-12

Some of the tags on the main site have "a bit" of a quality problem. and jump to mind.

They get hundreds of new questions a day, which can be broadly put into three categories:

  1. Duplicates (about 45%)
  2. Badly researched/off topic (about 45%)
  3. Genuine questions (only 10%)

Now in the recent elections, one of the top PHP users (aka Gordon) won and the tag should have a little more mod activity on it.

However, the is still in pretty bad shape.

I'm proposing temporary tag moderators, who have limited moderator powers extending only to a few tags, and are appointed only for the time the tag needs maintenance.

These users will not have full moderator powers, but instead be a hybrid between 20k users and moderators.

Their powers

  • Binding close votes for questions on those tags
  • Binding delete votes for questions on those tags
  • Ability to migrate questions to appropriate other SE site, if available (for example, Android.SE from the Android tag)
  • Unlimited reviews for questions on that tag, in addition to the review limits like 20 edits reviews.
  • Ability to handle only Not an Answer and Very Low quality flags for questions on those tags.

Selection Process

Such "Tag Moderators" should be elected through an election, in which users with a gold (silver?) tag badge may nominate themselves. Only tags with a certain number of questions a day should require such moderators (100/day? 200?)

Users nominating should have a deep and specialized knowledge on that tag, and be familiar with the Stack Overflow Q&A model.


Benefits

  • Reduces workload for moderators
  • Improves tag quality

Downsides

  • Creates a bit of a temporary power inequality between users
5
  • 1
    Why not cut through some of the process and just give those powers to anyone with a gold tag badge?
    – blahdiblah
    Mar 16, 2013 at 21:23
  • @blahdiblah Because some tags have a lot of gold badges. There is no need for so many users to be given those powers. Mar 16, 2013 at 21:24
  • 2
    Randomly sampling some pages from the android tag (50 per page), your numbers just aren't there. I'm seeing 5-7 questions closed (non-duplicates), 0-1 migrated, and 0-2 closed as duplicates on each page I look at.
    – animuson StaffMod
    Mar 16, 2013 at 21:38
  • @animuson A lot of what look like genuine questions are actually very very badly researched or duplicates. As an example, we have 128 questions with NetworkOnMainThreadExcpetion in the body. There is only one reason you will ever get that error, and it can only be solved by a grand total of 3 methods. All of those questions could be answered with a single answer copy pasted word for word. Mar 16, 2013 at 21:40
  • 1
    @RaghavSood Let me know what the canonical answers to those questions are. FWIW, I've been filtering my close review activity on android for weeks (months?... time is slipping away from me) now. If you cast close votes and flags I will see them. Mar 16, 2013 at 23:57

2 Answers 2

15

No. One does not need to be an expert in a topic to be a moderator. I'm an first year undergrad physics student, with a little more than high school level chemistry knowledge. Yet I moderate the Physics and Chemistry sites (and not badly, if I say so myself). I'm not an expert in either, I'm a novice. I wouldn't be able to answer most of the questions on Physics. Still, moderating the site has nothing to do with how good you are with a topic. Familiarity is enough.

The SO mods could indeed spend more time in these tags. But, wait: there's a reason that they spend less time here. Most of an SO mod's moderation comes from the queue. While using the site as "normal" users (asking, answering), they get a very small window to the site, though they probably are killing bad things on sight.

If the tag has a quality problem, adding mods won't help1. You need to get your community to start participating more in community moderation by flagging and voting to close. Point them to this guide or this long guide on Physics to get started.

1. Especially if you do it by rep/badges. Rep has little to do with how suited a user is for moderatorship, besides the familiarity with tools. Rep in a tag has nothing to do with how well you will moderate the tag.

3
  • 1
    While I agree that moderating overall is not dependent on subject knowledge, users who have been on that tag for a while, and answered enough to get gold badges will be much much better at detecting duplicates and problems regarding that topic that aren't appropriate for SO in their current form than users who have little to no experience there. Mar 16, 2013 at 21:36
  • 4
    @RaghavSood: That's just a slight advantage. Also, they already can vote to close as duplicate/off topic (it only takes one vote to push it to a queue). Make them use the close vote review queues. There's a shiny "filter" option there, show them how to use it. Mar 16, 2013 at 21:39
  • 3
    Again, mods don't go about patrolling their tags for bad stuff. Mods clean up what's in the flag queue, and not much more. Patrolling the tags is for community moderation. Mar 16, 2013 at 21:40
7

In general, more moderators is better than not enough.

This feels like a lot of overheard either way, though. An election isn't a trivial thing to run, even if we aren't necessarily talking about the same style of elections as we have now for site-wide moderators.

Are those percentages based on data or are they just your perception of how things are? The community should be able to take care of things... and if it's not, more community members stepping up would be a better way to go than electing "super users" and managing their powers.

To quote from the Theory of Moderation:

We designed the Stack Exchange network engine to be mostly self-regulating, in that we amortize the overall moderation cost of the system across thousands of teeny-tiny slices of effort contributed by regular, everyday users.

Let's not lose sight of that. Appointing additional moderators, no matter how or with what restrictions, goes against the basic principle of the network - most moderation should be community-driven and not dictatorial.

3
  • For the android tag I've mostly eyeballed the statistics from experience over the past year or so, but I'd be willing to run proper numbers if it helps. Mar 16, 2013 at 21:33
  • @RaghavSood I'm not sure it would make a big difference, but having more accurate numbers probably wouldn't hurt. (Note that the Data Explorer doesn't give you access to deleted content, so the numbers will be a bit skewed either way.)
    – Adam Lear StaffMod
    Mar 16, 2013 at 21:36
  • Yeah. I'll probably go through the next 100-200 questions as they come in manually when I have some free time and compile the numbers. That way I can track deleted questions too. Mar 16, 2013 at 21:38

You must log in to answer this question.

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