Your health data is the most personal information you'll ever have, yet it's scattered across systems that were never built for you.
I believe that should be the other way round. Your data should live with you, under your control, so you decide how it is used. Any AI that accesses it should work for you and you alone.
Self-sovereignty is a principle I believe in, and it's why I am proud of what we are building at GetMine.
Mike retweeted
Watch Clojure: the documentary here!
youtube.com/watch?v=Y24vK_QD…
It's an over an hour long documentary about ... #clojure!
I'm apparently ranked #1 on the GitHub Commits Leaderboard :-) ghcommits.com/u/mikera
Been going over ten years now, and @TweeGeeMee still comes up with some gems. Though I think this one is better suited for Halloween than Valentine's week @RogerAllen ?
Anyway, some of us were doing generative AI long before it was mainstream :-)
There’s a lot of discussion right now about context graphs and AI systems of record. Context is important. But in multi-agent systems, execution fails long before representation does.
If agents are going to operate across real systems, we need ordering, authority, and runtime guarantees, not just better memory.
I wrote about where I think the stack is converging.
Would love feedback from folks thinking about context and system-of-record layers.
open.substack.com/pub/chirde…
#multiagent #EnterpriseAI #covia_ai
Well... I built a thing.
Managed to write a working mini-RDBMS running on the @convex_world lattice in less than a day using @ApacheCalcite
Here's it pretending to be @postgres via @beekeeperdata
Feels like we’re early in defining the infrastructure for this layer. Curious to compare notes with others building for multi-agent coordination and shared context, the shape of this stack matters (esp to @mikera and I as we have been at it for a while :) )
Mike retweeted
Live coding a full game level using only the REPL. No restarts, no reloads, just pure flow.
This is how I build levels while the game is running.
If you like game dev, Lisp, or live workflows, this is for you 👇
Mike retweeted
Check out my latest article: Context Graphs Aren’t Enough: Why Multi-Agent AI Needs an Execution Substrate linkedin.com/pulse/context-g… via @LinkedIn
@ashugarg @JayaGup10 - thanks for the inspiration
#AI #multiagentsystems
This is a scandal if true.
Honest citizens have been told they are doing their bit by recycling, but they could actually be making things worse?
Are any UK politicians doing anything? If not, why not?
@Conservatives @reformparty_uk @UKLabour @TheGreenParty
yes, asian and african rivers produce 95% of ocean plastic but nobody asks where the plastic comes from.
plastic recycling is a scam. always has been.
the industry knew since 1974 it “cant be justified economically”
they funded the recycling campaigns anyway because the alternative was banning plastic.
only 5-6% of US plastic actually gets recycled and europe isnt better.
the EU exports 1.1 million tonnes of plastic waste per year, 3 million kg leaving every single day.
31% goes to turkey, 16% to malaysia, 13% to indonesia, all labeled “recyclable”
most plastics cant even be recycled to begin with (thousands of types, different chemical properties). and for the few that can, the output is lower quality and more toxic than virgin plastic.
you literally degrade the material each cycle until its worthless.
so what happens to the other 95%?
western countries ship it to southeast asia and africa under the label “recycling exports” with the receiving countries promising to recycle it in their stead for a price, but those countries dont have facilities either. so they burn it or dump it in rivers.
every 20 minutes, a 10-tonne truckload of plastic enters the ocean in indonesia alone.
it may look like indonesians pollute more but that’s because we ship them our garbage and they have no infrastructure to handle it.
china used to absorb it all. when they banned imports in 2018, the west just redirected to countries with weaker regulations. malaysia, vietnam, philippines. the map below shows the result.
you sort your trash so you feel like youre helping, the plastic goes on a boat, gets burned in a village without emissions controls, poisons their air and water, flows into their rivers, enters the ocean, and in your food and water.
you feel good about the plastic you recycled and never think about it again but you end up eating it anyway.
then we ban straws and plastic bags and call it environmentalism
If it is truly ethical, Artificial Super Intelligence will be kept quite busy prosecuting ecocide cases in the near future.
World leaders might want to think about that?
Indonesia's military is RAZING 2 MILLION hectares of Papua's irreplaceable rainforests—the LARGEST deforestation project on Earth—for a greedy sugarcane mega-farm, dooming 42 dazzling Birds of Paradise species (with 1 in 5 already on the brink of EXTINCTION) and wiping out biodiversity hotspots FOREVER! 🐦🔥
Ignoring Sumatra's nightmare: Over 1,500 DEAD in Asia's deforestation-supercharged floods and landslides, with Sumatra losing 4.4 MILLION hectares of forest since 2001—turning rain into RIVERS OF DEATH! 🌧️☠️
This eco-genocide fuels climate chaos—RT if you're FURIOUS and demand a HALT! #PapuaForestMassacre #DeforestationDisaster #BirdsOfParadiseInPeril
ft.com/content/27a6bd…
It's dumb to move all your codebase to a single language. There is no "one size fits all".
Confident Prediction (99%): It will turn out that there are languages *far* better than Rust in the future, particularly with respect to AI's ability to write good code in them.
(A strong candidate would be a Lisp with a TypeScript-style optional type system)
wow. Microsoft wants to use AI to wipe out all C and C++ code by 2030 and replace everything with Rust.
Their new “North Star” metric: 1 engineer, 1 month, 1 million lines of code.
Microsoft Distinguished Engineer Galen Hunt writes in a post on LinkedIn.
“My goal is to eliminate every line of C and C++ from Microsoft by 2030. Our strategy is to combine AI and Algorithms to rewrite Microsoft’s largest codebases.
Our North Star is ‘1 engineer, 1 month, 1 million lines of code.’ To accomplish this previously unimaginable task, we’ve built a powerful code processing infrastructure.
Our algorithmic infrastructure creates a scalable graph over source code at scale.
Our AI processing infrastructure then enables us to apply AI agents, guided by algorithms, to make code modifications at scale. The core of this infrastructure is already operating at scale on problems such as code understanding.”
---
Source Linkedin post
linkedin .com/posts/galenh_principal-software-engineer-coreai-microsoft-activity-7407863239289729024-WTzf/
Hugely important topic.
Most importantly, we need open / decentralised economic systems to ensure all the upside isn't captured by a few billionaires with monopolies.
The economic benefits of advanced AI risk being captured by an elite few, while the costs fall on workers, communities, and countries.
Windfall exists to change that. Our mission is to ensure AI prosperity is shared by all, not captured by a few.
windfalltrust.org
The Lattice is the ultimate system of record.
Decentralised. Unlimited scale. Self-healing. Cryptographically verified.
@convex_world
















