Meta Stack Exchange is where users like you discuss bugs, features, and support issues that affect the software powering all 158 Stack Exchange communities.

What is meta?
Here's how it works:
  1. Any Stack Exchange user can ask a question
  2. The community provides support, votes on ideas, and reports bugs
  3. Your voice helps shape the way Stack Exchange operates

Something I ran into on SO today: OP asked a question, and got and accepted an answer (mine). Then 14 hours later a different user (not the OP) edits the question, so that now it is asking for something slightly different, and proceeds to comment that my answer is wrong and provides his own answer.

Is this OK behavior? It seems like a ploy to get rep: edit an old question, so that it bumps up to the front page, and so that the answers that were already provided look wrong and your new answer gets votes for being more correct. (None of the new visitors are going to bother to check the edit history.)

Should I roll back the revision to the question? I don't really care, OP has already accepted my answer, and I have better things to do than get in an edit war. It just feels a bit rude, especially since it came from a high-rep user who should know better.

(It is this question, if anyone is curious.)

share|improve this question
Instead of saying "a different user", just use my name. – brian d foy Mar 24 '10 at 20:08
To be clear, there's really no need for me to get rep. I edit almost every Perl question, and answer less than 25% of them. – brian d foy Mar 24 '10 at 20:09
I think @Kip was trying to discuss the general case of editing posts. – Randolpho Mar 24 '10 at 20:11
@Randolpho: it's easier to discuss a specific case. If we come to something resembling a consensus, we can always generalize later... (editing this question to match the answer, of course!) – Shog9 Mar 24 '10 at 20:14
No problems. I was failing at making a joke with that comment. – Randolpho Mar 24 '10 at 20:20
You know, you could have just asked me if I would mind changing "repeated" to "duplicate" in the title. I wouldn't have cared, really. I only changed the title because it wasn't a question. I didn't really care when Shog9 changed it. – brian d foy Mar 24 '10 at 20:53

Here's what happened:

  • I saw the question and read the answers.
  • I pointed out that some answers, including the accepted one, were limited.
  • I provided my own answer for the more general case.
  • I pointed out in my answer that I use this example in my Learning Perl book. I've actually thought quite a bit about this particular problem for the book and the classes I teach.
  • As I have done with the thousands of other Perl questions, I made the title an actual question.
  • I did not edit the body of the question, change the inputs, or change the desired outputs.
  • Despite allegations to the contrary, I wasn't trying to kick this to the front page. I answered and edited the title at essentially the same time, just like I do every morning when I go through the new Perl questions.

People don't have a problem with just the content of my answer, although they are downvoting me because they don't like that I made the comment to Kip's original, accepted answer. Eric Storm also commented with essentially the same content, and by the comment vote count, many people agreed with that. I don't think Kip's answer gives the questioner enough information to solve his potential problem, just the immediate problem. One of the biggest problems that newbies have with questions is that they don't know the scope of their problem. That's why they typically ask imprecise and short questions with very limited sample input. Answering exactly what a newbie asks leaves them with only half the stuff they need to know.

However, some people misinterpreted what I did and took offense. I don't think there is any actual argument here about the actual question, the proper answer, or the proper title. However I did things, no matter how many times I've done this before, this time people thought I was a jerk. Once they decided I was a jerk, that's how they framed everything and how they voted. That's why no one bothered to try to work within Stackoverflow to solve this problem. There never was an edit war. There was a progression of edits, each by different people, that each improved on the last edit. Since they thought I was a jerk, they thought there would be an edit war and made this pre-emptive strike.

I'm not really concerned about that. I want to give the best answer to the question, and make the question as maximally useful to other people as possible. I've been doing this for many, many years and as part of the core Perl documentation. I'm a writer, and writing is editing and considering intent, etc. I'm not particularly attached to the exact words I use for any particular question title. At the same time people question my ability to intuit the actual question, they are quite confident in their own ability to interpret my actions. Most of this seems to stem from their fear of me gaining reputation points. That is, I don't think anyone would care if I did the same exact things without being in a position to get rep.

As part of that, I'm going to remove all of my comments from the SO question. Since you are afraid of my rep, I'll also make the answer community wiki. You can fight it out here. The future public doesn't care about our squabbles; they only want the answer to their questions.

Other than that, I'm leaving the question alone. I think the current title is fine. I would have thought it was fine before this tempest in a teapot.

share|improve this answer
Now I think you're overreacting a bit, although it's understandable given the circumstances. I think the whole question arose only because you tried to generalize the question title -- however valid that may be, the fact that others had answered the original question and been accepted makes it a questionable action. Nothing else you did could be argued with. – mmyers Mar 24 '10 at 21:49
@mmyers: Except Kip also both edited the question and answered (something I do too; it shouldn't be considered bad by itself, though of course has the potential for abuse). If anything, Brian restored the original meaning, rather than generalizing. – Gnome Mar 24 '10 at 21:56
You know some people are overly focused on the game when brian's thoughtful answer gets flagged as spam. FWIW, I decided not to spend any time on that question because a less than stellar answer had already been accepted: After all the answer consisted solely of a pattern with no real explanation. Anyway, this is indeed a tempest in a teapot. It would be nice to occasionally see @Kip and others fix titles and question bodies to generate better content for SO. After all, that is what we are: Content generators so Joel and Jeff and others can earn money from ads. – Sinan Ünür Mar 24 '10 at 22:07
When you change someones question, you have to try to avoid changing it's meaning, which of course involves checking what the OP really asked for instead of just making changes to others changes. In this case it looks like you changed the meaning of the question to match your answer. Even if that wasn't what you indended, you should take better care when editing someones question. – Guffa Mar 26 '10 at 8:31

Rather than approaching this generally, let's stick to the specifics of this question.

Original title: Highlight the double Words
Kip's title: Regular expression to highlight duplicate words in a string
Brian's title: How can I match and highlight repeated consecutive words with a Perl regular expression?

Brian correctly captures the OP's intent of "double words", as I read the entire original question, rather than "duplicate words". The latter implies that for "A B ... A", A would be a duplicate. (The title change is the only edit Brian made.)

Shog9's title: How can I highlight consecutive duplicate words with a Perl regular expression?

Shog9 removes the redundant "match and" and changes "repeated consecutive" to "consecutive duplicate". This is perhaps a better wording, but it maintains the same clarification that Brian made!

Brian's comment on your question had nothing to do with his title edit, and you're conflating the situation, then taking offense. Providing a better answer to a question that has already has an accepted answer should be encouraged—especially when the OP is relatively new to the site and appears to have made a snap judgement by accepting within an hour of posting. I see no problem.

It's fair to say you misunderstood, and I'd just move on.

Regarding bumping: His answer had already bumped the question. The question edit was about 20 minutes later, so while this could conceivably be primarily for the purpose of bumping, a look at his activity page shows he was reading at least one other question and came back to edit the title. As Brian shares in a comment below, he was working on something related during that time, but he shouldn't have to account for his time. I often work on multiple things at once, as I imagine most do. (And if I couldn't, I'd never be able to use MSOFU. :P)

share|improve this answer
Ignoring the bad vibes that arose in comments, I think this situation highlights the value of editing a question to reflect your interpretation of it vs. just making your assumptions implicit in your answer. "Duplicate", "double" and "repeated" share some of the same meanings, and while the example helped to clarify the intention it's not inconceivable that a later reader might have misunderstood the question without Brian's addition of "consecutive". – Shog9 Mar 24 '10 at 21:29
Actually, during that time I was writing the bug report I mention in my answer with perl5porters:… – brian d foy Mar 24 '10 at 21:40

I don't think there's a difference. Tellingly, the input and desired output haven't changed. They are phrased differently, and the question approaches form a slightly different angle in terms of phrasing, but the intent and meaning are clear.

Of course if that wasn't/isn't the case and the changes do fundamentally alter what is being requested, then it seems a somewhat underhanded abuse, and it should be flagged for moderator attention.

share|improve this answer

I think Brian thinks that the OP was asking the wrong question to solve the problem (which is absurd given the limited context of the question).

I would roll back that title edit.

That being said, Brian probably didn't have malicious intent, but this is definitely an activity that should not be encouraged. I'd want a mod to peek at this one for sure.

Edit re my comment in there: I have no intention of editing his questions, I was just making a point.

share|improve this answer
I can't wait to see the fallout of your question-edit war with Brian.... pops some popcorn – Randolpho Mar 24 '10 at 19:51
@Randolpho: See the edit of my answer here. – Jon Seigel Mar 24 '10 at 19:52
Awwwwwwwwwwwwwwwwwwwwwwwww – Randolpho Mar 24 '10 at 19:53
Just use my name. You don't need to feel bad about pointing out that I'm the one you're arguing with. – brian d foy Mar 24 '10 at 20:10
As for your comment about editing questions, um, that's the point of Stackoverflow. There's no point to make. If you think you have a better or more clear way to phrase any of my questions, I'd like to those edits. As a long time author, I've learned not to become attached to any particular phrasing. – brian d foy Mar 24 '10 at 20:12
@brian: This was written before I started commenting in that question, but I updated this answer for you. – Jon Seigel Mar 24 '10 at 20:12
It's not that I think he was asking the wrong question, but an imprecise question based on a limited input sample. – brian d foy Mar 24 '10 at 20:13
@brian: its more meta without the names though. This question is only useful if someone encounters the situation again, and there's a small chance you may not be involved. ;) – Jimmy Mar 24 '10 at 20:15
@brian: Yes, clarity is good; however, putting words into someone's mouth is presumptuous and, quite frankly, somewhat rude. – Jon Seigel Mar 24 '10 at 20:16
@brian: If you had answered just as you did but with a note explaining that your solution was for the general case, and if you hadn't then changed the title of the question, I would have have found that perfectly acceptable. – mmyers Mar 24 '10 at 20:19
@mmyers: 10000000% agree. – Jon Seigel Mar 24 '10 at 20:20

The original question specifically asks to match "double words", so that is pretty clear.

To be really complete, an answer can contain solutions both to match double words and multiple words, but it's not correct to say that a solution that only matches double words is wrong.

I don't think that the editing of the question was malicious, but it doesn't look good either way.

share|improve this answer
The "double" was removed from the question before I got there. – brian d foy Mar 24 '10 at 20:08
Even more, it was Kip who changed "Highlight the double Words" into "Regular expression to highlight duplicate words in a string". I'd expect "duplicate words" to imply "2 or more", but maybe my English fails me...? Then, given the OP's example on how to highlight, I'd say @brian's "repeated consecutive" is better than "duplicate words". – Arjan Mar 24 '10 at 21:04
@brian: Then you were just clumsy not to look at the edit history when editing the question yourself. If you base your edits only on others edits, you can't claim to know what the OP really asked for. – Guffa Mar 26 '10 at 8:23
It's not that I didn't look at the history, just that I didn't make the edit that removed it. I don't want to be accused for more than I actually did. Again, people are assigning a lot of malice to me without reason. You are quick to attack me personally, for what personal reasons I can't say. – brian d foy May 4 '10 at 21:53

On the one hand, he did change the character of the question.

On the other hand, he has potentially made the question more generally useful.

IMHO, the latter justifies the former only if it allows the question to better serve as a source for another (I'll do things like this occasionally upon finding a question that isn't an exact duplicate of one that's just been asked, but contains answers that are applicable to both). Otherwise, it's jumping the gun at best, and mildly abusive at worst.

Obviously if the OP comes back and indicates (via an edit or a comment somewhere) that he really did need to match > 2 consecutive duplicates, then it's ok.

share|improve this answer
The ends can't justify the means. The OP may have specific data where only a single duplicate should be matched. I do agree with everything else you said. – Jon Seigel Mar 24 '10 at 19:52
@Jon: I'm referring to cases where the answer(s) to question A effectively answer both questions A and B (say the question "how can I print a number in any base between 2 and 52" gets an answer that demonstrates how to print a number in an arbitrary base not limited to <=52 - this would then answer a future question regarding printing in bases 2-64, etc) – Shog9 Mar 24 '10 at 20:00
Agreed. That wasn't the case for this question, though, where the edit specified that a more general solution was required. – Jon Seigel Mar 24 '10 at 20:06
@Jon: yup. That's why I edited it back to the prior requirement... – Shog9 Mar 24 '10 at 20:07

You must log in to answer this question.

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