6

A gold badge for a tag gives a user additional review power, to help manage low-quality questions having that tag.

At https://stackoverflow.com/help/badges?tab=tags&filter=gold, we see that there are many tags with only one person having a gold tag.

Wouldn't it help site quality, if the formula took into account the total popularity of a tag? Presumably on any tag, there are a handful of people "most expert" on that topic - even if the topic itself isn't popular enough for them to reach 1000 score (about 10K) rep on 200 answers.

Some ideas for alternative criteria:

  • Top "N" scorers for answers of a tag.
  • Threshold adjusted based on tag popularity. Would make sense to have a "sliding scale", with some minimum. E.g. the least popular tags might still require 200 score on 100 answers. (If no one has that much, then it probably isn't a very useful tag - perhaps its a variant of a more popular tag.)

Possible downside: will this make it "too easy" to get gold privileges on some tags?

Idea to minimize downside: Minimum reputation requirement, perhaps 5K total rep?


One question that comes to mind is whether the badge would be "taken away" if the tag becomes more popular over time, thus the badge holder no longer "meets the criteria". I say "No" - once you have a badge you always keep it. Taking it away would be hurtful, and serve no useful purpose: if someone was an expert early in the lifetime of a tag, its good to reward them / encourage them to stay active.


Note that I'm not thinking about tags with very few questions. They would fall below the minimum threshold.

I'm more interested in the "middle ground". There are quite a few "niche" tags that have a few thousand questions; not a hundred thousand or a million questions, like the top few tags. Very important tags to those using them; just not a ton of programmers that do.

————————- UPDATE

Independently of any privileges gained, I think it would be worthwhile “recognition” of those knowledgeable about a topic.

It doesn’t feel right that a handful of early answers on a very common topic gets one most of the way to a gold badge, whereas the top three answerers on a niche product may still never qualify for a gold badge.

This suggests a possible alternative criteria: the position one has within top answerers for a tag, could contribute over time towards a gold badge.

9
  • 2
    Counterpoint/frame challenge: gold badges are the opposite of what qualifies one for a dupe hammer. To get a gold badge, you have to answer many questions. And it's easiest to do it by answering duplicates rather than closing them. Dupe hammer prevents answering questions.
    – VLAZ
    Dec 23, 2021 at 23:53
  • @VLAZ - Huh? I think you pointed out that its easier for people to get that gold badge, if they answer dup questions instead of closing them. What does that have to do with the desirability (or not) of making it easier for the top answerers of a "niche" tag to get the gold badge? Dec 23, 2021 at 23:57
  • 3
    Your entire argument relies on a gold badge giving a user a dupe hammer. I posit that gold badge shouldn't be a criteria for a dupe hammer because it's a bad criteria for it. Ergo, frame challenge - gold badges being made easier to get do not automatically make for better quality. Therefore, the right question would be what should the criteria for a dupe hammer be.
    – VLAZ
    Dec 24, 2021 at 0:01
  • Your counter-argument boils down to either a) its a criteria that could be abused, or b) its a criteria that interferes with achieving dupehammer ability. Isn't that equally true of the current criteria? [Therefore is not relevant to my suggestion.] Feel free to suggest an alternative criteria. Lacking an ideal solution, I think this adjustment would be a major step forward for "niche" tags. Dec 24, 2021 at 0:14
  • 4
    What you are not taking into account, is that the top answerers on any given tag are highly committed to that tag. Giving them more ability to manage the tag is a clear benefit to the site. If anyone abuses this, there are mechanisms to deal with abuse. (Your counter-argument falls into the fallacy that leads to the aphorism "the perfect is the enemy of the good".) Dec 24, 2021 at 0:22
  • 3
    "What you are not taking into account, is that the top answerers on any given tag are, inherently, highly committed to that tag" It's top answerers - Where do you get the direct link to "top people who close questions on the tag"? I don't. There is no such link I have witnessed. Anybody is free to answer a thousand questions and never cast a close vote. Conversely, people who seem to be very active with curation, don't answer as much instead preferring to dupe close. "[Therefore is not relevant to my suggestion.]" - I said it twice already: frame challenge.
    – VLAZ
    Dec 24, 2021 at 0:27
  • The perfect is the enemy of the good. I have made a suggestion that I believe is likely to lead to an improvement in tag management. Your counter-arguments are valid, but miss the point. Most notably, which you have not responded to, the "frame challenge" applies equally to the current situation. I propose to tweak that situation. Is this a perfect solution? Obviously not, as you correctly point out. Yet nothing you say addresses the possibility that this would improve the current criteria. That is, I assert there is positive correlation between top answerers and the privilege. Dec 24, 2021 at 0:32
  • 3
    I think this is a good idea, it would not only help curate niche tags on Stack Overflow but it would also be beneficial for 'regular' tags on other SE sites that get less traffic than SO (similar to Beta sites where the requirements are already lowered).
    – Marijn
    Dec 24, 2021 at 11:00
  • 1
    That list (in the link) is particularly long and especially useless; with no ability to sort by name or number, and the central text is almost an exact duplicate that should be a heading with more useful info in the center (something that differs for each entry).
    – Rob
    Jan 30, 2022 at 12:57

0

You must log in to answer this question.

Browse other questions tagged .