I like to edit my answers. To clarify, illuminate, add examples, add links, point to new information as it crops up, etc.

It's highly irritating that my effort to improve my answer turns it into a Community Wiki.

Could anything be done to keep me from turning into a collective?

  • 11
    Heh... 9 edits, the last 6 each made within a half-hour of the previous edit (one 6 minutes after the previous... two minutes faster and it'd still be yours). You should really consider either editing a lot faster, or a whole lot slower! – Shog9 Jul 24 '09 at 23:27
  • Yeah. Clearly my timing sucks. – Nosredna Jul 24 '09 at 23:52
  • 4
    I've decided to deliberately not care when my edits push me CW. – Nosredna Jul 24 '09 at 23:54
  • 3
    I'm going to force this post into Community Wiki :) – Jarrod Dixon Jul 25 '09 at 3:24
  • 28
    You don't have to. I'll eventually do it by accident. – Nosredna Jul 25 '09 at 3:43
  • 3
    And I just did this on SU.....sigh. – Troggy Aug 4 '09 at 21:49
  • Related: meta.stackexchange.com/questions/46078/… – Jon Seigel Apr 27 '10 at 17:04
  • it also makes sense, since you're likely editting your answer because of comments from the community – seanmonstar Apr 27 '10 at 17:58
  • 1
    Is some circles, it is called the "community wiki of death". FYI. – MPelletier Apr 28 '10 at 22:33
  • 1
    I only found out about this after my edits turned my post into community wiki. I understand the reason but where's the trust for established users? – Reinstate Monica - Goodbye SE Sep 13 '10 at 17:27
  • 1
    Four years and no one has replied to the last paragraph with "Resistance is futile"? – Tobias Kienzler Aug 8 '13 at 13:07
  • This no longer applies. – Peter Mortensen Oct 25 '17 at 22:22

12 Answers 12


I concur with this one; edits from the original author shouldn't count towards community-wiki-ness.

  • 4
    Utterly disagree. The idea is to stop people bumping their question to the front page constantly, and gaining rep via this. Auto CWiki'ing is a reasonable disincentive to this – dbr Jul 25 '09 at 12:29
  • 64
    Then, surely, the question shouldn't get bumped to the front page as the result of an edit. – moobaa Jul 25 '09 at 23:45
  • 22
    @dbr then address the issue of bumping directly instead of using a weird side effect penalty. Two ways to solve this are only re-bumping if a significant time has passed since the last edit, or simply limit the # of times it can be bumped. Or, remove the bumping entirely (which I don't like, often legitimate edits make a question answerable) – TM. Aug 27 '09 at 16:52
  • What if there are no other answers at the time of the edit? How can it possibly be unfair to anyone if those edits did not count towards wiki-nature? – David J. Liszewski Jan 20 '11 at 20:22

Someone just pointed out in a comment that this post has been CW'ed. Now I'm not so much of a rep whore that this really upsets me, but it seems counterintuitive, as it effectively punishes me for making my answers better. In this case, it seems a bit unfortunate that I apparently disowned an answer with 87 upvotes. (my third-most repped answer)

I understand the rationale for turning the post CW after X edits, but it seems a lot of good answers get caught in the crossfire.

A couple of possible modifications to the rules could be:

  • After the first N edits, instead of turning the post CW, just stop bumping it when it gets edited. I generally edit my posts when I become aware of errors or flaws in the answer. I don't care about getting the post bumped, I just want to make my answer better. So you could just stop bumping them once they reach the N-edits threshold.
  • Use some kind of heuristic to determine whether this is "an attempt to bump the answer", or of it is a genuine edit. A lot of edits in a short time (say within an hour or two) seems like genuine improvements. And if a really long time passes between edits, people probably aren't trying to game the system by bumping their posts. Most likely, they received a comment or similar, which made them re-read their answer and update it. But if a post is updated every few days, or every week, it might be suspicious. Of course getting such a heuristic right might be tricky,but at least you've got plenty of test data to experiment with. ;)
  • Look at how much previous edits have bumped the question. Editing a question which is already on page 1 of the Recent list probably isn't an attempt at getting it bumped. Editing one which has fallen further behind is more suspicious, so that might count towards CW-ificiation. Or perhaps look at how much time has passed since the last activity in that thread. If someone else edited another answer 5 minutes ago, which bumped the thread, then that makes it likely that I simply rediscovered the thread, read my answer, and decided to make a few improvements. And the effect of my edit is almost nil, since the thread was already bumped those 5 minutes earlier.
  • Or perhaps simply add a checkbox when editing: "Should this cause the answer to be bumped?" iF the user checks this, the thread gets bumped, and he gets one strike against his CW limit. Keep doing this and it gets CW'ed. If he unchecks it, the edit is made, but the thread is not bumped, and it doesn't count towards CW-ification.

Just a few ideas. This seems like one of the few cases where SO actually discourages positive behavior. I tend to edit my posts quite a lot, and I think that's a good thing. My answers are better than they would be otherwise.

  • 1
    Except for the first one these are all really great ideas! – Oliver Giesen Aug 23 '09 at 16:16
  • 10
    +1 Especially the last bullet point. Offer a choice between bumping & increasing the CW strike count, or not bumping. – MarkJ Apr 27 '10 at 12:51
  • 3
    Just FYI, apparently you can flag your answer to ask a moderator to un-CW it. Just in case you're interested, and haven't done it yet. – Justin Morgan Feb 13 '13 at 22:10

The reason for having the questions/answers automatically going into Comunity Wiki status, was to prevent people from editing Their question/answer just to bump the question to the first page.

The only thing is, it doesn't really solve the problem, you could just go edit someone else's answer, and it would bump the question up. Or add another answer to the page, perhaps with a dummy account, and have it edit it's own answer repeatedly.

I personally would prefer it not bumping, to it becoming a Community Wiki.

If anybody wants to know where this rule came from, it was discussed on podcast #20. I have included the relevant portion of the transcript.

Atwood: -

Somebody wrote a bot that would just revise posts every minute to keep it on the top of the stack. Actually there is a certain amount of people doing that still, which I'm trying to discourage. They way I like to discourage things is, where possible, creating rules in the system that make that behavior not desirable. Not negative, necessarily, but things happen that make it not worth much to you.

So let me give you a specific example in that scenario. You have the user who's just editing their own posts every three hours so that it's always on the top of the stack. We have this concept now, it's actually implemented though I talked about it in previous podcasts, of the community owned post. Because one of the great divides in StackOverflow is that we have this ownership system where you get voted up and down, your content gets voted up and down, and that affects your reputation. You own stuff, so when you post something you own it. Then you contrast that with the Wikipedia model which is that nobody seems to own it, and we're trying to do both of those things.

At the transition point we came up with a couple rules. The initial rule I had was that edits by four different people will cause a post to switch from being owned by Joel, for example, to being owned by the community user. At that point you don't lose any reputation that you got up to that point, but any future upvotes on that content don't go to anybody they go to the content. I think this is the way it should be. Ultimately you're voting on the content more than the person anyway, so hopefully people are ok with this.

Seeing that people kept editing stuff over and over I bent the rules a little bit and said ok, if you edit your own thing more than N times then it also becomes a community owned post. There is no real value to the user, in terms of getting additional reputation, to bump stuff up to the top of the stack anymore because if you edit your own thing enough you won't get any reputation from it. It behooves you to only edit it once, or however many times you need to edit it but hopefully no more than once, and just let it sit there and have people find it organically and naturally the way it is supposed to happen.

  • That is basically what I just wrote... – jjnguy Jul 23 '09 at 22:20
  • You can only bump using other people's posts if you have over 2k rep...which the vast majority don't. – jjnguy Jul 23 '09 at 22:20
  • 18
    +1 for limiting bumping (over auto-wiki-convert) – Robert Cartaino Jul 24 '09 at 1:14
  • 9
    Why not just stop bumping minor edits... say, less than 20 characters changed. That allows for typo corrections, while still having an effect for material changes to a question. – Lawrence Dol Jan 20 '10 at 7:42
  • @Soft you can put a lot of spam in 20 characters. – Earlz Apr 28 '10 at 19:31
  • 3
    @Earlz: What does that have to do with bumping the question, community wiki, or anything else in this post? – Lawrence Dol Apr 29 '10 at 13:55
  • @Earlz What spam can you fit into 20 characters? What does spam have to do with whether or not a post gets bumped? – endolith Jun 6 '11 at 20:23
  • @end @soft in 20 characters, you can replace two link references with bit.ly links. Looking back on this now. Maybe just not bump unless there is a link changed or added? – Earlz Jun 6 '11 at 21:16

I edit compulsively. I doubt there's any post over a few paragraphs in length that I haven't edited at least a half-dozen times.

And yet, I can't remember a single instance where I've accidentally triggered CW mode. Indeed, before SO put the checkbox on the edit screen, I struggled at times to force answers I wanted CW into CW mode!

See, I edit in spurts. A big change, followed by several small changes as I proof-read. And all that happens within the 5-minute grace period, so most of them don't even show up in revision history (much less the more lenient CW trigger).

So, my advice then, is... save up your edits. Make 'em count. Open a text editor, and make your changes while you're researching, then submit them en-mass. Not only does this reduce the chance that you'll inadvertently trigger CW, it provides those reading the front page with more of a reward for clicking through to see what changed when the question gets bumped by your edit...

  • 27
    I've hit it by accident. At least a warning would be nice so I'd know how close I was getting. – Nosredna Jul 24 '09 at 0:53
  • I think it took me about 8 or 9 edits to turn one of my answers to wiki. I'm not sure what the threshold is supposed to be. – gnostradamus Jul 24 '09 at 2:44
  • Yeah... I've wanted to make a new CW-faq for a while now, but really have no clue what the exact behavior is anymore. It's plenty hard to trigger accidentally with my workflow, that's for sure. – Shog9 Jul 24 '09 at 3:12
  • 5
    I was a magazine editor. I can edit an answer as many times as I can edit a code file. :-) – Nosredna Jul 24 '09 at 4:31
  • 2
    My advice is pretty much identical to this: rather than editing a hundred times for tiny stuff, make 4-5 edits that really matter. It's a saner behavior for the entire system. – Jeff Atwood Apr 27 '10 at 9:00
  • 9
    What a bad answer, why not just say "work around the system's bugs"? – Evan Carroll Apr 27 '10 at 14:21
  • 5
    @Evan: because that doesn't actually offer anything to anyone. Such a trite answer dismisses the problem and mocks the asker, implying that the former was well-known and the latter already had a solution but was unwilling to make use of it... It brings to mind the "just use jQuery" answers posted to Prototype questions, or the "just use Windows" answers on Linux posts - rude and unnecessary. Small wonder then that your answer similarly proposes discarding the entire platform instead of attempting to make it work... – Shog9 Apr 28 '10 at 1:34

I edit like crazy. If my answer reaches the top of the list, I edit even more. There is always something I forgot to add, something that could have been said better, a sentence best followed by a code example or refinements based on comments.

While I usually refrain from editing other people's posts unless some glaring error remains after leaving a comment to point it out, I edit my own religiously. I've been nailed by the same thing a few times, actually I became rather upset the first time. I'm at 9k something now, I'd be well over 10k if I actually earned the rep from every up vote.

I really don't care about rep anymore. Well, sure I do care, but not to the point that I won't correct or add something to make my answer better, even if it means not seeing a number on a web page increment.

Due to having my site in my profile, I've actually received email regarding some of my answers, not just comments, especially on controversial topics. Make edits when you need to correct or augment something, even if it means your score at the top of the page doesn't increment as much as it would. The answer is what matters, if you care enough to edit in the first place.

Besides, you still earn badges.


Kaveh brought up some valid points, points I've been considering for a while during some thoughts I've had on community wiki.

  1. it doesn't seem to be very effective in preventing too many small edits,
  2. it is abusing a feature which seems to have been designed with a different goal and using it as a punishment,
  3. it has side-effects like reasonable edits over reasonable periods of time causing a post to become CW.

1 isn't as much one I follow as much as, it didn't do much about too many small edits. It was just "Wiki, we're done here", and that's it. No one is told that it happened, no one is told why it happened, and no one is told what shouldn't have happened. The other two, especially the second, are very much on point about the issue.

It is my pleasure to announce that we will no longer automatically turn your edited answers to community wiki. Or any posts at all - all of the automatic triggers in the system have been removed. In their place, we have setup a number of alert flags that are given to moderators in the event of too much revision activity or too much answer activity. The point of these will be to help identify when their is abuse, and have it handled much more appropriately.

  • So that's what you were doing all this time! :-) – Shadow9 Apr 9 '14 at 20:59
  • @Shadow Only one part of it. There will be more to come. ♪ – Grace Note Apr 9 '14 at 21:03
  • I'll take your word for that! Any hint of what is coming? :) – Shadow9 Apr 9 '14 at 21:22

My proposal is this:

  1. Increase the amount of times the author can edit a post before making it CW. I suggest making this 8 or 10.

  2. Edits that are done by the same user(including author) within 5-10 minutes(to taste) within each other are not counted and will not rebump the question.

  3. The first 15 minutes a question/answer is posted, none of the edits count.

1 will help with improving answers/questions as more research is done. 2 will fix the minor-edit problem for whenever you make an edit and then discover you misspelled something, and then after discover your formatting is a bit wrong. 3 will fix the problem for when someone posts a poorly formatted question on a popular topic and 6 people edit it at the same time.


The reason this is in place, if I remember correctly, is to prevent people form endlessly bumping their posts with edits.

I agree that it is a pain though.

  • 16
    So it could be fixed by not bumping me up if I edit too much. – Nosredna Jul 23 '09 at 22:07
  • 5
    I agree with that suggestion. After a certain number of edits, the post should just stop bumping. – jjnguy Jul 23 '09 at 22:09
  • 10
    I would prefer it not bump me up, than for it to go into CW status. – Brad Gilbert Jul 23 '09 at 22:09

Not bumping through edits would prevent sediment churn, which means old (and possibly wrong) information would be left to fester and die in the bowels of the system -- unseen and unloved. That's much more toxic to our community than someone sacrificing a few reputation points.

Our CW threshold is quite generous; I suggest adapting your workflow to do fewer, larger edits instead.

Remember: getting good information out into the world, and getting answers, are the real goal, not arbitrary points -- so we have to favor processes that further that goal.

Also, being a hybrid system, we have aspects of wiki (information-first) and traditional bulletin board / blog (owner-first), but there's always a tension in balancing those aspects of the system.

  • 3
    But what if you opted for what seems to me to the best of both worlds? I would suggest you allow editing to bump up until the point where it meets the current criteria for being converted to CW, then it stops bumping but does not convert to CW. Perhaps that flag could then clear again after a set amount of time, although I don't know if that's necessary. – Gregyski Aug 22 '09 at 16:37
  • 1
    Could you instead just limit the # of edits that create a bump? Or put in some sort of rate limiting? – TM. Aug 27 '09 at 16:54
  • 15
    Jeff, I'm a bit surprised that you haven't implemented a rate limiter for this. You're already doing this for votes and other parts of the system; but churning a detailed post into a CW that's permalocked is a bit harsh to the poster; it's like saying "we're rewarding your hard-won effort with a slap". A rate limiter would solve the issue entirely and prevent the kind of question/answer churning that causes bumpage. If you have delays between each edit, or exceed a number of edits in a given timeframe, there is a cool-down that would be required before you could save the edit. Simple. – Avery Payne Sep 5 '09 at 16:17
  • 11
    The same person editing a single post can turn it into CW? I think that's ridiculous. What about a "single" person coincides with "community"? – sliderhouserules Jan 14 '10 at 8:14
  • 2
    This idea breaks down on SF and SU where a user is troubleshooting an issue. The process you recommended was for the owner to edit the original question as more troubleshooting is completed. Doing that eventually causes the owner to change though. – dlux Apr 27 '10 at 20:24
  • 1
    @dlux I disagree -- we've done a ridiculous amount of troubleshooting / editing in this question and it isn't wiki yet. serverfault.com/questions/104791/… – Jeff Atwood Apr 27 '10 at 23:35
  • 6
    Could you please add a warning when this is about to happen? I had no idea editing your own answer to clarify some points would turn it into Community Wiki. I wrote a script, which was the original answer, then I made updates with notes and examples of the script's usage. I don't feel like those kinds of edits should push someone's answer into a Community Wiki status. – Christopher Parker Jun 9 '10 at 21:23
  • 4
    Also, to rebut your comment about "arbitrary points". I would typically agree with that statement, however much of the site's features depend on accumulation of these "arbitrary points". Good answers will earn points naturally based on their merits. Why punish those authors who make meaningful edits in order to make their answers better? Shouldn't you want to encourage this behavior, and reward it with more site features? – Christopher Parker Jun 9 '10 at 21:31
  • 5
    This seems very much at odds with your usual policy of encouraging good behavior. Usually, rep is not "arbitrary", but is an important carrot to encourage us to make the best possible Q&A site. Here, someone has pointed out a real problem in the rules, which discourages good behavior (I want to improve my answer, but I don't dare do it after the first couple of edits, because I'm tired of my answers being CW'ed). It may not be a problem for you, with your workflow, but for some of your users it is a problem. Are you saying those users should just stop improving their answers? – jalf Oct 14 '10 at 12:56
  • 4
    How about, instead of just going by your experience, look at the data (as you've done in so many other cases). How many answers are "auto-CW"'ed due to excessive edits by the author? If you sample a few random ones, do any of them seem like attempts to game the system? Do any of them seem like "bad" edits that should be discouraged? If not, why does the system currently in place discourage it? There are so many possible improvements that could be made to the system, to both prevent people exploiting edits to bump their answers, and also avoid punishing people for improving their answer – jalf Oct 14 '10 at 12:57
  • 1
    I'm surprised you propagate this as a means against unseen “old (and possibly wrong) information”, while in fact you discourage people from updating and possibly correcting that information. This doesn't make a lot of sense to me. If you want “sediment churn”, then don't punish the ones doing the churning. – MvG Oct 5 '12 at 19:47
  • 1
    Why not just stop bumping after ten edits? Or stop bumping from that specific user's edits? The auto-CW trigger seems needlessly punitive, since the user's behavior isn't known for sure to be undesirable. – Justin Morgan Feb 13 '13 at 22:47
  • As it stands taking time to research and make good answer will not give the most reputation. Why? because people come by, see a semi decent answer, up vote it (even if it wrong, but it appears to be good at first glance) and then they are gone for good. Because of that the initial answers my get three times more reputation than one that comes two minutes letter that better, more complete, or just plain correct. A similar problem happens with down votes to questions. They come, see there is a problem, down vote and are gone for good. Even if the problems are fixed, they don't come back. – Theraot Dec 17 '13 at 12:48

I just accidently triggered the auto community wiki by editing my answer to this question:

What should I know to select a power supply?

I just wanted to point out other good points from other answers even though mine was accepted as the answer. I guess i should have just used comments. And you cannot even rollback. I guess I know from now on and there is no way to reverse it.

  • Yeah, those would have made a lot more sense as comments... Oh well. Now that you've made it CW, i can edit in links to the answers you're responding to... ;-) – Shog9 Aug 4 '09 at 21:59
  • 1
    Too bad. You lose. :-) – Nosredna Aug 5 '09 at 1:14
  • 1
    i did lose – Troggy Aug 5 '09 at 4:41
  • on the contrary, you have won knowledge .. of how the system works! – Jeff Atwood Aug 7 '09 at 9:14
  • 16
    @Jeff: Honestly, I could strangle you for these snotty comments on this particular topic. I am confident that you will "see the light" on this one eventually, like you did with meta... though it totally escapes how you cannot have groked this already... :( – Oliver Giesen Aug 23 '09 at 16:22
  • 1
    @Oliver.. and your comment will be deleted in 3....2.... – Earlz Apr 27 '10 at 16:50
  • 1
    @OliverGiesen: Maybe he doesn't want to make this change to the source code because he's already edited it 9 times... – endolith Nov 23 '11 at 18:19

I was wondering, now that CW for questions has become a mod privilege shouldn't we also update this automatic triggering?

I understand the problem that Jeff refers to but it seems to me that it is not a good way of preventing too many edits:

  1. it doesn't seem to be very effective in preventing too many small edits,
  2. it is abusing a feature which seems to have been designed with a different goal and using it as a punishment,
  3. it has side-effects like reasonable edits over reasonable periods of time causing a post to become CW.

Can't we come up with some other way of discouraging too many small edits? E.g. how about limiting the number of edits a user can perform on his own posts over some period? in a way similar to limits on the number of questions a user can post?

  • 2
    This should be a question, not an answer. Ask a new question referring to this one and asking to have the debate reopened. You might also choose to strengthen your argument with some research and statistics (using data) – JDB still remembers Monica Oct 2 '13 at 19:17

I think we should continue to leave the CW bumping. The system works fine as it is.

I have seen quite a few instances where people abuse the system with editing and bumping, and if the system didn't automatically change the post to CW, they would gain a lot more reputation for this abusive behavior.

  • 2
    So, in other words. If it didn't bump the page, they would have no incentive to do that? – Brad Gilbert Jul 24 '09 at 1:25
  • No incentive to do what precisely? – GEOCHET Jul 24 '09 at 1:41
  • Make it auto CW – Brad Gilbert Jul 24 '09 at 22:39
  • 2
    Are you sure? I mean if I intentionally bump a really bad question why should this automatically mean that anyone would upvote it? Just because it's the first item on the front page? – Oliver Giesen Aug 23 '09 at 16:19

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