Hi friends,
The http://alt-text.org alt text library project needs a new leader, I have brain cancer.
I built a working, scalable, proof of concept library of shared alt text with fuzzy matching.
I'd like to connect with the #accessibility dev community. I want to hand the project off to a team or a leader if anyone is willing to take it over.
Github: https://github.com/alt-text-org
WIP MVP: a site designed for writing alt text with a private library: https://my.alt-text.org
Boosts appreciated
Wondering why I urgently need to retire my MIPS-based Debian box, it turns out to be the fault of Rust... and LLVM, and Debian. Sort of, anyway: https://lists.debian.org/debian-devel/2019/08/msg00201.html
I understand the local decisions of each project. But the emergent outcome is nuts: that whole classes of hardware are deemed obsolete (in some specific way) just because optimising compilers -- you know, the things that are supposed to help make software faster/smaller -- have themselves got too huge.
Today, we filed our final brief in Hachette v. Internet Archive, the publishers’ lawsuit against our library. For four years we've been fighting for library rights—what our founder, @brewsterkahle, calls “a battle for the soul of libraries in the digital age.” https://blog.archive.org/2024/04/19/internet-archive-stands-firm-on-library-digital-rights-in-final-brief-of-hachette-v-internet-archive-lawsuit/
The death of physical media gives me so many mixed feelings.
On the one hand, it's kinda wasteful isn't it. All those plastic boxes.
On the other: you don't know when digital will just be taken away from you. And at some point it probably will be taken away from you somehow.
Makes me nostalgic for the times opening presents when younger, and even only a few years ago, and it was some cool game or Bluray/DVD.
If you are a qualitative researcher and you put data into any LLM, but especially Chat GPT to "help" with analysis, you are being conned and violating privacy and data protection ethics writ large.
If you are REVIEWING qualitative research and the authors discuss using chatGPT in their methods and you think this is somehow more rigorous and reliable than traditional inductive methods then you are not qualified to review that or any other qualitative study.
If you are an IRB staffer and you approve studies which will put participant data into chatGPT or any other LLM that is not completely internally controlled and disconnected from cloud services. WHAT THE FUCK.
This is journalism. They think lasers work like in science fiction, where you shoot a beam, and there is like this small beam ejected that flies towards the target at slow speed.
RE: https://wetdry.world/users/avesbury_rosetta/statuses/112259750743540070
A reminder that on Monday, when the Solar Eclipse occurs, there will be earthquakes.
It's not because of the eclipse. It's because today alone there were almost 300 earthquakes. And last year there were an average of 2,000 earthquakes a week. They are pretty common. We just don't notice them because they don't always occur near us.
But when they happen during significant events, it presents an excellent money making opportunity for people with no morals. So there's scary rumors about earthquakes on Monday. And there will be. Maybe there will be 300 earthquakes. Maybe some will be significant. It most likely has nothing to do with the eclipse, because it happens daily, with and without an eclipse.
TLDR: earthquakes are common. And tying it to a celestial event is a great opportunity for charlatans.