15
\$\begingroup\$

See the question What's a good method of giving a certain monster resistance to ranged attacks made against it? as a recent example. The question has been fundamentally edited in order to get it reopened, and bears little resemblance to its original form. In cases like this, where there is no input from the original querent, it is impossible to known whether the new form of the question comes anywhere near close to useful to solving the querent's problem.

I've seen an uptick in this type of edit, where people will make massive changes to make a question fit the site, but in doing so they effectively completely change what the actual question is.

Is this kind of edit justified/OK/desirable given that there is a good chance the question we end up with is so different it doesn't actually help the querent?

\$\endgroup\$
  • 1
    \$\begingroup\$ So long as the OP is OK with the edit, and it does improve the quality of the post, then I don't see any real issue... The main thing is that the OP should be the one to make the edit, or at least post their approval of it \$\endgroup\$ – Ben Jun 20 '18 at 5:39
  • 3
    \$\begingroup\$ @Ben that's entirely my point. The examples I'm concerned about are those where it isn't the OP making the edit, and they never indicate whether or not they are OK with it \$\endgroup\$ – Wibbs Jun 20 '18 at 5:44
  • \$\begingroup\$ As a tangentially related point, I think I've noticed a significant increase in the number of edits generally on questions and answers in the last few months, driven by a relatively small set of users. I don't necessarily believe its a bad thing, but would be interested in any stats as to whether I'm right \$\endgroup\$ – Wibbs Jun 20 '18 at 16:54
  • \$\begingroup\$ I am the culprit that caused this, and I am sorry. I thought the only problem with the question was that it was too wordy and needed to be shortened, but it turned out to be more of a fundamental problem than that. \$\endgroup\$ – Southpaw Hare Jun 20 '18 at 22:07
  • \$\begingroup\$ @SouthpawHare Hey, it’s okay! This is how we learn. :) \$\endgroup\$ – SevenSidedDie Jun 21 '18 at 18:18
  • \$\begingroup\$ Wibbs, I have taken to directly addressing via a comment that I have edited, and a request to OP to review and take further action if needed. for the past few months. My record on that was uneven, and I have tried to now make it a personal SOP. The intent is to sort of the OP to try and evoke a response. I think it's helping, but maybe not. \$\endgroup\$ – KorvinStarmast Jun 21 '18 at 19:58
22
\$\begingroup\$

Heavy provisional edits that try to be faithful to the original question are warranted if the author is asked if they approve. Often, it is best done only after the question is already closed.

What happened here wasn’t helpful — the intervention caused more harm than good:

  1. It wasn’t closed, so justification for an overhaul to make it reopenable didn’t exist yet.
  2. The 3 existing close votes were for it being unclear — except when it’s a formatting and spelling issue or similar problems, edits by non-authors can’t fix clarity problems. That means the edit wasn’t even justified as preventative against further “unclear” close votes — and it didn’t prevent them.
  3. The author was not asked if the edits reflected their problem and continued to be useful to them. This is another reason to wait to do an overhauling edit until the question is closed: it means there is a pause during which the edit can be made and feedback can come from the author.
  4. The author did not approve of the changes and wrote an angry answer about it, and tried to fix the problem by putting the removed material into an answer. This answer had to be removed by mods, further complicating the situation and requiring explanation and diplomacy to try to head off an escalation from a frustrated user unfamiliar with how the site works.

A teaching moment was missed here, a question thrown into turbulence, and the hold process was interfered with by unrelated changes that didn’t at all affect the hold vote outcome. The question had not been clarified, the hold voters’ votes lost in a reopen that was done based on a now-obsolete revision, and the only person who can provide clarifications is busy being angry and diverted away from providing the requested information.

How to overhaul a question

A measured, patient, and collaborative approach to overhauls has worked for me for years. It involves not doing anything unless sure, not worrying about preventing close votes, and explicitly inviting the post’s author into the iterative, collaborative process that SE uses to revise questions. It focuses on improving and least change, reserving drastic change only for where it’s needed to polish the core of the question and raise it up above noise. It’s about making a question a better expression of its true self. It reserves edits to make it suitable for RPG.se’s topic rules only for when the change is superficial and leaves the heart of the problem unchanged

As with most guidelines and methods, there are always exceptions, but below is the general approach that fulfills the above principles.

Rule zero in overhauling a question: Have patience. Haste makes waste. Then:

  1. Understand the underlying problem first. If you’re not sure, stop.
  2. If there are close votes, consider whether the overhaul can be done faithfully while addressing the concerns of the close votes. If unsure, stop.
  3. Edit with an eye to expressing the problem faithfully and clearly. If making a question on-topic can’t be done without remaining useful to the problem confidently identified in (1), stop.
  4. When done editing, immediately leave a comment saying you’ve overhauled and asking whether it accurately represents their problem.

    Ideally this includes five elements: 1) that you’ve overhauled the question to 2) fix <topic problem here>, and 3) asking whether the question is still true and useful with 4) a note that if not they can roll it back including 5) a link to the post’s edit history.

This measured, patient, and faithful approach had polished many questions at RPG.se over the years. It’s tested and successful, avoids alienating askers, and gently draws them into how the site works with a positive experience.

\$\endgroup\$
  • 4
    \$\begingroup\$ I try to do this, and have been putting effort into being better about point 4 lately. I recently had HICC take a look at my edit, and we decided that my edit wasn't helping, so he made a better one. That positive engagement model that you have laid down in this answer is the kind of SOP I'd recommend to everyone. \$\endgroup\$ – KorvinStarmast Jun 21 '18 at 20:01
  • 1
    \$\begingroup\$ Is there actually a way to "provisionally edit", subject to a querent's approval, or is that just shorthand for "make the edits, leave a comment telling the querent what you've done and saying it's cool to revert"? \$\endgroup\$ – Glazius Jun 21 '18 at 21:45
  • 1
    \$\begingroup\$ @Glaziuz Sorry, yes, that’s what I meant. There’s no technical feature for it. \$\endgroup\$ – SevenSidedDie Jun 21 '18 at 22:37
10
\$\begingroup\$

I would say that such edits are not justified.

The core purpose of this community is to answer the community members' questions. This means working with the members to discover what those questions are and whether they are an appropriate fit for the format of the community. It does not mean guessing the intent of the members so that they can be answered. To use a poor analogy, we are not a puppy mill for answers, but rather a bespoke breeder of answers.

We have an established precedent of not guessing at the intentions of users when it comes to guessing at system tags. I think that the guidance of waiting for the user to clarify their intentions is equally applicable in this scenario as well as the tagging scenario.

\$\endgroup\$

You must log in to answer this question.

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