antirez - Network

Cool, $200 for a model that can't remotely match $20 Claude Sonnet. This will be a huge hit I guess.

Cool, $200 for a model that can't remotely match $20 Claude Sonnet. This will be a huge hit I guess.

BREAKING I may add SSL to my blog in the next days.

BREAKING I may add SSL to my blog in the next days.

Yesterday I was in a call with my daughter's teachers about an exchange project. They reported they are unable to do exchanges with UK...

Yesterday I was in a call with my daughter's teachers about an exchange project. They reported they are unable to do exchanges with UK because families refuse to send children here in Sicily :( This doesn't happen with France, for instance. They don't understand causes to fix: UK friends, hints?

For many years people said that Hacker News was ruined, full of noobs for the most part, and I kept saying: it's not that bad! Not like...

For many years people said that Hacker News was ruined, full of noobs for the most part, and I kept saying: it's not that bad! Not like the first years, but it's still a cool place. Well now, in 2024: OK YOU WERE RIGHT, HAPPY NOW?

Concurrent programming is hard, but also so oddly satisfying and the design space is larger than expected. Once I'll see this program...

Concurrent programming is hard, but also so oddly satisfying and the design space is larger than expected. Once I'll see this program that I'm writing working very well with read/write locks, optimistic insertion and the like, I'm sure it will be a good few seconds (after which life sucks as usual).

I really hope that the BlueSky firehose will remain open. It's such a great live dataset to do all kind of experiments!

I really hope that the BlueSky firehose will remain open. It's such a great live dataset to do all kind of experiments!

nomic-embed-text via ollama seems to work great for small amounts of text, but the embeddings are apparently a bit less semantically sharp...

nomic-embed-text via ollama seems to work great for small amounts of text, but the embeddings are apparently a bit less semantically sharp with 2000/3000 tokens. Anyone noticed this as well?

Was playing with random projections for dimensionality reduction but had to stop and think: how can today CPUs be *so damn fast*?

Was playing with random projections for dimensionality reduction but had to stop and think: how can today CPUs be *so damn fast*?

Please note that I'm also at @antirez.bsky.social. I see Bluesky gaining traction. I'll post here and there for now, but I hope one...

Please note that I'm also at @antirez.bsky.social. I see Bluesky gaining traction. I'll post here and there for now, but I hope one will win and the community will be back glued together.

HNSW: you write the first implementation and think, wow, that was simple, just following the paper general idea. A few weeks and iterations...

HNSW: you write the first implementation and think, wow, that was simple, just following the paper general idea. A few weeks and iterations later: wow, I had to invent half the algorithm that was missing in the paper.

I pip live my life the --break-system-packages way.

I pip live my life the --break-system-packages way.

My notes about this are here: https://news.ycombinator.com/item?id=42202451

My notes about this are here: https://news.ycombinator.com/item?id=42202451

With Redis 8, there is a return on investing in the community edition (the code you just download from Github), probably what made this...

With Redis 8, there is a return on investing in the community edition (the code you just download from Github), probably what made this possible is the license change. Anyway interesting new features / improvements:

https://redis.io/blog/redis-8-0-m02-the-fastest-redis-ever/

My comment on HN about open hardware in the context of 3D printers. Worth sharing I guess, since this dynamics are going to be central in...

My comment on HN about open hardware in the context of 3D printers. Worth sharing I guess, since this dynamics are going to be central in the future I guess.

So @josefprusa just unveiled a printer that will change the short term story of the 3D printing market, storngly limiting Bambu advantage....

So @josefprusa just unveiled a printer that will change the short term story of the 3D printing market, storngly limiting Bambu advantage. Yet I believe that us loyal MK4 and then MK4s kit buyers (or MK4s directly) with purchase proof, should get kit at lower price.

I love this trick when traversing graphs where I can't visit already visited nodes. At every scanning: graph->epoch++, then as I...

I love this trick when traversing graphs where I can't visit already visited nodes. At every scanning: graph->epoch++, then as I visit a node, node->last_visit = graph->epoch. Much faster than a collection of visited nodes, but threading hostile... Other alternative: bitmap + node_id, but sucks.

Musk didn’t just ruined a social network. The whole IT field lost a fundamental virtual place of information sharing and discussion, and I...

Musk didn’t just ruined a social network. The whole IT field lost a fundamental virtual place of information sharing and discussion, and I am sure this will slow down the field for some time.

In case it's useful for you as well, Word2Vec converted in a sane binary format. https://huggingface.co/antirez/word2vec-simple-parsing

In case it's useful for you as well, Word2Vec converted in a sane binary format. https://huggingface.co/antirez/word2vec-simple-parsing

Fortunately today there are good embedding models with simple APIs (for instance Ollama now supports them), and one can just take a list of...

Fortunately today there are good embedding models with simple APIs (for instance Ollama now supports them), and one can just take a list of words in txt file, process them with the embedding model, and store the result in whatever format.

Word2vec is a freaking list of words associated to a few floats each. It is hard to think at something simpler (regardless of the complexity...

Word2vec is a freaking list of words associated to a few floats each. It is hard to think at something simpler (regardless of the complexity of the vectors training itself), yet go figure what utterly complicated file format became the standard for such a simple thing... :(