This paper presents our research towards a near-term future in which legal entities, such as individuals and organisations can entrust semi-autonomous AI-driven agents to carry out online interactions on their behalf.
https://arxiv.org/pdf/2409.04465
Alchemy is great. It essentially is trasforming from a lower state to a higher state. Bitcoin can be considered an example, Satoshi transfomred inert hard drives into precious numbers, and digital gold, via proof-of-work. Jung and Newton were also great alchemists.
I'm studying one of the great unsolved alchemical texts, which says something similar, and also about language, mathematics and the geometry of conic sections:
https://en.wikipedia.org/wiki/Monas_Hieroglyphica
"Micro$oft BitNet a blazing-fast 1-bit LLM inference framework that runs directly on CPUs.
You can now run 100B parameter models on local devices with up to 6x speed improvements and 82% less energy consumption—all without a GPU!"
https://github.com/microsoft/BitNet
in this case, but anyone can host videoes and playlists anywhere
and anyone can run a server
#solid doesnt have nodes, just a normal web server with a few modifications or upgrades ... this is completely compatible with the regular web and regular cloud
Yes, I like it. Very much user controlled "small data" setup. 100% contrast to the web giants "big data" approach. Not sure anyone other that me will want it, but Im glad to have it.
Regarding selling, because #solid has access control built in, what you can do is ask for a payment. And when that payment is recieved you can allow access to the content for a specific npub.
They are not quite web sites, they are web servers. Most relay are web servers without a website. Its a bit like using a car radio. Listening to some music, then walking home.
You know he means it. Jack is a great CEO.
nostr:nevent1qvzqqqqqqypzp382htsmu08k277ps40wqhnfm60st89h5pvjyutghq9cjasuh38qqy2hwumn8ghj7erfw36x7tnsw43z7un9d3shjqpqwq2fqtslazua4caxfllvv7qdlrp8zlssuqefrw7p8svfqh6w59vsuwwydt
Good call. Pretty much every sentence in that abstract is propaganda and/or wrong.
However bitcoin is not a single purpose thing. It has a few different properties and functions. And it was designed in that way, not knowing which one(s) would take off. Now we know more.
An important missing column is how easy it is to identify when people are using these chats
nostr:nevent1qvzqqqqqqypzqhjlc9p5ey5teh96d7qpskw4ywp5zzfjjxvql5mwxwm5zcun6k3vqy2hwumn8ghj7erfw36x7tnsw43z7un9d3shjqpqhk3605cnzkmpaalrm9mhnazyy4ls6n6jdm3kd2sqxvqezx9hekzsevflly
#sandymount is the first truly permissionless dev environment for both humans and machines. Powered by #solid for personal storage, and #nostr for identity and realtime updates. I wake up in the morning with the idea for a music player. First the machines design me a user interface. Next they create a data structure to use. After that build the interface together. Import your songs. And start listening. Future ideas can be user to user songs, and pay to listen.
https://image.nostr.build/a68d8f1ca6935bc9283c237f3b938b7d66534562a43372e16c97800bd8e7ddf4.jpg
They are just web servers. you have to run a web server first, to run a websocket server.
There are millions more web server that can pop up on clearnet or onion because its easier, and harder to track
The opposite is true. Mastodon network is growing (new servers popping up) and nostr network is shrinking, due to neglect.
Nostr relays do NOT popup when needed. They just die as they are not supported. And we move ever closer to a model of a few central relays.
Mastodon's problem is they are tied too much to one domain. They can move domains but it's hard. What they could to do is add a nostr key to their profile. Just like Nostr added NIP-05 to their profiles. Then users got best of BOTH worlds.
#ditto is doing this.
Nostr is like a car radio.
People go to sit in a car and listen to music.
They dont even know (yet) that the car can turn move or drive.
Because they are happy listening to the music.
Martin Kleppmann is very good at this stuff. He also helped design bluesky. I honestly thought nostr was better than bluesky at the start. But they have worked really hard to improve every aspect.
#sandymount after 1 week has exceeded all my expectations. Technologies that work well on their own, explode in utility when combined together. Each tech has its advantages and limitations. But together they remove the weaknesses from each other, without compromising the benefits. #happydays
https://sandy-mount.com/
Main issue is the worse clients hold back the better ones. Well, we expected this. What I didnt expect, was that the worse clients have the louder voices then the better ones. Lots to observe and learn from.
Searchlores is perhaps the best website I ever read. Fravia+ was one of the greatest, if not the greatest, of the cypherpunks. Left us far too early. When I read his essays today, and compare it to the fake cypherpunks of today, I still feel nostalgic. Mirror it wide and do the world a favour.
https://github.com/opsxcq/mirror-fravia
Well, this was unexpected. Yesterday with #sandymount I gave away 2 million testnet sats by giving the utxo and privkey in a capability URI.
Instead of someone taking the 2 million sats, two people actually used the fuacet, and solved a capcha, and ADDED to the balance!
Now there are 2077311 testnet sats still unclaimed!
https://image.nostr.build/d87cae904670344ade082b16599f61a203558aa94cdec7a7a08dfcdfbb1635d0.jpg
#bitcoin doesn't need a soft fork. Any soft fork risks splitting into two chains, both of which could eventually fail. The current features of Bitcoin are more than enough and barely fully utilized. Be cautious of anyone pushing for a soft fork—there are adversaries out there
Nostr relays are web servers. The only reason they work is because they are web servers. Most are written from scratch, which means that 99% of the features of the web are not implemented. Reversing that, it means that relays can have 100x more features, that have been developed over the last decades. Once you realize the relays are just web servers, there's no nostr "winning" or "losing", the web has already won. We're just making it better.
The very best coding models and workflows, right now are roughly Satoshi level intelligence. Imagine what could be built with a new Satoshi, or 2, or 5!
Quite a few out there, and there's ever moving league tables. The prompts are quite important. Also everyone has a different workflow. I'm using nostr and personal storage for mine.
Cmon. We both know "centralized" is a spectrum. Nostr dev is pretty darn centralized rn. As is the relay network. Can we not just say the good things they are doing such as good UX, good onboarding, good infrastructure, good devX and try and improve nostr. Instead of pretending nostr can be better without putting in the work?
It's easier for new users. So they are growing and retaining users. For example bluesky started off smaller than nostr. Today they got more users in 1 hour then nostr did in it's entire 4 years. They also have good infrastructure. Many more things can be learnt.
its very good for development of new ideas, like smart contracts
and the coins can make a solution you dont need to get regulatory approval
not a lot of hashing is burnt on testnet4 because the difficulty reduces to 1 after 20 minutes
You can actually log in to nostr with your PUBLIC key. Then you can post stuff, use apps etc. But so can everyone else. This might be completely useless. Or there might be some use cases there with anonymity and plausible deniablity.
People ask me why I am so bullish on #ditto?
Simple: it's the next evolution of relays on Nostr. As one of the first daily users and contributors to Nostr, I’ve seen every early relay – they were barebones, just enough to get things going. And sure, they did their job, but they were like turning on a web server with most of the power features, like scaling, storage, and caching, switched off.
Ditto changes that. It’s not just a relay—it’s a super relay. Imagine if the full potential of the web met the full potential of Nostr. That’s Ditto. The team behind it already knows how to scale to millions of users, and their design is built to handle that scale. On top of that, Ditto’s interface is probably best-in-class UX, right now.
Super relays like Ditto aren’t just keeping pace with the growing demands—they’re quietly taking leaps forward, pushing boundaries, and driving the network to evolve.
It's pretty hard to measure temperature. Even with the most modern readings, because most of the earth is water, and we cant measure the deep ocean. The Argo network is under funded. Satellite data from the polls can vary. That's even with modern space technology. Before that we had proxy methods such as tree rings and ice cores which are even more sketcy.
But you can safely say that CO2 is a greenhouse gas, that traps heat. And since energy cant be created of destroyed acc. to the well-tested 2nd law of thermodynamics, you can assume that the heat goes somewhere. And that C02 has a log effect on temperature, with possible feedback mechanisms, mainly water vapour.
Figuring out temperature involves a lot more detective work that people realize. And scientists dont want to admit that, because it's red meat for the skeptics. Then there is the sensationalism in the media which doesnt help.
We probably have warmed about 1.5C in 150 years and c02 probably is a greenhouse gas that traps heat, on land, but in also some strange cycles in the ocean. There is an awful lot of detective work that is needed to come to that conclusion, but it's prolly just about right.
Whether this is catestrophic is anyone's guess. Probably not, but those wanting to be cautious have a point. Using it as a tax strategy is non-sensical, tho. If humans were able to solve simple problems, climate change might be a good one to tackle. But we are barely able to stop ourselves nuking each other, in which case we'll need all the heat we can get. Maybe we'll get smarter when the machines help us think for ourselves...
Oddbean new post about login | logout
Notes by melvincarvalho | export