tired_fedora

Not The Onion?

I wonder if a strict divide into planner and worker agents really is the best design for swarms. In the spirit of stigmergy, swarm intelligence emerges from every agent performing both of these roles according to their current context. I also wonder how much more performance could be gained by updating model weights as the swarm churns, either of every agent on their own or of every agent class. This would be the truer representation of learned behavior than the field journal described in this article, and it would also implement lessons from context drift and ‘LLM sleep’ from earlier this year. Impressive publication, nonetheless.

It really is :D ! Source: https://en.wikipedia.org/wiki/Antarctic_English

Fully agree. That said, the Antarctic research stations do have their own subculture and their own accent, so that much will likely occur on a lunar base, too.

The EU will need to form a proper federal state or be torn apart in the persistent attacks of the other major world powers that actively try to prevent just that unification.

5 lines in and it reads like slop.

I believe they are just running some sort of network detection. I also got this message yesterday, but when I switched servers in my VPN it went away. Sad to see reddit go down this path. But that’s why I’m here.

I’m doing my part 😁

Absolutely delightful article about how limited our digital color space is, the history, technology, and physics behind that, and vivid examples where to encounter those exiled colors in our physical environment. Very approachable language.

Can we normalize tldrs under bot posts please?

If Chromium becomes incompatible with privacy, the only real and broadly accepted alternative is FireFox. Which implementation, and as always in these kinds of discussions, that depends on your threat model: On desktop, I am very happy with LibreWolf. Mullvad Browser is also great, especially with Mullvad VPN, though it breaks pages a little more often than LibreWolf. On Android, I am quite happy with IronFox.

This article criticizing AI gives me strong LLM vibes. Especially this sentence, which addresses the reader directly, as if in response to a prompt: “The split you describe is real, but it’s not new, and it’s not caused by #AI, it’s the endpoint of a long enclosure of commons → platform capture (#dotcons), trust → contracts, sharing → surveillance + monetisation and public space → login walls.” by Hamish Campbell on OMN (or is it?)

I often feel a little ‘legislative paralysis’. On the one hand, I want as little government interference in the free web as possible. On the other hand we can see first hand that web anarchy collapses into web oligarchy. I guess the EU is demonstrating that targeted legislation, like one click unsubscribe or one click cookie denial, can improve the web experience and privacy even beyond their borders. Baby steps… When do we get one click delete all my data? And when does a single page start caring whether my browser sends a Do not track request or not? Until then, it’s back to private privacy measures… Even if that’s an uphill battle.