23

Ever since suggested edits were rolled out, the Retag privilege has been a bit... Weird. You can retag questions at any rep-level, even with no rep (or account) at all - you just need a few people to approve your changes. Then suddenly, you don't - but any other changes you make to a post still goes into a queue for approval.

We recently made this a little bit less weird by rolling it into the same UI as other edits - but this just leaves the question of whether or not this is really needed at all.

As animuson notes,

This has confused users for too long. The only reason it existed was because users couldn't suggest edits once upon a time, and retagging was a legitimate privilege that let them edit the tags only. Now that users can suggest edits and access the edit screen, having separate functions seems incredibly pointless, and as stated, the system knows when a tag-only edit is made. It's time to say goodbye.

Reasons to remove it

Reasons to keep it

  • It can motivate users without the reputation to perform full edits to perform more retags
  • It can reduce the load on suggested edit reviewers somewhat
  • It's still somewhat useful on sites (child metas) where suggested edits aren't enabled

Bullet #1 needs some evidence. Here's a graph of retags by reputation level (in 100-rep increments) over the past 90 days on Stack Overflow, and here's a graph of all edits excluding retags by rep level.

It's not huge (and it doesn't last) but there is a noticeable bump in activity from folks who've just gained the privilege. However, it's worth noting that this bump is far less pronounced since we merged retag and edit (graph) - indeed, by simplifying the interface, we may well have removed the thing that motivated people to use it in the first place.

So, is this worth the added complexity?

11
  • 5
    Is it an option to only keep it for child metas where suggested edits aren't an alternative?
    – Tim Stone
    Commented Jun 1, 2013 at 1:18
  • Anecdotally, I've rarely used the privilege, if only because if I'm invested enough in a site to get 500 reps, 2000 aren't too far off. But admittedly I have a low sample size to base this on.
    – Tim Stone
    Commented Jun 1, 2013 at 1:25
  • Like @TimStone, I've rarely used it too, but for me it's because if the tags are wrong, it's rarely the only thing. Perhaps tag-only edits should be restricted to 2k (or even 10k, with the inline editing).
    – Kevin
    Commented Jun 1, 2013 at 2:15
  • 3
    Also, with the title of the post "Do we still need the Retag privilege?" I don't know whether to vote up or down to show my support for eliminating the privilege.
    – Kevin
    Commented Jun 1, 2013 at 2:16
  • 1
    @Kevin You can vote in the traditional sense: "This question shows research effort; it is useful and clear."
    – Antony
    Commented Jun 1, 2013 at 6:40
  • 1
    It might not be huge but it's still 750 questions that someone else doesn't have to re-tag... assuming that it's been done well. Can you change your chart to the number of posts which were retagged and the tags were not subsequently changed within, say, 48 hours. If this number is still 750 then it's worth keeping the privilege, 273k questions each year have been given the "proper" tags. If the number is 20 then it's probably not worth keeping. Commented Jun 1, 2013 at 9:29
  • 2
    Child metas should just allow suggested edits (and display the counters on the main site as well). Commented Jun 1, 2013 at 11:37
  • Yes! I made something a little less weird!
    – Undo
    Commented Jun 1, 2013 at 14:48
  • 1
    The results aren't dramatically different, @ben. The graph flattens out slightly, but there's still a bump at 500 rep and it's still a 700-some increase over the past 90 days. FWIW, the bump amounts to 3K+ retags over the past year, but is practically non-existent since the May 8th change - it's possible that by rolling this into "edit" we've already removed the thing that was motivating folks gaining this privilege.
    – Shog9
    Commented Jun 1, 2013 at 16:20
  • 1
    If it's 3k/year maximum posts that are being "improved" then I really don't see the point in maintaining the code necessary to keep this running @Shog9. But then you already knew that didn't you :-)? Hence the question... Commented Jun 1, 2013 at 22:44
  • @Shog Is this all my fault?
    – Undo
    Commented Jun 30, 2013 at 21:02

6 Answers 6

19

Yes, the retag privilege is useful on medium-to-heavy-traffic sites. It is useful because retagging allows a question to be directed towards the people who are best suited to understanding it, possibly editing it into shape (when that requires knowledge of the content matter and not just of English), answering it or closing it.

I don't think the retag privilege is useful on low-traffic site where many users read all the questions. It probably doesn't help so much on Stack Overflow now that suggested edits are reviewed very quickly. On intermediate sites, where many users filter questions on tags, it's very useful to have questions' most important tags sorted out as quickly as possible.

To give a concrete example: when I joined Stack Exchange, I started participating on Super User. I gave up after a few months. My primary reason (not the only one, but the most decisive one) was that I got to see an awful lot of posts that I didn't care about. I'm a Unix guy, so I don't care about all of these Windows questions. But I wasted a lot of time sifting through many Windows questions that didn't have tags that allowed me to ignore them. Situations like these need more retaggers, and quicker retags. Removing the retag privilege only makes the situation worse.

The user interface was recently made very confusing when retagging and suggested edits were merged. Retags now have exactly the same user interface as suggested edits, which in itself isn't a bad thing. What is bad is that there is an incorrect explanatory text that tells the user that his edits will be peer-reviewed (no, that isn't the case for a tag-only edit), and also that the user must enter an edit summary which is then scrupulously ignored.

There is a simple fix to the UX confusion, which is to make it less confusing. Tell users with the retag privilege that tag-only edits will take effect immediately, and don't require an edit summary.

Of the reasons against you cite, we've seen that #1 (unnecessary) and #2 (confusing) are wrong. #3 is no reason at all: just fix the bugs. Of the reasons for, #1 (motivation) is corroborated by the data you give for SO, #2 (load on reviewers) is not critical but still nice, and #3 is also a good one (there are plenty of metas where all the questions are tagged [discussion], and what do you mean you can have more than one tag?) — though admittedly suggested edits would do in that case, if only they were possible. But the most important reason for is that retags are useful.

Keep it.

6
  • See my edit - the motivation bump seems to be drying up with the removal of a separate "retag" link on each question. It's still there, but it's very small. It'd be great if we could think of a way to encourage folks to edit more without creating a frustrating user interface.
    – Shog9
    Commented Jun 27, 2013 at 21:58
  • 4
    “The motivation bump seems to be drying up with the removal of a separate "retag" link on each question.” Really, you don't see the obvious solution? Commented Jun 27, 2013 at 22:00
  • 2
    @Shog9 I find it much more frustrating to go through the normal edit UI, than having the separate link, to be honest. For instance, as mentioned, the extra link never required an edit reason - the combined UI nags you into one, even though all you did was "edited tags" - and that's all that actually shows up.
    – user98085
    Commented Jun 28, 2013 at 0:28
  • @Gilles: that depends on the problem we're trying to solve. Retagging as a sort of training wheels for editors was (along with Community Wiki) a good idea - give folks access to some of the tools early, in a context where they can't do too much damage. But that strategy is largely superseded by suggested edits, which get folks editing - and retagging - much, much earlier. When you had your lousy experience on SU, the earlier system was still in effect - but obviously, it wasn't enough. If the goal is to encourage more retagging - lots more retagging - then let's think about that problem.
    – Shog9
    Commented Jun 28, 2013 at 2:45
  • 1
    @shog9 would releasing the awesome 10k inline retag privilege work? Or is there spme reason for if being an elite-only feature? I've always found that featire to be really useful and I don't see much scope for abuse. (I can post this on the new-500-privilege meta post if there are no major objections) Commented Jul 2, 2013 at 23:24
  • 1
    @Shog9: Retagging had a value in itself - not only "as a training wheel". Commented Jul 4, 2013 at 8:29
6

I don't see the need for the privilege.

The only place it might be useful is very early on in public betas, but generally I don't think it's really necessary there as typically several people pretty quickly (first 2 weeks usually) amass 1000 rep necessary to approve the edits.

Typically a question needs more than just a retag and only allowing someone to retag without approval discourages them from making other needed edits. The privilege as currently formed actually encourages behavior we don't want rather than encouraging proper behavior.

5
  • The privilege encourages behavior that we do want, which is to retag questions so that they at least reach the right users. What's next? Reject suggested edits that don't solve world hunger? Commented Jun 27, 2013 at 21:24
  • keep in mind that while there are enough people to review suggested edits in public beta, retagging on their metas is impossible unless you happen to be a mod.
    – Amelia
    Commented Jun 27, 2013 at 22:29
  • @Gilles behavior we don't want -> retag without actually fixing any other issue with the post. Retagging is great, only retagging but ignoring other issues not so much.
    – wax eagle
    Commented Jun 28, 2013 at 2:52
  • @waxeagle I'm perfectly fine with retagging without fixing other issues. It helps because at least this way the question reaches the right audience. Sure, fixing all the issues is best, but fixing the tags is better than doing nothing. Commented Jun 28, 2013 at 9:33
  • @waxeagle And not editing anything rather than "just" editing tags is better, yes? A mere tag edit pushes the post back to the front page and enables better filtering. Of course we should encourage people to fix the entire post if they can. But there is no harm in letting them fix minor parts of the post, if they can.
    – user98085
    Commented Jun 28, 2013 at 9:35
6

I can easily strike out two of the reasons for keeping it:

It can reduce the load on suggested edit reviewers somewhat

While that may be the case, the current load on suggested edit reviewers is very manageable, unlike the load on close vote reviewers.

It's still somewhat useful on sites (child metas) where suggested edits aren't enabled

I prefer having suggested edits on all sites anyway, since bad edits by newer users are always possible.

1
  • Being unable to edit a question's blatantly incorrect tags on a child meta that is still in beta is painful to watch. especially when there are no actual tags yet, and you are trying to organise everything.
    – Amelia
    Commented Jun 27, 2013 at 22:27
4

I've only been an active member after the suggested edit was introduced.

I've always felt it was a very strange privilege. I've only used it once or twice to the best of my knowledge, the rest of the time, any retag was combined with other edits. In general, most posts need more work than just a retag, so the post will need to go to the review queue anyway.

I'd say ditch or replace the privilege.

2

Another reason why the original retag priviledge was useful: retag only edits get rejected as being too minor.

If you want people to maintain the tags (that is if you value the tags), you should make it easy for them to do just that.

3
  • 1
    For what it's worth, I would have approved such edit. I understand the frustration, but nothing we can do now and you'll get new privilege soon. :) Commented Jul 4, 2013 at 12:08
  • 1
    @ShaWizDowArd I'm hoping for this one but I'm not holding my breath. Commented Jul 4, 2013 at 17:01
  • 1
    @Gilles it was removed after much discussion, doubt they'll undo it. Commented Jul 4, 2013 at 18:00
0

I mentioned some of this in a comment, but I think I'll make it a full answer.

I don't recall ever having used the retag (only) privilege. I'm not saying I never did, but I don't recall any and it would have been a very rare occurrence. The reason is that it is exceedingly rare for tags to be the only thing worth fixing on a post. If the tags are wrong there is almost certainly something else wrong too.

Therefore, I will go a step further and say not only drop the retag privilege, but also forbid tag-only suggested edits. Suggested body edits already require 6 characters changed in the name of encouraging substantial edits, so keep encouraging substantial edits by making them do more than simply change the tags.

1
  • 3
    Yuck. Tag-only edits are very useful. Sometimes a post only needs to be retagged. Sometimes retagging the post helps funnel it towards the people who know the subject matter enough to edit it. I know I use the retag privilege a lot (mostly the 10k one, because relatively speaking I don't spend so much time on sites where I don't have edit privileges), and I benefit from others' use of the privilege. Commented Jun 27, 2013 at 21:23

You must log in to answer this question.

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