While there are some very popular tags that generate many upvotes, there are also some tags that are technically challenging but hardly ever generate more than one or two upvotes per answer, hence making it almost impossible to get tag badges.

Examples are Spring (1 gold, 5 silver for 4,668 questions) and Maven-2 (1 gold, 2 silver for 2,894 questions) as supposed to mainstream tags like C# (54 gold, 207 silver for 127,069 questions) and Java (40 gold, 111 silver for 81,277 questions). The following diagram shows that it's a lot harder to get Badges in the unpopular tags: e.g. in C#, a Bronze badge is assigned for every 128.22 questions while a Maven-2 Bronze badge is assigned every 723.5 questions. alt text

I do however believe that in the spirit of the recent discussion Will high reputation in Stack Overflow help to get a good job? for more exotic technologies, specific tag badges will go further towards impressing potential clients than overall rep. (If I have to hire a spring consultant, I'll prefer one with a Spring Gold Badge over one with an overall 150k rep). So my suggestion is to make tag badges for exotic answers easier to get by checking for both upvotes and accepted answers.

I have hacked together an OData Query to calculate the average score of an accepted answer, the result is that the average accepted answer has a score of 3.071917 votes. So I guess 3 is a rate we could work with here, but I'd suggest 2.5, because it makes for nicer numbers. So I would like to modify the tag badges as follows:

  • Bronze: 100 upvotes or 40 accepted answers
  • Silver: 400 upvotes or 160 accepted answers
  • Gold: 1000 upvotes or 400 accepted answers
  • 1
    yup, similar, but not identical. I want to modify the existing badges, not add new ones Dec 2, 2010 at 9:39
  • 2
    I second this--it makes no sense to not get credit towards a tag badge if your answer is accepted by the asker, is correct, but doesn't get upvoted. I mean, a Struts 2 question might only be seen by a dozen people. Nov 26, 2011 at 2:45
  • I third this :-) In my limited experience of SO (so please read this comment with that in mind), I also feel the rule not only applies to obscure or exotic languages but also extremely 'popular' languages where a lot of entry-level programmers ask questions... e.g. JavaScript. Don't know if others have observed this or would agree. If it's right, it suggests a bell-curve on a chart of popularity (x-axis) to good questioning/answering/usage on SO.
    – guypursey
    Feb 15, 2013 at 15:14

1 Answer 1


This proposal has been officially declined at Cross Validated Meta where animuson wrote, in part:

This change potentially makes it twice as easy to get the tag badges. If tag badges didn't mean anything like the others, we frankly wouldn't care, but gold tag badges give users heightened privileges on questions with that tag, and we have to be careful not to make it too easy for that privilege to be obtained. Allowing this change would likely mean we'd have to modify the criteria for the tag badges to make them more difficult to obtain based on the new system, so any progress you'd make by counting the accepted answers would likely be diminished by the increased criteria for the badge.

  • This leads to the current situation where gold badges in some major tags (Java, JQuery etc.) are all but worthless because almost everybody has them, whereas tags for more exotic technologies (e.g. AspectJ) will never have gold badges, while almost everybody will eventually end up with gold badges in nonsense tags like "string". This is an unfair and unnecessary disadvantage for people specialized on certain technologies. Apr 26, 2017 at 15:56
  • Yeah... it was already proposed to have some kind of differentiated badge criterion, but the proposal never got much support. For one thing, nobody came up with an understandable and abuse-prone way of adjusting the thresholds...
    – user315433
    Apr 26, 2017 at 16:49

You must log in to answer this question.

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