Exactly like humans.

Now, Claude is down. I wouldn't be surprised if it was because we people who have both tried offloading their work to it.

How is Hermes different in that regard?

Also note, that RAG and even vector search are more paths of the early days that didn't prove too valuable. Just let your agent search it directly and optionally create an index as a default entrypoint for common topics.

Fair points. Agreed.

If my coworker was part of a clone series of 100 million units, requesting a character sheet would be pretty reasonable

These are $1 Trillion dollar companies that can't produce explicit details on how their products work? It's nonsense.

I think if they could explain how they work, their strengths and weaknesses, they would reveal to the world whose data they've been appropriating.

That's another thing altogether. They can characterize the behaviour without quite giving up who and where the data comes from.

Admittedly, yes, there's some overlap there.

They would have to admit 'seen it in the training data' as a factor, and that opens a can of worms.

Maybe it is just me, but by far most of the time, when I want to archive something from the internet, it is information and information is best served in an absolutely minimal text format like html or md.

You make it sound as if "coding" was a distinct thing with clear boundaries in the technical world. But this critically misses the fact that coding agents dramatically lowered the barrier to controlling everything with a microchip. The only thing that exists "outside [the reach] of coding agents" is purely the analog world and that boundary will get fuzzier than it is perceived to be.

It seems like many people have the same or similar ideas. I was thinking of using a tool similar to bookmark-managers as the foundation of a new web. Where you subscribe to RSS-feeds of specific (or clusters of) people to specific topics as the "follow" primitive and you publish your own feed(s), which bookmark-managers btw. already allow. The missing pieces are commenting on the feeds of friends and a layer of federated ML for ranking, which the user controls by simple sliders that set the mark for dimensions like retrieval-vs-discovery, hightrust-vs-highnovelty, recency-vs-trendingimpetus and so on.

That's like saying Spotify wouldn't be a good example, because it outsourced music production to musicians and just not a meaningful point.

I thought that early coding assistants came to be written in some Java/TypeScript, because AI companies just had web-devs playing around and then made it a product even though the languages being such a misfit for terminal. Why did you decide for TypeScript?