In Recursive conversion from ExpandoObject to Dictionary<string, object> the OP omitted the keyword private, and after receiving answers, edited the post to include that keyword. I rolled it back because it was mentioned in Jesse's answer, and left a comment, pointing the OP to the What to do when someone answers page in the Help center. The OP then replied:

Sorry I did not know. I feel like this was just a small mistake I made which may have caused some misunderstandings. Do I really have to ask a new question just for that (becaues I feel like my main concern didn't have to do with that at all) or how would you recommend me clearing that misunderstanding up to the concerned question? 1

I checked with Jesse to see if the revision was okay, and the response was:

I think the revision is minor enough, I could remove a single line from my answer to keep it salient. 2

Should we re-apply the OP's changes?


3 Answers 3


No the question should stay as it is,
because not only would this edit invalidate two answers (first, second) but I think we would send the wrong signals as well. It would look like it could be allowed to invalidate answers if one just asks an answerer of the question if he/she is ok with it.

This could lead to extensive comments and could weaken the enforcements of our rules.

  • 1
    \$\begingroup\$ The problem with this question was that more context became available through comments and edits, while answers were already made. I feel the OP should close and make a new question with updated context and a better specification of what the code is about. \$\endgroup\$
    – dfhwze
    Jul 3, 2019 at 4:47
  • \$\begingroup\$ @dfhwze OP's cannot close question themselves, only delete. However, if they do delete the question answerers are probably gonna give up on answering and just think "Oh, he changed my mind on asking for a review. Oh well, then there's no use for me to continue to write one. <deletes all progress of writing an answer>". \$\endgroup\$ Jul 3, 2019 at 7:48
  • \$\begingroup\$ @ Simon Forsberg Well I have taken down my answer because the initial question did not correspond all the comments and some edits. \$\endgroup\$
    – dfhwze
    Jul 3, 2019 at 7:54

Why do you need to come to meta?

In this case both parties involved are ok with it. Let them edit their posts however they want.

Surely we don't need a rule that says:

If an answerer agrees and removes part of their answer that an OP wants to change, then the OP can answer invalidate the older revision of the post.

If you actually think about it, that's already a 'rule'...

  • 3
    \$\begingroup\$ Perhaps all people that have answered should review the edited code. In this particular question, the OP made a new edit in which he added a use case of the code. With all comments and edits made, I feel my answer is almost entirely mute. Perhaps the question 'lacked context' and we all got too excited answering in the first place. \$\endgroup\$
    – dfhwze
    Jul 3, 2019 at 4:45

Let this revision through

The core issue is whether the questions and answers continue to make sense together. In this case it is easy enough to keep all that as true so it seems harmless to say let them add the private in this case.

Could it lead to a slippery slope?

The regrettable part of making a decision like this is that we've the rules more complicated which will make life harder for reviewers (code and meta) for the foreseeable future. I'm not sure how best to define an edit to the code of the question that would be minor enough to say that it was ok.

And it leads to deeper questions: what happens if we allow a minor edit to a question but the authors of the answers written before that edit decline to update their answers? Should we leave them alone or try to edit them into shape?

  • \$\begingroup\$ "I'm not sure how best to define an edit to the code of the question that would be minor enough to say that it was ok." And that's a surprisingly good reason to err on the side of caution and disallow this edit. A follow-up question would be more appropriate and won't lead to a slippery slope. \$\endgroup\$
    – Mast Mod
    Jul 3, 2019 at 11:58

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .