@jacob @anildash I feel like if I were an archivist, I would have smart things to say here about appraisal. Since what we’re doing here is, in a sense, deciding what belongs in the historical record, and I imagine the people who consent to such a bot are wildly unrepresentative —better networked and techier than the median user, and thus possibly replicating power dynamics that already plague the historical record. (Or: lulz actors spamming said record.)
Which isn’t to say you’re wrong. Consent is clearly really important, and there’s also all kinds of problems with people being included in archives without it. (Their content or their actual bones, e.g. https://projects.propublica.org/repatriation-nagpra-database/ .)
But I do wonder if the need here is actually _search_ or discoverability. Or maybe we’re trying to make search do all the things (onboarding, bringing the receipts…) when it shouldn’t.
Anyway. I’d check out Documenting the Now (https://www.docnow.io/), which has been doing social media archiving for a while in a way that’s really thoughtful and community-centered.