Elon Musk broke Twitter, and he's breaking research communities, too.
I'd be curious to on the ways that twitter was better than say the blogosphere/rss days. I can buy that my perspective is too nostalgic here, and if that's a subject you want to cover in any future posts I'd eagerly read it.
I'd imagine that part of it is that the producer time commitment is more manageable and managing a larger set of contacts is easier. I've found these properties useful in twitter for transit advocacy. On the other hand, my RSS reader works for me and on social media I am the product and my attention is being optimized by a central set of engineers towards ends that are not my own.
The late Twitter usefulness you're describing is really a product of the early Twitter culture--essentially before the guys who made it had any idea whether it even could be monetized, they just knew it might be popular and/or useful. Reddit is starting on the same death spiral now--the dumb people in charge now need to try and figure out how to squeeze blood from its rock convincingly enough to make it through an IPO, after which they'll cash in and run. I think what made early Twitter generative compared to the blogosphere/RSS was that it was a new platform (so new entrants) plus you could share and consume high-value information with a light burden in terms of content labor whereas the blogosphere took more work (of a kind largely discounted by anybody who wasn't in it).