RIght now lemmy doesn't calculate or display a user's "karma". And many think this a good thing (me included).
Interestingly, kbin does calculate karma, even for us lemmy users (you can all probably just search on kbin.social and find your karma now, +/- federation inconsistencies).
Whenever karma comes up, this fact often comes up, along with the identification of up/down voters, such that many lemmy users will probably know that they actually do have karma and can go look it up if they want to. Some lemmy apps/frontends are also reporting karma AFAIU.
So I think the question now presents itself of whether this is an issue we want users to have some control over, within the bounds of what can done over federation/AP of course.
I can imagine a system where karma is an opt-in setting of one's profile, and a protocol is established that any platform/client that understands up/down votes ought to respect this setting and that non-compliance risks defederation.
Though lemmy/kbin obviously lean more "public internet resource" than microblogging platforms like mastodon, I think it makes sense to value user health and safety here, and this seems like a not unreasonable option to establish a norm around.
Thoughts?
I appreciate the instance based approach. And it is also probably the easiest from a tech perspective, just as downvotes are optional for an instance.
But it would mean that the user’s wish to not have to worry about their karma would not be respected across the fediverse, and I think that that’s an issue worth considering.
In your proposal, though their karma is not visible on their home instance, it could potentially be visible in every other instance, and therefore of as much concern whether presented on their instance or not.