Occasionally I have done a search like this:

smirnov -[kolmogorov-smirnov]

to locate posts that mention a word that's likely connected to the tag, but doesn't have the tag.

The second post that it turned up, for example, appears to be a possible candidate for having the tag added.

I do this for a variety of reasons, but one is simply to add tags to posts where I think they need it so people can find them more readily, for example. (More selfishly it helps me locate posts I feel like answering at the time that I won't have found in previous forays with the tag.)

I was just doing this kind of search with the histogram tag now, and added a few (not every post that mentions the word needs the tag however and I am sometimes left to ponder when it's reasonable to add and when it isn't. This post is not about that quandary).

It occurred to me that maybe there's some issue with specifically searching for missing tags and (sometimes) adding them in this manner.

It's not something I do all the time, just now and then as the mood takes me (usually after noticing what seems to me to be an important but missing tag on a post).

So is there anything out of order with that kind of behavior? Is it a good idea? A mix?



A big thanks to AndyW and gung for their insightful, helpful, high quality analysis, and useful links.

By way of summary so far -

  • there's a problem with doing it for a single tag on a large number of posts without consultation

  • even with a smaller number of posts, there's a problem with doing it on more than a couple at a time in quick succession because of the impact on the front page - so it should be well 'spread out' in time.


1 Answer 1


This is IMO totally reasonable. In fact, I've seen other sites organize tag clean-ups to try to get the community involved in similar behavior.

About the only thing to be concerned with is mass retagging without community approval. Users are just expected to use their best judgement about the boundaries between mass retagging and cleaning up a few questions.

  • 2
    $\begingroup$ Thanks for the pointer there; that information pretty much speaks to my concerns - and makes it clear that a lot of it could be a problem, and the reasons why - and explains what to do (which is talk about it first if I had plans to do a lot). I don't think it is likely to approach those levels of retagging, but it gives me a somewhat better idea of how much to do it. It seems to me that doing 5-10 tag-adds with pre-existing tags should almost never be an issue, but doing say 30 without notice or discussion could be. $\endgroup$
    – Glen_b
    Commented Jun 7, 2013 at 12:38
  • 1
    $\begingroup$ I fully agree w/ everything here. I also look for posts that I know are about [such-and-such] (often my own), can't find them, ultimately do, & then re-tag them. I had read the linked meta thread long ago & agree w/ what's there as well. It would be inappropriate to do 50 at once, even if it's not to get a badge, because it floods the main page. I do bump old threads by editing (tags or otherwise) occasionally, but try to do only 1 at a time, & when traffic is slow (see also here: feature-to-edit-without-bumping). $\endgroup$ Commented Jun 7, 2013 at 19:57
  • 1
    $\begingroup$ @Glen_b, 5-10 may be a bit much, IMO. Most of the attention on this site goes to the main page (for indirect evidence of this, see AndyW's recent blog post: Voting behavior & accumulation of old votes on CV). I suspect even most of that attention goes only to the top ~10 questions on the main page. I suspect it's best to do only a couple, unless activity is pretty dead. $\endgroup$ Commented Jun 7, 2013 at 20:04
  • $\begingroup$ @gung I generally only do this only after I have dealt with every single question on the main page where I can make any contribution to at all - whether by answer, comment, edit, vote, flagging etc, that have appeared since the last time I read the main page - looking at old posts is something I do after everything else. ... And if old posts aren't regarded as something to spend time on why do several badges seem to encourage people to deal with old posts? (If I was after those badges, my approach would be different, but SO/SE plainly encourages working on old posts). $\endgroup$
    – Glen_b
    Commented Jun 8, 2013 at 0:11
  • $\begingroup$ @gung I'm not sure I actually see the effect AndyW describes as a particular problem, among other things for the fairly obvious reason that - because of the way new questions that are very like existing questions are closed - the most interesting questions are always going to tend to be asked early. Old questions on average should get more votes as new people join because they're often the most interesting, relevant, best edited and often best answered. It's not a problem in my mind, it's a feature. $\endgroup$
    – Glen_b
    Commented Jun 8, 2013 at 0:16
  • 1
    $\begingroup$ @Glen_b, I didn't mean to suggest that it was a problem. Re the blog post, what I was referring to is the fact that most of the votes come on the 1st day of a question, ie, when the thread is on the main page. My point was that bumping things off the main page can have a big effect. I don't have a problem w/ spending time on old posts--I do it myself. I just only deal w/ 1 or 2, or I do it when there's less overall activity. To put this clearly, I don't have a problem w/ your activities on the site. $\endgroup$ Commented Jun 8, 2013 at 2:04
  • $\begingroup$ @gung Thanks for all your responses; each is useful. I definitely agree that there could be an issue if I were to bump a bunch of old posts in quick succession (though on the other hand 'Community' is constantly touching old posts to bring them back into the active queue, so I think the 'in quick succession' would be the main potential problem there if the overall volume isn't large). It's not something I'd really considered (the effect on the active queue that is). $\endgroup$
    – Glen_b
    Commented Jun 8, 2013 at 2:10
  • $\begingroup$ @Glen_b, you're right, it's the "in quick succession" that is the potential problem. The 'community user' bumps 1 old post (>1 month) w/o an upvoted answer per hour in the hopes that it will finally be answered. $\endgroup$ Commented Jun 8, 2013 at 2:16
  • 3
    $\begingroup$ Retagging en masse can reflect abuse (although it is not necessarily so). For more about this, please review our last experience with this problem. $\endgroup$
    – whuber Mod
    Commented Jun 8, 2013 at 17:56
  • $\begingroup$ Personal opinion I would say 5-10 is ok, much more than that it is IMO getting into en masse territory. Note the prior problem was both the user created a tag (that was not without dispute in and of itself) and the user did it on 30 some questions. If he just applied an already existing tag to historical questions I doubt the incident would have been so noteworthy. $\endgroup$
    – Andy W
    Commented Jun 9, 2013 at 18:19
  • 2
    $\begingroup$ IMO = in my opinion $\endgroup$
    – Jay
    Commented Jun 9, 2013 at 22:30

You must log in to answer this question.

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