

Adding to this: a lot of the most heinous, truly “problematic” stuff ended up being demonstrably false accusations - like a feature may be opt-in, or at worst opt-out, but either of those is a far cry from being hard-coded as was claimed.
That said, the software truly is “opinionated”, pushing its own agenda even if you can opt-out of most of it - an instance admin still has to be wary of each new feature added, especially with the lack of transparency surrounding the announcements.
Although personally I think Lemmy is equally opinionated, just more in the sense of what features it decides not to implement at all.
Both platforms have their advantages, and it seems more important for an end-user to pick a good admin team that reflects their values more than picking a platform, especially if using a 3rd-party app than replaces the UI anyway. (Though secondary to that, I do think PieFed has the strongest edge, especially right now before the release of Lemmy 1.0.)

It’s too bad they won’t allow funding to go towards development of just the software itself, separately from the moderation and other costs of maintaining Lemmy.ml. As it is, Lemmy is a tankie platform first and foremost, and only secondarily offered for use by the rest of the world. You can see this in the development philosophy even: like even Reddit offers a modmail, notification upon moderation events, and ability to see deleted posts (PieFed offers some of this too), though Lemmy offers strong rights to instance admins, and somewhat lessened ones for moderators, but significantly weaker privileges to mere users (who I guess they feel don’t deserve to be told upon e.g. being banned from an instance, and instead merely have to find out the hard way?).