I’m considering implementing additional anti-spam measures against synthetic and manipulated engagement. There has been an uptick of accounts paying for engagement or using bots to create misleading social proof.

I should state that Linux overreports security vulnerabilities. Anything that could *theoretically* cause the system to crash for example is included, or a UAF which is not very exploitable. Only a subset of these vulnerabilities are exploitable in production systems, and much less have any confidentiality/integrity impact, and even then you usually need to have user access to the server. View quoted note →

Apparently, if you run GPT 6.1 Sol for long enough, its CoT starts getting weirdly dense: RL does things

GM nostr

"Several vulnerabilities have been discovered in the Linux kernel that may lead to a privilege escalation, denial of service or information leaks. [...] We recommend that you upgrade your linux packages." List of CVEs:

Debian alert DSA-6528-1 (kernel) [LWN.net]

Why are we making the user experience more complicated for everyone?!

ga nostriches 🌍🚀🍍👁️

Why can’t Claude models stop speaking in “sounding smart but isn’t” slop mode

Good morning Nostriches.

It seems that all the developers are missing the big picture on Nostr. Everyone doesn’t want a shiny app with everything or AI AI AI AI. It needs to just work. Hmm 🤔

This is a good time to start using nostr.land. The aggregator relay combines events from hundreds of relays, removing spam while ensuring you never miss a note. The Event Cannon publishes your notes to hundreds of relays, even the small ones and the ones you don’t use. And you get instant backups of your data as a portable .jsonl file, anytime :) Only for 5K sat/month:

nostr.land

nostr.land

Welcome to the next generation of paid relays.

Passkeys are great, actually

What are your complaints with Nostr? Also please specify clients and platforms

Good software is not easy to build.

Unmaintainable mountains of slop