which is already too common in current society (I wrote a paper about it here: https://philpapers.org/rec/BINAKR). That's why, in my opinion, the Fediverse should set a counter-point here with something like a first-order logic language like ACE, which actually brings people closer to an actual understanding of computer science concepts like modeling and logic without hiding the complexity behind seemingly "magic", and could also result in better code.

The above demo shows a glimpse of how this could look like on the Fediverse. Imagine communities being able to form their own spaces on the social web through language! Simply using natural language will probably not be specific enough here. We always imagine everything getting much easier, but that's just the logic of digital capitalism that tries to sell us innovation as inventing yet a more easy way to get your coke handed to you, which can only lead to more and more environmental destruction. So, what will the language interface for the future digital look like? I think it will be more something like the semi-formalic language found in technical manuals, cooking recipes and judicial texts. Something like ace, in between coding, domain specific languages, modeling and natural language. And people who are experts at this and know the old technical stuff that no one understands anymore will be the new "coders". But maybe I'm wrong.

Repo: https://github.com/bluebbberry/AceCoding.social.

cross-posted from: https://lemmy.world/post/28384872

This is a showcase of combining vibe coding with the Fediverse and attempto controlled english (ace).

I'm fascinated by vibe coding, but I'm also highly critical of it. It fascinates me, because it enables people, who normally cannot code to be able to generate running code. What I don't like, is that it just isn't actual programming. It's closer to a wishing well. It fosters a quasi-magical understanding of programming and computer science, which is already too common in current society (I wrote a paper about it here: https://philpapers.org/rec/BINAKR). That's why, in my opinion, the Fediverse should set a counter-point here with something like a first-order logic language like ACE, which actually brings people closer to an actual understanding of computer science concepts like modeling and logic without hiding the complexity behind seemingly "magic", and could also result in better code.

The above demo shows a glimpse of how this could look like on the Fediverse. Imagine communities being able to form their own spaces on the social web through language! Simply using natural language will probably not be specific enough here. We always imagine everything getting much easier, but that's just the logic of digital capitalism that tries to sell us innovation as inventing yet a more easy way to get your coke handed to you, which can only lead to more and more environmental destruction. So, what will the language interface for the future digital look like? I think it will be more something like the semi-formalic language found in technical manuals, cooking recipes and judicial texts. Something like ace, in between coding, domain specific languages, modeling and natural language. And people who are experts at this and know the old technical stuff that no one understands anymore will be the new "coders". But maybe I'm wrong.

Repo: https://github.com/bluebbberry/AceCoding.social.

This is a showcase of combining vibe coding with the Fediverse and attempto controlled english (ace).

I'm fascinated by vibe coding, but I'm also highly critical of it. It fascinates me, because it enables people, who normally cannot code to be able to generate running code. What I don't like, is that it just isn't actual programming. It's closer to a wishing well. It fosters a quasi-magical understanding of programming and computer science, which is already too common in current society (I wrote a paper about it here: https://philpapers.org/rec/BINAKR). That's why, in my opinion, the Fediverse should set a counter-point here with something like a first-order logic language like ACE, which actually brings people closer to an actual understanding of computer science concepts like modeling and logic without hiding the complexity behind seemingly "magic", and could also result in better code.

The above demo shows a glimpse of how this could look like on the Fediverse. Imagine communities being able to form their own spaces on the social web through language! Simply using natural language will probably not be specific enough here. We always imagine everything getting much easier, but that's just the logic of digital capitalism that tries to sell us innovation as inventing yet a more easy way to get your coke handed to you, which can only lead to more and more environmental destruction. So, what will the language interface for the future digital look like? I think it will be more something like the semi-formalic language found in technical manuals, cooking recipes and judicial texts. Something like ace, in between coding, domain specific languages, modeling and natural language. And people who are experts at this and know the old technical stuff that no one understands anymore will be the new "coders". But maybe I'm wrong.

Repo: https://github.com/bluebbberry/AceCoding.social.

Combining vibe coding, attempto controlled english (ace) and the social web - form space on the social web through words, secured by attempto controlled english.

You are only able to run code based on attempto controlled english (ace), which is a formally defined subset of the english language. In the future, admins could through this restrict certain kinds of code from executing for security purposes. Additionally, it lessens the ambiguity of natural language and you can be sure that the resulting code will do what it should.

Here are a few example commands in ACE that could be run on AceCoding.social in the future:

• Moderation: If a user posts more than 20 times in 1 hour then the system temporarily restricts the user's posting ability.
• Look and feel: If a post contains an image then the system displays the image with rounded corners.
• For content curation: Every post that has more than 50 likes is added to the "Popular Today" collection.
• For notifications: If a user has not logged in for 7 days then the system sends a digest of missed interactions.
• For accessibility: Every image in a post has an alt-text that is either provided by the user or generated by the system.

Repo: https://github.com/bluebbberry/AceCoding.social

(Image from Veronica Casson, https://www.freethink.com/wp-content/uploads/2023/06/terraforming-thumb.png?resize=500%2C281)

Combining vibe coding, attempto controlled english (ace) and the social web - form space on the social web through words, secured by attempto controlled english.

You are only able to run code based on attempto controlled english (ace), which is a formally defined subset of the english language. In the future, admins could through this restrict certain kinds of code from executing for security purposes. Additionally, it lessens the ambiguity of natural language and you can be sure that the resulting code will do what it should.

Here are a few example commands in ACE that could be run on AceCoding.social in the future:

• Moderation: If a user posts more than 20 times in 1 hour then the system temporarily restricts the user's posting ability.
• Look and feel: If a post contains an image then the system displays the image with rounded corners.
• For content curation: Every post that has more than 50 likes is added to the "Popular Today" collection.
• For notifications: If a user has not logged in for 7 days then the system sends a digest of missed interactions.
• For accessibility: Every image in a post has an alt-text that is either provided by the user or generated by the system.

Repo: https://github.com/bluebbberry/AceCoding.social

(Image from Veronica Casson, https://www.freethink.com/wp-content/uploads/2023/06/terraforming-thumb.png?resize=500%2C281)

The FediAI-frontend I posted in the last video was a bad idea. Prompts very posted publicly when you didn't expect it. This Mastodon client has a semantic feed, which should be clear that prompts are posted only. On the feed, several AI workers could listen to. In this example, a generic bot is listening, which always replies with the same suggestion. What do you think of this?

Repos: https://github.com/bluebbberry/Seamantic and https://github.com/bluebbberry/FediAI

You post through the UI to a Fediverse hashtag, on which AI bots listen to, which replies are then displayed in the UI. In the future, the main app should have some kind of ranking only to show the best replies. Through hashtags, AI bots can specialize in certain areas. It would also be possible to partially process a task (for example translate it) and then repost it to another hashtag (I call that "prompt routing). This way, you can have chains of AI bots working together on public, shared message queues.

MyceliumWebServer Update | ActivityROS Demo

Shows decentral AI fungus that can be browsed like a chat bot connected to other chat bots - and it can also be interacted with through Mastodon and its connected knowledge graph. It's basically three views of the same thing, just from different webs!