I think I’ve settled on the latter. Disagreement is maybe best communicated by the absence of an upvote? And downvotes work best when they signal something that is just off base, and while not reportable, is not appreciated at a broad cultural level.
Depends on what kind of post it is.
General discussion threads, sure - ‘up’ = ‘good content’, ‘down’ = ‘irrelevant’. Irrelevant could be because it’s not to do with the matter at hand, it could be hateful, trollish, whatever.
Post asking for a specific fact, like in ye olde askahistorian? Up = correct, down = incorrect. Doesn’t matter how well written or how good the intent is, downvoting for disinformation.
One of the things that Slashdot got right was being able to upvote / downvote with a reason. (Perhaps only being able to upvote / downvote occasionally too, which stops brigading.) Made it possible to filter on why things were good, save ruining your fake internet points when you were mistaken about something as opposed to being an arsehole.
Interesting! I’ve kinda thought this myself, that having a sort of sentiment meta data attached to online actions would be an interesting way to go, kind of as a substitute for the body language and gestures we use and pick up on in real life.
Enter: the wheel of upvote options and the multidimensional spectrum of downvote options. Don’t worry, I’ll ask Google to analyse my life history and feed it into the emote-i-vote.
Come to think of it, I like the attach emoticon thing in GitHub (and lots of other social media? But I’ve liked it in GitHub) to get a relatively convenient and concise expression of “I like your message in this particular(ish) way”
Yea emoji reactions are fun. Calckey has them and it works well.