infosex.exchange <3

You are probably looking for the infosec.exchange Mastodon instance

This host is mostly for my random stuff, and in little part acts like a well-intentioned placeholder for the typosquatted domain.

Discoverability and Archiving

Currently I'm using this host for saving the items from my own feeds to the Wayback Machine and provide in-links for search engines. I hate that I have to do this, but the non-sense ideology of Mastodon pretty much ruined the search feature for Fediverse as a whole, and this wasn't changed by the fact that they owned their mistake and implemented search eventually.

Yes, I (or anyone else) could do similar things with other peoples published feeds, regardless of the tantrum. No, you can't defederate this, because the process doesn't rely on an instance.

Gluttony Section for Search Engines

Fair warning:

If you are a person involved in creating splash windows for first browser startup and we ever meet in person, I will hurt you.
this post | permalink
@stf The main issue that I don't plan to move to Emacs, although I'm sure org-mode can do literally *everything* (khm..Unix philosophy...khm) :D
this post | permalink
@brouhaha This sounds strangely similar to the no-search fedi crowd...
this post | permalink
[RSS] unpacking iDRAC9/iDRAC10

https://trouble.org/?p=1467
this post | permalink
@stf Several editors highlight "TODO" specifically, but I want a way to format custom sets of markers, possibly depending on the extension/project I'm working on. Another example is marking findings in my notes with different severity labels.
this post | permalink
@stf How does org-mode help with customizing highlighters?
this post | permalink
@paniash @tarsius Good point, a simple regex-based highlighter would probably cover most of the use cases!
this post | permalink
@eingemaischt right??
this post | permalink
This seem like a systemic reverse-centaur: you can't build "AI-driven inspection systems" with general-purpose LLMs, you can only use LLMs then validate *their* output with other, reliable[1] methods. Whoever decided to use LLMs like that is unfit for the job.

[1] ML-based image classification, trained on a very narrow set of labeling data can be more reliable than a human. Try to achieve the same numbers with LLMs, I'll wait!

RE: https://aus.social/@perkinsy/116825631938165912
this post | permalink
The Deadliest Design Mistakes in History

https://www.youtube.com/watch?v=uFbxUPbZovM
this post | permalink
Next Page