Well, on DPReview, posts could be edited after other people replied. Just not by the original author - unless they were a moderator.
It was possible for a moderator to silently - at least from the user side - gaslight users by changing their posts.
I had a problem with this on the Studio & Lighting forum, with Ellis Vener. Now Ellis Vener is knowledgeable, and generally helpful, but he's rather partisan about a flash trigger (Raven) apparently made by some mates of his. And apparently sometimes he edited comments to make these comments - in his opinion - read better, or make more sense.
He made some false - physically impossible - claims about a new version of this Raven trigger, which I briefly contradicted. As it happens, the Raven thing sounds quite good to me, but I don't need what it does.
Two people replied to my post: Ellis Vener, and a UK-based photography journalist. Evidently at nearly the same time.
In Vener's reply, I was surprised to see a quote of my post slightly different from what I had written. He generally agreed with my comments. I checked my post, and found that it did not match what I thought I had written. Which I found quite disturbing.
However, in a post that appeared a moment later, a second post quoted my post without modification.
Here's what I posted, on DPReview (quoted): www.dpreview.com/forums/post/66710572
Here's Ellis Vener's reply to my post (with edited quote): www.dpreview.com/forums/post/66710433
Here's my post (edited by Ellis Vener): www.dpreview.com/forums/post/66710209
Now, in this case, Vener's editing is trivial, if a bit petty.
But in another case, if I said that I didn't write what was in a post with my name on it, most folk would call me a liar.
So:
I believe it is essential that the last editor of a post be publicly visible, including in the case that the editor was a moderator.
Some forum software allows a reason for edit to be entered, and displayed below a post.
If edits by moderators are allowed, they need to be logged, and the moderators of the moderators should review them - such edits should be very infrequent.