We recently encountered a spate of rollback war. Which got started under the misconception that guidelines which govern low-quality question editing is the same as improving and fixing an answer.

Were the improvements/edits made on the answer in the acceptable range?

Had the author of the answer approved them would it be fine?

And in general what are the guideline of improving/editing an answer?

To what level can an answer be edited/improved and improved?

Do the guidelines of question improvement apply to answers too? (On SO it doesn't.)

  • What's the final conclusion of this discussion? There's apparently no rule prohibiting the kind of improvements I made to the post! In fact stated SE policy encourages substantial improvements to the posts! – infatuated May 31 '14 at 10:05
  • 1
    I'll have to see into it @infatuated. I can't make a statement right now, the improvement are at a level that the side-by-side diff is almost rendered useless. I'll have to read both the versions then I'll be able to give my statement. The community is welcome and encouraged to do that though. – Bleeding Fingers May 31 '14 at 13:35
  • Thanks! Actually I'm aspiring to know whether such improvements will be allowed/welcome in future. Because there are many other posts that would make a rich contribution if substantively improved in phrasing and language. – infatuated May 31 '14 at 14:06

Without commenting on the specifics of this particular post (God-willing, I'll find the time to do a proper critique of the edits and post it as a separate answer), I'll throw in my two cents on the general editing of answers (community wiki notwithstanding).

First, for context, I quote the following from the help centre:

When should I edit posts?

Any time you feel you can make the post better, and are inclined to do so. Editing is encouraged!

Some common reasons to edit are:

  • to fix grammatical or spelling mistakes
  • to clarify the meaning of a post without changing it
  • to correct minor mistakes or add addendums / updates as the post ages
  • to add related resources or hyperlinks

The second point here is, I feel, the most important: To clarify the meaning of a post without changing it.

Voting on answers is of vital importance to this site. Not only does it reflect the actual expertise of the poster, but it reflects how useful they are to the community as a whole and by extension what powers they have to actually moderate the community.

If an edit to an answer substantially changes it, this system destabilizes. If, for example, the edited post demonstrates expertise that the original poster doesn't likely possess, the votes no longer reflect that; such post-edit votes would typically apply to the user who edited the post, but they're given to the original poster instead. Similarly, since there's no real way to distinguish pre-edit and post-edit votes, the score on the post no longer accurately reflects either; you'll very possibly just end up with a good post having a lower score than it deserves thanks to pre-edit voting.

This not only risks presenting the original poster as an expert (or non-expert) in a field incorrectly, but it also unfairly prevents the expertise of the editor from being reflected.

And if there is a fundamental problem with the post that renders it useless (such as it not actually answering the question, being pretty much unintelligible, or even being outright wrong), again editing is not the tool for fixing that. As above, editing such posts to make them useful incorrectly allows post-edit votes to present the original poster (rather than the editor) as more useful than he actually is; this can be highly problematic for the purposes of site-curation since user reputation is directly correlated to how much power that user has over the site.

The Stack Exchange model is built around the very concept of giving experts, useful experts, the tools and ability to moderate themselves, and editing should be done with this idea in mind: You're not trying to make the poster appear more (or less) expert, or more (or less) useful to the site, you're trying to clarify the poster's actual expertise and usefulness.

The difference between things like "fixing minor mistakes" vs "correcting a technically wrong answer", or "adding addendums/updates" vs "introducing new expertise" can be a bit of a judgement call, but it's usually fairly obvious to other users with domain expertise what a poster actually intended, or would've actually included had the facts been brought to their timely attention. Other experts are often the best judge of expertise and usefulness, and as such are often the best judge of what actually reflects that expertise and usefulness.

And if you find yourself wanting to make an edit that actually changes an existing answer on a fundamental level, chances are what you really want to do is just post a new answer instead.

  • 1
    Thanks for clarification, but does prohibition of "fundamental change" apply only to the meaning and the general theme of the answer or also to phrasing, grammar, and style? As for the post in question those were the areas that I improved upon with some elaboration on information already presented by the main author. The edited version of the answer neither makes a new answer nor fundamentally changes the general theme and main points of the original post. That's why I resisted roll-back. – infatuated May 28 '14 at 3:25
  • 1
    @infatuated That can be a very difficult line to define; in general I would stay away from significant phrasing/style changes, as they often carry particular nuance that may or may not have been present (or intended) in the original. God-willing, I'll tackle that if/when I get around to actually critiquing the edits, haven't had the chance to actually examine them yet. – goldPseudo May 28 '14 at 3:42
  • I see, thanks for you care, anyway! – infatuated May 28 '14 at 3:49

So far as the official stated policy for editing according to the help center guideline on Editing is concerned, which reads:

When should I edit posts?

Any time you feel you can make the post better, and are inclined to do* so. Editing is encouraged!*

Some common reasons to edit are:

  • to fix grammatical or spelling mistakes
  • to clarify the meaning of a post without changing it
  • to correct minor mistakes or add addendums / updates as the post ages
  • to add related resources or hyperlinks

Try to make the post substantively better when you edit, not just change a single character. Tiny, trivial edits are discouraged.

substantial improvements, contrary to the "tiny, trivial", are, in fact, encouraged. So that I think perfectly applies to the substantial improvements I made to the answer in question.

Yet, even assuming that the prescription against "fundamental change" as suggested by goldPseudo doesn't contradict the official policy of encouraging substantial improvements (because I am not completely sure if a substantial change can be non-fundamental), then my edits to the answer did not whatsoever fundamentally change the general theme, the main points and key implications of the original answer; Grammar, phrasing and style were the only areas I substantially improved while elaborating further on information already provided and referenced by the original author.

And as for the problem of fair expertise reflection, I think that's fundamentally a problem with the voting system. Maybe the site admins can work out a solution by distributing new votes justly among all editors whose improvements shape the final content of the posts. However, as for the post in question, I think it doesn't exemplify this problem as I highly suspect that grammatical, phrasing and style improvements represent expert contribution at all.

That's why I see the imposed roll-back as completely unjustified and demand restoration of my edits.

  • Such a vote distribution system would represent a significant change in the SE model, and would mostly just be trying to solve the same problem CW already does (which deals with such communal answers by giving nobody the reputation) – goldPseudo May 29 '14 at 21:38
  • @goldPseudo Right, so do you agree with my other points? – infatuated May 30 '14 at 3:28

After perusing* revision 5 and 8, I found only one statement worth commenting on:

Ja'fari School is thus distinguished from Sunni School in its recognition of the infallible scholarly and spiritual authority of Ahl al-Bayt which are viewed as the only legitimate inheritors and interpreters of the Prophet Muhammad's religion. Ja'fari School, therefore, draws upon traditions from all legitimate successors of Prophet Muhammad, a total of 12 Imams.

The author didn't have this added in their post. But since the addition of this statement didn't make any radical, incorrect and/or meaning changes to the posts, and gives future readers another bit of information to take with them. Therefore it definitely was a positive and healthy contribution.

When I first started comparing the two revisions I would jot down every excerpt and adjective that I thought were the editors addition. Those included adjectives such as "vicious", "undeserved", "supremacy" and some other phrases, sentences and paragraphs. And while I was jotting them down the adjectives I had made up my mind to advice the editor to not do that since it could come under the category of imposing one's foobar onto the author, although in the case of the answer under discussion it couldn't have been considered off seeing the author's post and the points made in it.

So you must be wondering why have I stated just one and removed the rest? That's because when I started adding my comments for each one individually I found (while searching for them in the revision page in order to get the context) each and every one of them, except the one mentioned above, was present in revision 5 too and that I was falsely assuming that it was the author's addition.

And even if they were, I found their presence perfectly appropriate and the only reason they were sticking on that 6+ item list of mine was that I wanted to highlight the fact that how they didn't essentially change the meaning of the post and were valuable.

About the overall edit:

Kudos! The effort that was put into editing that post must have been tremendous. It is commendable and highly appreciated. Wish to see more such community members.

That said, I recommend the post be unlocked and rolled back to revision 8.

* I tried to be as finicky as possible while comparing and analysing both the posts.

Some of the items that were struck out:

  1. yet never in decades to come, people would so widely turn to Ahl al-Bayt as best representatives of Islam and the Prophet's Sunnah as they did during Imam Sadi's life.

  2. Now the Prophet’s Ahl al-Bayt had to shoulder the Prophetic mission but they were denied the pivotal role they fully deserved and fully needed (and in fact were vested with by the Prophet) to carry on the Prophet’s mission in the Ummah and this was to be blamed on the self-proclaimed supremacy of much less worthy companions of the Prophet.

  3. got the opportunity to fulfill their pagan tribalistic dreams they'd been stealthily satanic desires

  4. Abassids’ undeserved power

  • Oh, I am actually amazed by your generous effort to examine my contribution, and I'm dearly grateful. Also delightful that one admin finally took the job. And also thank you for letting me know about another site policy, foobar non-imposition. But as for the "stricken" numbered list, I'm not sure I understand it. Are they the passages that need correction/revision?! I do notice that the third passage is confused! Or maybe they were the the excerpts that were initially suspected of foobar imposition that as you said were finally approved? – infatuated Jun 1 '14 at 5:50
  • 1
    I just provided them as reference to those who were curious what actual was on the list. Nothing important. (just realized that I had used the wrong word "stricken") – Bleeding Fingers Jun 1 '14 at 11:52
  • 5 items that were struck out: Did you miss an item, or did you just miscount? – goldPseudo Jun 1 '14 at 19:05
  • @goldPseudo It was a mess. I originally had like 7 on it. It involved jotting the points here, adding the comment in Emacs (It's-all-text), deleting all the items, syncing back here to find the items still in the review pane, got tempted to leave it for those who might be curious, copy-paste ... Fixed. Tada ;) – Bleeding Fingers Jun 1 '14 at 19:57

You must log in to answer this question.

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