If you want to actually improve archival access, you do that by spending money providing liveable wages and more resources to existing archives and professionalizing the thousands of volunteer-run local repositories. Spending money on AI/genAI isn't helping either.

And Os are often read as zeroes! OCR is useful but flawed. These tools that pretend that they can work with no human intervention on the backend whatsoever are ridiculous. That assumption always ends up making more work.

Indeed. And both of us know that that’s not the case.

And this is what librarians spend all our time trying to facilitate. Thinking about how stuff is organised, how can we help you find what you don't even know you don't know. It's literally the joy of our profession

We do have some very limited uses for computer pattern recognition—there has been some tools developed that can compare photographs and another that helps transcribe handwriting. But both require human confirmation to confirm and apply metadata, etc. It doesn’t make up stuff, it doesn’t falsify

My point is there is no substitute for Deep Work. For wrestling with readings you don't understand. For trying to make connections between 5 things you just read and the context beyond that. AI cannot do this and it robs you of the foundational skills you need in whatever career you choose.

Like, she’s the “Mother of Hubble”! She’s the driving force behind all those colorful, sublime space images from my childhood! Why did no one tell me about her??

Also, how did I never know about Nancy Grace Roman until today?