There’s nothing stopping a malicious user from doing that right now. Be aware that anyone who wants can already see your votes.

We already depend on trusting instances for a lot of what’s going on here, I don’t see why we shouldn’t be able to defederate untrusted ones.

1. Currently, any admin can modify any local user activity, can’t they?
2. Not really, your local instance may still hold the vote data for validation. And therefore could be ported and resigned.
3. Don’t see the problem.
4. Today, each instance decides whomever they want federation with. The ones who decide the criteria should be the same ones who decide whom the instance federates with.

Your first comment expands on both privacy and security. There is no privacy without some type of security.

Now to answer your questions: Yes and yes. Users from c/all were downvoting posts from a small community I’m a part of because they don’t agree with. I couldn’t see the posts from small communities that are important to me because of that. Now we have the possibility to sort by “scaled”, which fixes that. Sometimes there are discussions that are very relevant as to who is voting for what. But that discussion has nothing to do with privacy, which was your first point and went unacknowledged on your second comment.

Not only admins can see the votes, but anyone on Fediverse (except regular Lemmy users) can see them.

Security through obscurity is prone to failure when it is used by itself. If people want their votes to actually be private then another method of securing their privacy should be created.

User choice would be best indeed. The problem is that currently the votes are public but hidden from Lemmy regular users. Anonymize votes seems to be such a big problem the devs don’t even want to consider it.

I completely agree with the idea of more accountability. We are real people in acting public right here, we should be constantly aware that our actions have consequences. If you don’t want your pseudonym associated with a vote, don’t do it. It’s kinda like the opposite of 4chan, where instand of anonymous controversial content on top, here we have human-curated content being pushed up.

What it the instance signs the activity? Then it propagates to others instances after local validation. That way only local admins would have access to voting data. Malicious instances could still be defederated/blocked/have votes disregarded.