Hmm, this doesn't feel like the clearest criteria: > If user u is a member of p1 before time t and is found on platforms p1 and p2 at a later time t′. That use

Hmm, this doesn't feel like the clearest criteria: > If user u is a member of p1 before time t and is found on platforms p1 and p2 at a later time t′. That user is considered to have temporarily migrated from platform p1 to p2. Because this could mean two very different things: 1. They plan to use both systems, since they feel one or the other could shut down/take over as market leader/whatever and want to cover their bases. 2. Or they migrated to the new system, realised it didn't work for them, and went back. I'd classify myself in the first group, since I use Twitter, Mastodon, BlueSky and Threads concurrently. This also seems to include most people who publish their work online as either their career or main hobby, since they don't want to be stuck backing the wrong horse. Either way, it seems like some of the things the study highlights have quite likely explanations: - Less people from Mastodon return to Twitter/keep using it, since it's popular with both techies that prefer federated systems and open source software, and folks that disagree with Musk on a fundamental level. - BlueSky and Threads draw in influencers more than Mastodon, since the latter benefits from its Instagram integration and the former is only really possible to get an invite to if you're at least somewhat popular online/know the right people. - Mastodon saw less of a decrease in activity than BlueSky or Threads, because the people attracted to that platform weren't particularly focused on getting followers/high metrics. The average Mastodon user would probably be using forums or mailing lists a decade ago, whereas the average Threads user would probably be on something like Myspace.