Tootfinder

from my link log —
Don't make my mistakes: common infrastructure errors i've made.
https://matduggan.com/mistakes/
saved 2021-12-03 https://dota…

Another tale of the closed web: a common thing on Instagram is that it doesn't send out Co-Post requests when you do it too often or they don't like you. It's one of the aspects or levels of being “shadowbanned”.
I often work with various kinds of accounts to publish my essays on Instagram to get more exposure. For example, pages dedicated to criticizing the police (useless.cops) or accounts showcasing ICE activity in Minnesota (MNicewatch) for my essay on Renée Good…

> Respondenci podkreślali, że szczególne znaczenie mają nieformalne kontakty z wójtami, burmistrzami i prezydentami miast, a także relacje towarzyskie
Od @…
https://

Läsvärd text i Flamman om maktkampen i Byggnads (även om skribenten är tydligt på ena sidan i konflikten). https://www.flamman.se/en-strid-om-arbetarrorelsens-sjal/

En strid om arbetarrörelsens själ
Linn Svansbo: På helgens kongress avgörs vad ett fackförbund får vara

It's an absolutely beautiful Sunday. Perfect temperature, no mosquitoes. Yet I see and hear zero kids outside. I live across from a park and it's completely empty. Likely the same for the whole day until the ice cream truck melody lures them to the street.

TOPLAP is the *Temporary* Organisation for Live Algorithm Programming, but is 22 years old and still going. I think it was actually disbanded at some point over 10 years ago but it carried on by mistake. The problem with vaguely anti-hierarchical and distributed organisations is that they're quite difficult to disband.

How does an LLM know the difference between its own thoughts
and someone else's words?
https://role-confusion.github.io/

Prompt Injection as Role Confusion
LLMs can't tell who's speaking. We show they identify roles by writing style, not tags, and exploit this with CoT Forgery, injecting fake reasoning that models mistake for their own thoughts.

Boston Consulting Group study suggests 'embedding AI agents into formal organizational roles can reduce managerial oversight in AI-mediated work, and should be understood as a governance decision rather than a mere labeling choice' - i.e. more mistakes get through https://www.emmawiles.com/storage/ai_emplo

As I edit my writing, I've found it useful to have different layouts to catch mistakes. Earlier this year I moved to Obsidian for my a writing platform since I can tweak it with Claude or Codex. Here's an example of such a layout I created. It's in HTML, so I can't edit it (... for now).

OMG, I'm having a laugh with Opus 4.8. They trained it to be more honest, but now it's like Jim Carrey in Liar Liar. Earlier it was like "I accidentally deleted all the work I just did, but it's okay - I fixed it." And now it's admitting every little mistake it makes. It's hilarious.