If reddit hadn’t locked their API behind absurd paywalls, it would have been a cool project to try to make a browser plugin that gives accounts a “credit score” based on the factors you’ve been looking at, in order to let users quickly judge how likely an account is a bot.
It could let people adjust the metrics it uses to calculate that score in the settings, so even if it becomes popular enough for bots to start trying to game the system, people can adapt their scoring metrics themselves and share config profiles that they think are more effective at rating bots.
Might be something cool to see for activitypub/fediverse/lemmy accounts, but with the data available varying by instance it might be a little harder to calibrate a “catch-all” scoring config
If reddit hadn’t locked their API behind absurd paywalls, it would have been a cool project to try to make a browser plugin that gives accounts a “credit score” based on the factors you’ve been looking at, in order to let users quickly judge how likely an account is a bot.
It could let people adjust the metrics it uses to calculate that score in the settings, so even if it becomes popular enough for bots to start trying to game the system, people can adapt their scoring metrics themselves and share config profiles that they think are more effective at rating bots.
Might be something cool to see for activitypub/fediverse/lemmy accounts, but with the data available varying by instance it might be a little harder to calibrate a “catch-all” scoring config