Daeren

So a major battleground right now for the right wing anti-LGBT moral panic is public libraries. Public libraries are governed by boards, similar to the school board. These positions are often entirely ignored in local jurisdictions so basically anyone can get appointed to them just because they're willing to do it. There has been a coordinated effort in many jurisdictions by conservatives take over library boards.

When conservatives take over a library board, they try to "deprofessionalize" the library. Which is to say, to take power away from librarians and library workers and instead micromanage the manage through the board to push their right wing agendas and hamper the efforts of librarians to serve the community and create safe places for vulnerable communities.

BookRiot used FOIA to obtain tons of emails and documents from the Elmwood Park, IL public library detailing how a right wing board worked to ensure that "the complexion of the library doesn't change." Elmwood Park has been quickly becoming one of the most diverse communities in IL and the old guard wants to use the library to keep it white.

The job of the library board should be to hire a good well seasoned professional librarian as the director and do performance reviews of the director, while advocating for the library to city council or other funding bodies. But the director should be a librarian. Librarians and other library workers such as library assistants know what's best for the library because they work in it every single day.

If you live in some random suburb somewhere and you want an easy way to position yourself to do good: Join the local library board and be a left wing voice that can protect the community against advances of this right wing agenda. Be a vote for serving vulnerable and marginalized communities. Be a vote for diversity. Don't let the worst people in your county take over and ruin this public service. Don't let them ban books about trans people as has been happening in public libraries around the country.

Be a vote for the workers! To ensure that library workers are treated well, paid well, and safe.

I know a lot of people often feel like they can't do meaningful activism or good work unless they live in a big urban center. I'm telling you: you actually do live in a community. With kids and seniors and immigrants and disabled people and queer people like you who you just haven't met. The public library serves their needs because they live everywhere just like you do. If you're an adult who can donate some time to the library, look into taking over a board seat before some young earth creationist does instead. Before the only libraries left with queer books are the ones in major urban centers in solid blue states.

It's going to be very very different jurisdiction by jurisdiction how this works. You'll have to do research. If you don't know how to start, try asking a reference librarian at your local library.

Fun aspect of ancient history is when you look for political or cultural accounts the historians all say "no one can really say if this took place, or if this guy really existed, but we know that around this time a person may have instituted a policy of some sort that might have ..." and then when they're talking about a battle suddenly the detail level skyrockets and its like "thirteen guys moved in on the left, it was drizzly out, this spearman was left-handed, he ordered them to flank on the right and then changed his mind a few seconds later..." etc. Its the historical version of intuiting an artists fetish by how hyper-detailed the feet are in an otherwise totally normal drawing

if you knock the teeth out of the mouth of a Don Bluth character and plant them in the ground, a fully armed and trained legion of soldiers with Strong Opinions about fetish artists will spring from the earth.

A lot of the discussion around the huge automated statistical models that are conventionally called "AI" involves the idea of hypothetical future improvements to the system that will make it concretely useful for things that it's currently not very good at. While undoubtedly it will get better at the sorts of things it can already do, and may even at some point be able to render a hand that doesn't make me want to vomit, there's one important thing that gets glossed over in a lot of conversation because programmers think it's so obvious it doesn't even warrant mentioning and non-programmers may not be aware of it at all.

AI cannot be programmed. AI is not like science fictional depictions where you can just build three laws into it and have it unerringly follow them, and it's not like conventional software where it rigorously follows a minutely precise set of instructions. You cannot simply make the Bing chatbot more accurate by plugging in a big database of verified facts and rules of deduction, because it fundamentally has no idea what truth is. It is a statistical model of what people are likely to say on the internet, and it's so unimaginably huge that even a team of humans couldn't possibly manually correct it except in the most broad strokes imaginable.

You can't even tell it "this is what a statement of fact looks like" because to tell it anything at all, you need an approximately-internet-sized corpus of training data with annotations that accurately indicate that information and that doesn't exist. The only internet-sized corpus is the one they've already used, and it certainly doesn't have sentence-level semantic metadata. So you're stuck: you can push the statistics as hard as you want but they'll never really do what you want because you can never tell them what you want in a language they'll understand.

Yep. AI does not Understand what it is doing, it is assigning weights to inputs and outputs and wiggling them until they match the parameters that were fed to it to aim for, which is when we get back to the point that AI is sleight of hand over immense amounts of human labor to tag data, and that tagging data is where a lot of shit goes wrong and introduces bias.

There is also cases where the AI will infer connections you absolutely did not intend or are often spurious, because of random correlation over a large enough data set (like those charts that show a statistically significant correlation between ridiculous things), and God knows your training sets cause problems on your own. My favorite example i've found was a generator for portrait shots of people that had a gender slider, and when you set it to maximum Male, the pictures would have weird, spiraling black extrusions that ended in a bulb around the neck or mouth of the person.

This confused the hell out of me until I tweaked the sliders to make it more legible, at which point I realized that the training of the AI had correlated "is speaking into a standing microphone" into pictures tagged Male. Therefore, the ultimate expression of masculinity was the ability to warp reality and create a microphone so you never had to shut up and everyone had to hear you.

Which, like, if that'd been a purposeful art installation piece I'd have mumbled "alright there Banksy" but the fact it was an emergent result of people not curating their data set entertained me for a whole lot of reasons.