If it helps:

1. this is controlled by a user setting. I left the one that automatically "collapses" comments below a threshold at the default, but I disabled the one that "hides" comments by setting the threshold to -10000. So, far from taking away user power, it strictly enhances choices by providing new options, only at the user's behest.

2. it does have such a "reputation" feature, as too does life. Someone who constantly trolls others gets rather "known" for such. But crucially, it's a label - it doesn't hide anything, only enhances what is already there. And yeah it's a bit of an experiment, perhaps it won't work. Or perhaps it will be improved further? Based on the above and the responsiveness of the devs, I would expect complete control if features were ever added to actually do anything wrt this score.

Btw apps already have something similar, as too does PieFed, when adding a label for new accounts - bc people have asked for it, and it can be helpful to know when talking with someone that they are a new account (perhaps they are an alt, but it's something, and again it's just a label).

Yeah, I constantly get downvoted - and some of my posts are among the most heavily downvoted content existing in certain communities (but I also note that such things as Innuendo Studios The Alt Right Playbook got heavily downvoted by the same community as well so... I feel vindicated:-). So I mean it when I say that believe me I KNOW what you mean when expressing those concerns. Perhaps the experiment won't work out, or perhaps it merely needs tuning - e.g. so that any one post or comment doesn't weigh so heavily but rather only their aggregate (median rather than mean perhaps? or maybe only the binary choice of positive or negative total score, and even then perhaps not centered at zero but something more highly negative like -10?).

Also PieFed.social has defederated from hexbear.net and lemmygrad.ml, so those sources of downvoting are entirely removed. It also preferentially weights scores more highly feedback from those with high reputation already - which state I achieved in roughly a week and with only two posts, one a cross-post of the other even. So it's not like seniors are locking out the noobs.

Anyway yes there's enormous potential for misuse there, but it's also something that people have been clamoring for - so it's something that they are being responsive enough to try it out?

3. I'm not sure about the categories - but again the devs are very responsive so surely easy to change things? Also I've definitely joined communities that aren't in those, and while there are large federation issues with any non-Lemmy.World instance right now (I see the same from many instances including my 2 alt accounts elsewhere - so it has little to nothing to do with PieFed; especially after the enormous surge in content surrounding the USA election), I believe that they show up in the main feed.

4. I have never heard that before but I would support it - more "experimental" communities should be allowed, to try things out, a "safe space" if you will:-).

All of these are valid concerns - and seem like they are being worked on.

It doesn't need to be the answer. It just needs to be an answer for certain use cases. Both platforms can easily coexist. That's the beauty of federation.

Lemmy admins can already see who downvotes what, I'm sure they already ban accounts who systematically downvote their communities content

It's a tool. If some admins power trip, well report them on !yepowertrippinbastards@lemmy.dbzer0.com

PieFed communicates with Lemmy. Same content, different platform. That's one awesome thing about federation.

There is also mbin (fork of kbin), and Sublinks, which is API compatible with Lemmy so should be able to use Lemmy apps with it (from memory, this is what Beehaw are hoping to move to).

Hello from PieFed - yes it works.:-)