8

Matt Gutting added a very considerable amount of detail to this answer...

https://scifi.stackexchange.com/a/161520/20774

... turning it from a two sentence stub (that more closely resembled a comment) into a well-referenced and highly relevant answer.

It has 95+ upvotes and rolling it back dramatically harms the answer. I certainly would have rejected it (as conflicting with the OP's intent) if it had turned up in the review queue, but was I right to roll it back at this point?

Please note that I'm not asking if he improved the answer (I think we can all agree that he did) but whether such improvement is right and proper.

  • 2
    I agree wih your decision. The post received majority of its upvotes from the edit. I'm fairly sure it was at 30 upvotes when it was posted yesterday but jumped (sorry, skyrocketed) to 90+ upvotes. – Mat Cauthon Jun 14 '17 at 8:51
  • 4
    @Voronwë You're mistaken. The post was already on 80 or 90 upvotes before MG's edit. – Rand al'Thor Jun 14 '17 at 8:54
  • 1
    @Randal'Thor The post received 39 upvotes (going to 69) between initial vote count (30 up, 3 down) and MG's revision. – Gallifreyan Jun 14 '17 at 9:00
  • 11
    @Voronwë The consensus (SE-wide policy) is that we don't delete bad answers. We make them better. It's what Matt Gutting did. It's what SE is about. Period. – Gallifreyan Jun 14 '17 at 9:06
  • 2
    @Gallifreyan It received 39 upvotes during 2017-06-13, and MG's edit was in the middle of the day. From the post's timeline, it's impossible to tell how many of those upvotes were before the edit and how many after, but I can tell you from memory that most of them were before it. – Rand al'Thor Jun 14 '17 at 9:09
  • 1
    @Voronwë - I don't think it needed to be deleted. I do think that Matt Gutting's edits were worthy of being added as a new answer – Valorum Jun 14 '17 at 9:09
  • @Voronwë - That would be my feeling. – Valorum Jun 14 '17 at 9:11
  • 3
    @Voronwë Again, voting is completely up to the voters. Yes, SQB put more effort. Yes, they got less up votes. But no one can make people vote on the "deserving" answer, because that would be futile. Just leave the answer in its better form and move on, folks. – Gallifreyan Jun 14 '17 at 9:12
  • 4
    There are two problems here. 1) Was the answer good enough as it was, and 2) was the additional information added accurate and relevant. In my opinion, 1) the answer was fine as it was, as it answered the question. 2) the information added was accurate and relevant, and didn't change the meaning of the answer. As such, I would have left the edit as it was, even though the answer was good enough already. – Dr R Dizzle Jun 14 '17 at 9:35
  • 2
    @DrRDizzle - My problem is that the scale of this edit means that it's no longer the OPs answer, it's MattGutting's answer riding in an OP vehicle. – Valorum Jun 14 '17 at 9:51
  • 11
    @Valorum The edit certainly made the answer longer, but actually added literally 0 additional information - just "evidence". It's still very much Roseman's answer. – Dr R Dizzle Jun 14 '17 at 9:55
  • 1
    @Gallifreyan "The consensus (SE-wide policy) is that we don't delete bad answers. We make them better. [...] It's what SE is about. Period." Do you have a citation for that? I genuinely do not know of any such consensus or documentation that we're meant to do anything like that. My impression is we have no obligation or pressure to improve bad answers, we just downvote them and/or remove them. – doppelgreener Jun 19 '17 at 10:35
  • 2
    @Braiam Whether to edit or not isn't the question, it's whether SE compels us to edit every single low quality post to make it good quality. I don't do that. Gallifreyan made an extreme statement suggesting a fundamental part of the Stack is doing that. I'm fairly sure that's not the case. – doppelgreener Jun 25 '17 at 21:45
  • 1
    @Machavity - I'm not overly happy with the title change. This question was about the specific rather than the general – Valorum Jun 27 '17 at 15:40
  • 2
    @Machavity - My concern is that changing the title gives carte blanche for good vandalism – Valorum Jun 27 '17 at 15:42
21

We should leave Matt Gutting's revision in place. Don't roll it back.

I'm inclined toward that Matt Gutting should have posted their own answer rather than revise that one. I have a personal rule of thumb that I don't put more effort into revising a post than the author put into that post in the first place, and I think in future situations like this, Gutting should post a new answer.

However it's now 2 days later and the post is at +95. Rolling back that edit at this point is just destroying useful information and disruptive to our mission to “build a library of detailed answers to every question about science fiction or fantasy.” The opportune time to roll back has long passed, so let it stick and move on.

Overall both users are 10k+, and the author is 30k+. Rep at that point is just a number. I'd be slightly worried if the user was brand new and was unlocking privileges whilst not having really learned to use the site effectively yet, but I don't have to worry about that for a moment in this scenario.

  • 1
    I don't think the age of a post is necessarily a defence against being fixed/improved, nor should it prevent these sorts of rollbacks. I actioned this as soon as I saw it. – Valorum Jun 14 '17 at 9:50
  • 3
    Whilst I agree with the principle that age shouldn't prevent a post from being improved or problems in it resolved, it remains the case that in this circumstance the rollback is disruptive and sets us back in our mission rather than helping us move forward in it. – doppelgreener Jun 14 '17 at 9:52
  • 1
    If we encourage Matt to post his info into an answer on the original question, we preserve the info and prevent the OPs answer from being co--opted. – Valorum Jun 14 '17 at 9:53
  • 6
    At this point I don't see the need, and the answer's been collaborated on with acceptance from the original author (like Gallifreyan pointed out) rather than co-opted. Mission accomplished, good progress made. The effort spent separating it out is at this point unnecessary and disruptive. – doppelgreener Jun 14 '17 at 9:55
16

Oh wow. I didn't realize I'd be upsetting the hornet's nest on this.

My rationale in changing it in the first place was that it was clearly an insufficient answer. If it had been flagged and come up in the review queue I would have been given the opportunity to edit it and mark it acceptable. Granted it wasn't flagged, at least not that I saw, but it was at the point that I would have flagged it. Instead I just skipped that step. Was that appropriate? Well I certainly thought so at the time, in the name of producing the best possible answers. Now, I'm not sure. I can definitely see the opposing side.

I upvoted the answer immediately after submitting my edits. Mine was the 84th (net) upvote: the question stands right now at 95 net upvotes. That's a considerable but not overwhelming increase.

All that said, I honestly have no particular feeling in what the outcome of this discussion should be, and if the decision is to roll back the edits that's fine, I won't object. I'll post the identical answer as mine. But to me that doesn't seem fair to the answer's author. I'd like to see them weigh in.

  • 4
    We have had (at least since I joined, and presumably for years before) a problem with "good vandalism", where someone comes along and defaces an answer with sensible, helpful and relevant information :-) – Valorum Jun 14 '17 at 10:45
  • 1
    For me, it boils down to the fact that had your info not been correct, we'd slam this in a heartbeat. So the wider question is, is the fact that a large edit contains correct info enough to save it from the chop? – Valorum Jun 14 '17 at 10:46
  • 4
    I'd like you to post it as a new answer so I can upvote your effort and not the OP who posted a stub (which I'd like to downvote). – Valorum Jun 14 '17 at 10:48
  • @Valorum what do you say to my argument, that if an answer is particularly bad, people who see it in the review queue are given the opportunity to edit it, with no input from the OP, who then (I believe) gets all the benefit of, or blame for, the subsequent quality of the answer? – Matt Gutting Jun 14 '17 at 11:12
  • 1
    I am forever improving people's answers by adding quotes and links. What I strive to do is to only add info that I'm confident that the OP is likely to be aware of and nothing that involves independent research of any note. – Valorum Jun 14 '17 at 11:19
  • This for example could have been added to SQB's answer but I'm reasonably certain that he doesn't have that book. I'm also mindful of the fact that I deserve credit for finding it. – Valorum Jun 14 '17 at 11:22
  • 4
    @Valorum I am new to this site, but wouldn't you slam any edit in a heartbeat with incorrect info, regardless of the size of the edit? – Philipp Flenker Jun 14 '17 at 13:02
  • 1
    @MattGutting: I'm one of the people Valorum occasionally chides. Ever since I had one of the stub-answer people complain I'd substantially changed their answer, I've been trying to recognize when I've gone to far, and to write my own answer even when I feel like it might be "stealing" the answer. It's difficult. – FuzzyBoots Jun 14 '17 at 13:19
  • @PhillipFlenker - I thinks that's also fair comment – Valorum Jun 14 '17 at 13:31
  • @Valorum I guess the other issue for me personally is that I (as a Middle-earth fan) don't feel I put much effort into the edit - it might look like a significant amount of work, but I don't feel like it was. It took a couple of minutes, no more. – Matt Gutting Jun 14 '17 at 18:56
  • 4
    @MattGutting - It must have taken at least twice as long as it took for OP to scribble down their answer on the back of a metaphorical napkin – Valorum Jun 14 '17 at 18:57
  • @Valorum Perchance :-) I don't know. – Matt Gutting Jun 14 '17 at 18:57
14

I wouldn't have rejected it in the queue. It doesn't change the meaning of the answer - it adds relevant information. It's always good to improve a bad answer and make it good - it's better than leaving bad content around.

I know that I've certainly added supporting quotes to answers that I felt needed them, sometimes resulting in adding several thousand characters to the answer. I feel that you should not have rolled it back, as it only vastly improves, not changes.

  • I agree that it adds relevant info, but there's no indication that the OP was even aware of this info, nor did the editor attempt to communicate with the OP – Valorum Jun 14 '17 at 8:59
  • 1
    The edit automatically sends an inbox message to the OP. If they disagreed with the edit, they could have rolled it back themselves. – Mithical Jun 14 '17 at 9:00
  • 1
    Just because someone likes an edit, doesn't mean it's appropriate. – Valorum Jun 14 '17 at 9:20
  • Would you roll back this edit of mine (2-4)? – Mithical Jun 14 '17 at 11:08
  • Those are some bloody huge edits. – Valorum Jun 14 '17 at 11:17
  • 2
    @Mithrandir - That edit had disturbed me at the time. I'm just not the type to rollback vandalism. I leave that to people like Valorum. – ibid Jun 14 '17 at 12:53
  • @ibid - Cheers for that. – Valorum Jun 14 '17 at 13:32
  • 1
    @Valorum - Also, Mithrandir had sought out and received the approval of the OP. – ibid Jun 14 '17 at 14:10
  • @ibid - The very next comment indicates that the OP had no intention of using those quotes, nor doing the research required. – Valorum Jun 14 '17 at 14:21
  • 1
    @Valorum - The important thing is that the editor sought out the OP's approval for the edit and that the OP indicated that they accepted the edit. – ibid Jun 14 '17 at 14:27
  • 1
    @Valorum - I don't think it's relevant how large the edit was - over 6k characters added or only a few - as long as it doesn't go against the OPs intent, it's fine. E.g., the first edit (mine, BTW) is fine, the second is not. – Mithical Jun 14 '17 at 14:35
  • @Mithrandir - The size of the edit is largely irrelevant. If the edits represent more than 90% of the combined answer, you've probably gone too far – Valorum Jun 14 '17 at 15:06
13

Does it make the answer better?

Yes it definitely does. Another person with <2k rep had suggested the same edit as you, and I had rejected it on the grounds that it actively harms the answer

Was Matt Gutting right in adding those parts?

It's not up to us. If it's their decision to add those parts to an otherwise (not my words) low quality answer, instead of posting an answer of their own - it's their decision.

Do we have the right to roll the edit back because the answerer "doesn't deserve the upvotes"? No! This site is not about upvotes and reputation - it is about building a complete collection of questions and answers. Matt Gutting's revision made said collection better, so there is no point in rolling back the edits because of the feeling of justice.

I had done similar things myself (e.g. here) with an otherwise low-quality post, and I had added information to other people's answers without asking first. the way I see it - when I edit, I offer my help with source search. If the OP decides to roll it back - then it's their decision, but at least I took a stab at making the post better.


If Daniel Roseman decides to roll back the edit - that's their decision, since it's their answer. Then the edits fall under "conflicts with OP's intent", and then it's not up to us.

  • 3
    I disagree. Users don't have the right to utterly co-opt someone else's answer. At the moment, it's MattGutting's answer that just happens to be sat on top of DanielRoseman's. – Valorum Jun 14 '17 at 8:57
  • 13
    @Valorum I see it differently - Matt Gutting offered help with sourcing - Daniel Roseman accepted it. The end. – Gallifreyan Jun 14 '17 at 8:58
  • 1
    I can't see where that offer of help was made. Was it in chat or a deleted comment? – Valorum Jun 14 '17 at 9:00
  • 6
    @Valorum The post was edited. The answerer has the opportunity to review the edit and roll it back if they deem necessary. Does one have to ask for permission to add relevant information when they're not changing the meaning of the post? – Gallifreyan Jun 14 '17 at 9:05
  • 2
    Ah, I see. So because the OP didn't reject the edit, that makes the edit acceptable? I think that conflicts with a whole bunch of policies. – Valorum Jun 14 '17 at 9:08
  • 3
    @Valorum Links or didn't happen. – Gallifreyan Jun 14 '17 at 9:09
  • scifi.meta.stackexchange.com/a/5165/20774 perhaps? – Valorum Jun 14 '17 at 9:14
  • I noticed you've rolled back my rollback (which I've now rolled back) It would be appropriate to wait until we have a more settled consensus from the community. – Valorum Jun 14 '17 at 9:19
  • I've temporarily locked the post to prevent further rollback wars. Once this meta discussion has reached a conclusion, we can re-edit as necessary. – Rand al'Thor Jun 14 '17 at 10:03
  • 3
    @Valorum I disagree with that policy. Firstly, for story-ID answers all the details would come from the story itself, and that by definition can't be wrong (what can go wrong with a one-line answer as long as they don't change the title of the story?). For this answer, I agree with DrRDizzle that Matt's edits don't actually change the answer, only support it with data that is known to be correct. – Gallifreyan Jun 14 '17 at 14:24
  • 5
    @Valorum the Meta answer you link to says among other things: "To me, edits to answers or questions by other people should be restricted to things like spelling, grammar, formatting, and things like additions of links to sources/references that were already mentioned but not actually linked, or reasonably helpful images." That's what I felt I was doing. – Matt Gutting Jun 14 '17 at 17:23
  • 1
    @MattGutting - Airily waving a hand and saying "it's in the book" isn't a mention. – Valorum Jun 14 '17 at 17:27
  • 5
    @Valorum The answer states that this is a description given in the book. Maybe if the answer said no more than "Tolkien uses the phrase in the book", I'd agree with you. But the author mentioned that the phrase was used and gave some context to what the usage was. I'd count that as a mention. – Matt Gutting Jun 14 '17 at 17:35

You must log in to answer this question.

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