What is meta? ×
Meta Stack Exchange is where users like you discuss bugs, features, and support issues that affect the software powering all 144 Stack Exchange communities.

Stack Overflow has quite a lot of tags with small number of questions.

Most of the time, it's very hard to get even a bronze badge for tags with less than 500 questions and it's literally impossible to a get silver or gold badge (there are just not enough questions and views to do so). On top of that, the number of views and upvotes in these small tags are quite low most of the time. So, users doesn't get too much reputation for their efforts either.

As result, users are disincentivized from participating in small tags.

Does it make sense to introduce a badge for being one of the top 5 answerers for small tags as an incentive for activity in small tags?

I suggest:

  • a bronze medal for being in top 5 answerers for small tag
  • and silver medal for being a top answerer for small tag.
  • definition of small tag: tag with more than 100 questions (we don't want to start giving badges for answering one question in a 5-questions tag) and less than 1000 questions.

BTW waffles in this (kind of unrelated question) question Remove the 25% requirement from the "Unsung Hero" gold badge wrote:

The concept here is to give these "poor users" that participate in tags that are not wildly followed or upvoted or happen to answer less popular questions some extra incentive to keep on making Stack Overflow better.

And I believe that's exactly what will be accomplished by such badges.

Kind of related question:

Unpopular Tags and Accepted Answers

Update 1 (Amount of small tags on Stack Overflow)

I did a StackExchange query and found that there are about 7k of tags with 100 to 1000 questions (it looks like this is a little bit too broad) and about 3.5k of tags with 200 to 1000 (this is more reasonable).

Update 2 (Efforts to get to top 5 for a small tag)

Just to put it in perspective. I work on two-three tags like that. I spend about a year actively investing in them: - closing bad question - retagging questions - answering questions (including doing some research to do that).

And I got to top 5 for all them. However, it wasn't a piece of cake. As I mentioned, it took me a year, I had to answers 40-50 questions (in some cases 25% of total number of questions in tag). So, it's definitely not "everybody gets a trophy".

Update 3 (Gaming the system)

Concerning badge hunters. I believe it will be reasonably hard to game the system here. As example, it will require creation of new tag, retagging or posting there hundred questions. Getting upvotes on these questions to get to the top. I would say it's too much work for getting one bronze medal. There are way simpler methods to do so. As example anybody can write up 100 semi-trivial questions and wait until they will reach reach 1k and 2.5k views.

Update 4 (Quantification of the problem and the solution)


I can't image the way how you can quantify decrease in activity on StackOverflow because of such problem and changes which will bring new badges.

Actually, I believe huge amount of features on SO is implemented based on feeling of multiple people vs actual statistical proof.

The only statistic which I can show is that there are several thousands on such tags and people who work on them.

share|improve this question
Is it not due to their less popular nature that they receive less activity? I believe SO naturally gives more attention and thus rewards for those questions the community is demanding answers for the most. Making it an "everyone gets a trophy" style of rewards would be detrimental to the overall site value. –  Leon Stafford Sep 15 '13 at 3:58
100 is too low. There are only 64 questions in the Forth tag. –  Peter Mortensen Sep 15 '13 at 7:53
@Leom Popularity isn't a direct measure of value, though. You can spend hours writing a very valuable, elaborate answer for a lesser-used language, and gain maybe an upvote or two; while a trivial jQuery answer can net hundreds of upvotes. Trying to alleviate that a bit isn't "everyone gets a trophy" at all. –  Pëkka Sep 15 '13 at 13:57
@LeonStafford: It's not about "everyone gets a trophy". It's about incentivising people answering questions in less popular tag to make SO a better place :) –  Victor Ronin Sep 15 '13 at 17:17
@PeterMortensen: First of all, the number 100 is arbitrary. On other hand, I didn't get relation between 64 question in Forth tag and 100 being too low. –  Victor Ronin Sep 15 '13 at 17:22
I take your points and in the case of writing the elaborate answer for a lesser used tag, I still believe the natural drive to show expertise will not prevent people from answering these questions. It would be beneficial though to show their "authority" in the subject matter, once proven. What the algorithm required for this though, I am unsure. ie, "Answered 50% of all {obscure} tagged questions" would not be great when only 2 questions exist. –  Leon Stafford Sep 15 '13 at 21:22
@notPekka I think you meant LeonStafford instead of Leom –  SheetJS Sep 16 '13 at 0:17
@VictorRonin FORTH is not a trivial programming language. It dates back to the 70s and is still actively used today. –  SheetJS Sep 16 '13 at 0:19
@Nirk: I know about Forth. I just wasn't sure why 64 questions in Forth tag is an argument for "100 is too low". –  Victor Ronin Sep 16 '13 at 0:27
@VictorRonin if you accept the idea that forth is a sufficiently important subject matter that it should be considered popular, then the 64 questions becomes an upper bound for the badge requirements –  SheetJS Sep 16 '13 at 0:30
@Nirk: Oh... got it. I believe you meant you meant lower bound? I suggest to call small tags everything between 100 and 1000 questions –  Victor Ronin Sep 16 '13 at 0:48
I like the idea, and suggest looking at the constable badge as an idea of how it could work. The point is that early participation should be readily invited to grow those communities and provide resources where there isn't much. I have found a lot of success after animuson consolidated the 4 google visualization tags into one (the participation increased visibly). –  jmac Sep 16 '13 at 23:50
This proposal is calibrated for the trilogy. On many other sites, tags with fewer than 1000 questions are the norm rather than the exception. –  Monica Cellio May 1 at 15:00
@MonicaCellio You are right. The idea could be modified to be reasonable for other sites. However, I feel like there should be some explicit incentive in participating in small tags on large sites. –  Victor Ronin May 1 at 15:37
@VictorRonin I realize you posed this on MSO originally (and then it became MSE). You might want to edit to either limit the scope or propose some adjustment for non-huge sites. Thanks. –  Monica Cellio May 1 at 15:39

3 Answers 3

users are disincentivized from participating in small tags

No, users are not specifically incentivized, but right now there's no disincentive.

Beyond that, your request fails to convince me that there's currently a problem.

You believe there's a disincentive. Even if that's the case, what is the impact on the site? Are these questions going unanswered? Are there programmers leaving the site because these questions aren't getting a lot of answers/votes? Are experts in these fields leaving the site due to lack of interest in their favorite tag?

And, assuming that any of the above is actually happening and can be quantified, is it a bad thing we don't want to have happen? Will adding badges reverse the situation significantly?

Not that I'm in charge, but I need more convincing than, "I've been working really hard, but I'm not getting enough rep/badges for my work, so please make some for me," before I get on board with your plan.

share|improve this answer
Obviously, I don't have statistics and most likely there is no way to quantify both "Are there programmers leaving the site because these questions aren't getting a lot of answers/votes" and "Will adding badges reverse the situation significantly". If you apply the same logic to any other problem, there won't be badges at all. It's not only me working hard and not getting enough rep/badges. There are other X thousands active members who work on small tags and have the same problem. –  Victor Ronin Jan 27 '14 at 17:41
"most likely there is no way to quantify" consider querying the data explorer. Determine the top, say, 100 tags that you consider important to programming but unpopular. How you do this is up to you, since it's, as far as I can tell terribly subjective. Once you have that, check out the top 20 people in each tag per your badge recommendations. Now graph their overall site activity over the years. Is it increasing, decreasing? Are they just answering those tags, or are they active in other tags? Have we lost some over the years? I don't think it's unanswerable. –  Adam Davis Jan 27 '14 at 18:05
I do think that Stack Exchange has really slowed down their pace of adding badges, and have a very, very high bar for a new badge. What little speculation you've so far provided is hardly enough to give serious consideration to. You really have to have a very convincing argument why these new badges are necessary, and not just a nice thing or good idea, otherwise you're not likely not make much impact on the people who actually make the decision. –  Adam Davis Jan 27 '14 at 18:06
I think you are right. There is a way to gather some statistic (probably it will require to do a comparison with people who answers popular tags). And such statistic, if it will prove that people who participate in small tags contribute less over the time would be very serious argument. I will think about gathering such statistic. However, it may take some time for me. –  Victor Ronin Jan 27 '14 at 21:23

Existing requirements for tag badges could be kept the same for small tags, while aggregating scores from subordinate tags to their hypernym. However, our tag taxonomy doesn't have subtags and parent tags.

To start "simple", a tag A on which you don't have enough score for a tag badge could inherit score from tags A* which have its name as prefix (and whose score was also not used for a tag badge).

In your case, I see you have score scattered over 20 android* tags, 6 ios* tags, 3 iphone* tags.

"Cheaper" tag badges for small tags may make sense, but AFAICS tag badges "cost" the same on less popular StackExchange sites where it's harder to reach as much traffic. A tag with the same amount of questions and answers is "smaller" on a smaller site, so how would you keep consistency?

share|improve this answer

I agree with you on face value of the proposal, and presently for less used tags it would be good to have lower thresholds to give people the same chance to earn badges as people in busy tags.

However, this could be detrimental to Stack and the current users of the unpopular tags.
Stack sites are very well known and busy sites and especially SO with LOTS of users, which means unpopular tags are very likely naturally unpopular, with a small number of questions/answers due to a small number of people having knowledge of them.

So, adding easier to get badges could also potentially:
1) Entice badge gamers;

2) if 1 is true, Q&A quality could degrade as it may entice people without any clue of the tag - or they'd be there already. There wont be enough good Q&A for bad Q&A to be lost amongst the good stuff, and also not enough people to vote them down;

3) If 2 occurs, then the lower tag threshold would need raising as popularity rises, and I can't imagine Stack wanting to regularly maintain badges, or most of the community as there are much better things that mod/staff/dev time could be spent on;

4) If 3 occurs, it's not fair some people getting a badge at 100 X, then someone else having to get 150 X, then a month later it's 200 X, and so on. While this has happened a few times at Stack, it's once in a blue moon, not a regular occurrence;

With regards 1), as the others kinda require it to be true, while I'm not saying you're a badge gamer, you are here wanting to change badge thresholds so you can earn them. That, and from what I've seen, I think there is quite a significant number of users who do want and try to get badges.
Hell, there's blogs and forums discussing how to get easy badges and rep on Stack sites...

Also, while there are badges hard to get on unpopular tags due to the tags being unpopular, there are many badges hard to get in the popular tags because the tags are popular. So an overall review and revamp of many badges would need to be considered, implemented and monitored to keep it fair across the board, which, to a degree, is your argument here.

I know it's nice to get badges, but I don't think it should be an incentive to be here or do anything on the Stack sites, just a bonus that happens naturally from being here learning, answering and helping others and joining in with discussions etc.

One way your proposal could be introduced is to perhaps test it on just a few unpopular tags first, and see what comes of it.
But again, that is quite a bit of work for Stack. Not sure it'll be done...

share|improve this answer
Afaik it's encouraged to actively pursue badges. I don't see any issues here, only updvoted answers count towards the badge. Yes, they wouldn't get anything by posting low quality answers. There needs to be some algorithm to determine thresholds. I don't think it's that unfair though, participating in a new small tag and helping it gain momentum is risky and should be rewarded. Otherwise beta tester badges would be unfair too because people to late to the party can't get them? –  kapep Sep 15 '13 at 14:12
Actively pursue badges is fine, actively pursue badges by asking questions/answering in tags you know nothing about is not fine when people introduce quantity that's low level quality. It's detrimental to quality of the site and the tags. –  James Sep 15 '13 at 14:22
Right. My point though is, I think most people know they won't get any badges/rep for low quality answers. Either they would need to increase quality or wouldn't bother posting them at all. If they don't know, that certainly is a problem but it concerns the whole site and not only small tags. –  kapep Sep 15 '13 at 14:44
@James: I completely agree with kapep. This new badges aren't free. They will require to do quite a lot of work in small tags and they will require good answers. –  Victor Ronin Sep 15 '13 at 17:19
I believe your concern that badge hunters going after obscure tags flooding the ratio of good vs bad quality questions is valid, though not without solutions. I do tend not to want to add extra incentives to the obscure tags. The mere fact that they are obscure is an incentive enough for those who are experts in the subject matter. As an authority, SO will rank high in search results for such questions, attracting more and more niche experts to the community, where they, too will want to showcase their talent to the world. –  Leon Stafford Sep 15 '13 at 21:17
@kapep - "people know they won't get any badges/rep for low quality answers": this happens all the time now. Loadsa people get upvotes on crap questions and answers. As per another recent meta thread, people who should by 2k rep be trusted, are robo reviewing to earn badges. Of all users, there will always be a chunky percentage who game the system, don't care, post rubbish and get away with it. –  James Sep 16 '13 at 6:05
@VictorRonin - "This new badges aren't free". Well, no, the point (of this question) is to lower requirements needed to earn them, and the mere attraction of them being "relatively" easy (easier than other tags!) may lower the standard of the tags from bad Q&A attempting to earn them. "They will require to do quite a lot of work in small tags and they will require good answers". A requirement of hard work to actually earn badges does not hinder the potential that some people will still post bad Q&A in an attempt to attain them.... All I'm suggestion is it should be a consideration ;) –  James Sep 16 '13 at 6:20
If activity level in these tags is frozen now, why would it suddenly increase? If lower requirements for badges does this, you have to ask if quality would also be present in the activity increase. I'm not against it, just pointing out some potential issues. It could improve the tags and bring more people in who do actually ask ok questions and half decent answers (or better). I don't know, of course. I do know people, however, and while I do not have a crystal ball, I wouldn't be surprised if you saw the tag activity increase but at a cost of quality. As always in meta, just my 2p worth :) –  James Sep 16 '13 at 6:29
@James: I understand your concern. You say that people will start posting lower quality answers to questions. I don't think it will be sustainable trend. These answers won't be upvoted, so they won't get to the top of of the tag. As result they will see that doing so, doesn't bring them anything. However, I hope it will lure enough people how will start writing reasonable answers to smaller tags. Please look at this: stackoverflow.com/tags/mdm/topusers I am literally the only persons who were answering questions for this tag for last 30 days. I think it could be improved. –  Victor Ronin Sep 16 '13 at 17:05
James, I came to SO to get answers on google-visualization questions, and found very little here. 8 months and 143 questions later, I'm the top answerer in the tag despite knowing nothing beyond the basics when I arrived. If badges help people put in the effort to answer basic questions, that's great! It creates a community which attracts experts like asgallant who is going to pass me in value to the tag after only 2 months. And that's great for the people asking questions, and great for SE. –  jmac Sep 17 '13 at 0:02
As I said "I'm not against it, just pointing out some potential issues. It could improve the tags and bring more people in who do actually ask ok questions and half decent answers (or better)." –  James Sep 17 '13 at 2:30
I know @James, I was sharing my experience. You may also be interested in the fastest gun in the west problem. Even if the quality of the answers is 'low', even a quick and dirty answer is better than no answer... –  jmac Sep 17 '13 at 7:58

You must log in to answer this question.

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