Let me propose the following general corollary to the "wiki-like nature" of the site:
If a post is written in nonstandard English or nonstandard mathematical notation, and
- there is a risk that readers without special knowledge (unrelated to the mathematical field in question) will not understand it, and
- it is clear and unambiguous to someone with that special knowledge what is meant (i.e., no risk of changing the meaning of the post),
then it is acceptable for someone who recognizes this to edit it into more standard English / more standard notation without further ado.
In the instant case I think there can be no serious dispute that singular "they" is more standard and understandable than synthetic coinages such as "zie". Therefore, according to the rule above, Matt's edit is appropriate, and shouldn't have been rolled back simply as a procedural quibble.
Of course, it is also likely that dfeuer used "zie" precisely because (for reasons known only to themself) they don't consider singular "they" an acceptable solution to the gender-inclusive-pronoun problem. Thus, they might consider the edit to "deface" their post, ascribing to them a linguistic choice that they don't approve of. This consideration would be stronger if dfeuer had used a full name rather than a pseudo-pseudonymous handle. But in any case it would have been safer to change to "he/she" instead of singular "they". "He/she", if slightly clunky, ought to be unobjectionable and understandable to everyone.
The solution to this, however, should have been to edit the post further into "he/she" rather than rolling it back to "zie".