What is meta? ×
Meta Stack Exchange is where users like you discuss bugs, features, and support issues that affect the software powering all 150 Stack Exchange communities.

I recently answered a question that was several months old, and my answer was accepted. I thought this would earn me a badge, but upon looking, realised the badges were issued for upvoted answers, rather than accepted answers, fair enough I suppose as an answer may have already been accepted. However I did think it strange that there seems to be no gold version of this chain (e.g. 25 upvotes on an answer provided 90+ days after the question was asked). Does anyone know why there is no such badge?


Using EdoDodo's data query, here are some potential awardee numbers:
25 votes, 90 days = nearly 900 users
50 votes, 120 days = 250 users
100 votes, 120 days = 60 users

So now another question presents itself: How many users should a gold badge be awarded to, approximately?

share|improve this question
Feel free to ask for such badge.. edit one of the tags to badge-request and add full description of the badge. –  Shadow Wizard Jul 5 '11 at 7:55
You may want to run some queries on data.stackexchange.com to see if any existing users would qualify. That would give you some idea of it's potential to be awarded for future users. –  jonsca Jul 5 '11 at 7:56
And then pretty soon @JeffAtwood will be walking down the halls shouting, "Bring out your dead..." –  M. Tibbits Jul 5 '11 at 8:25
I've made a query that will take two inputs (the minimum days between question and answer, and the the minimum score of answer) and returns a list of people who would have received the badge, and the number of times they would have received it: data.stackexchange.com/stackoverflow/s/1555/… –  EdoDodo Jul 5 '11 at 8:35
If your answer was actually any good, it would have gotten an upvote at the same time that it was accepted. It's unclear why you think you deserve a badge for providing a mediocre (at best) and late answer. –  Cody Gray Jul 5 '11 at 8:50
Wouldn't you get a badge for the 25 upvotes anyway? Isn't that enough? –  Bo Persson Jul 5 '11 at 9:01
Well, in my opinion the revival/necromancer badges are great since they encourage people who have something new to add to an old question to do so. It seems like it would be a good idea to add a gold version of them. –  EdoDodo Jul 5 '11 at 9:16
@CodyGray I only joined the site last week. Clearly your answer was even later than my own, as you have not written it yet. I wasn't saying I deserved a badge, only that I thought I had earned one, but in fact was mistaken. This led to my investigation and the question above re: gold badges. –  Nicholas Jul 5 '11 at 10:14
Yeah, upon closer inspection it looks like I got bogged down in the introduction and didn't completely understand the point you were making at the end. I still don't think a badge is necessary here. You already get a badge for answers with 25+ upvotes. –  Cody Gray Jul 5 '11 at 10:21
@EdoDodo That's great. I have amended my question with some sample results. –  Nicholas Jul 5 '11 at 10:23
Well, gold badges should be difficult to obtain, but not impossible. In my opinion, (100 votes and 120 days) is way too difficult to obtain, and (25 votes and 90 days) is a bit too easy. Perhaps (50 votes and 120 days) would be a good choice? –  EdoDodo Jul 5 '11 at 10:41
While I, in principle, agree with @Cody -- there probably shouldn't be a new badge -- I can understand / appreciate a desire to encourage users to revisit and improve old answers / questions. –  M. Tibbits Jul 5 '11 at 17:45
@EdoDodo, I reused some of your ideas in my new answer. –  Nemo May 1 at 21:06

2 Answers 2

I think there is still a need of incentives to answer many more old, unanswered questions. Current badges don't help much: a bronze badge doesn't matter; getting 5 votes on a tumbleweed tag/question is very unlikely, however good your answer might be.

So, I propose: Persephone (gold badge, can be awarded multiple times): Posted 25 good answers to 90+ days old unanswered questions.

It's a mix of Revival, Necromancer and Socratic, with many details from Illuminator. In particular, Socratic is awarded for doing many times something good which didn't necessarily score much. The difficulty should be about the same, I think, despite the numbers being different. The name is also taken from Greek history, this time of a woman; alternatives could be Flamelicus, Zombie Army, Lich King or the original Reanimator.

(Other past suggestion were Highlander, not precisely defined; and The Long-Haul Trucker: Answered n Tumbleweeded questions with at least one upvote.)

See the query and play a bit with the numbers. I get only 38 "winners" on SO, with this necromancer earning it 6 times. However, there are 504 users with 10 or more: with some encouragement they could go the extra mile and clean up a bunch of old questions.

Maybe the badge would scale better and achieve wider impact if it required more answers but less upvoting? People would answer questions worrying about quality rather than upvoting-reactivity of the asker. This alternate version requires only a 0 score where the question had no previous answer: currently 27 winners on SO with the same truly unsung hero at 11 times (!).

An additional requirement for "maintained a positive answer record" could be added, similar to Socratic, but trying to post 50 garbage answers probably gets you banned way earlier than you get the badge. I just realised: a problem is that the badge might be awarded right after an answer is posted, whatever its quality, if it's the first and score 0 is the minimum in such case; perhaps we should only consider answers older than X days, to give people a chance to downvote them.

Update: I did some more runs of the query on multiple sites and it seems to be pretty general. Some sites would have very few awarded, but they would whatever the settings: perhaps just few unanswered questions, or nobody working as unsung hero/necromancer yet. On the many sites where a certain number of users would be matched, the criteria seem fair: wheter the min score for already-answered questions is 5 or 1, or min age is 90/45/10 days, the number of answers qualifying of course varies, but the pattern is always the same. I mean, we would not be unfairly excluding people who only focus on questions X days old or already-answered etc.; altruists hunting for unanswered/old questions to address seem to do so across the board. :)

But then, I'd use double checking of the query.

share|improve this answer
You can call the badge "Lich king". –  Deer Hunter May 2 at 5:32
@DeerHunter, let me add it to the alternatives. :) –  Nemo May 2 at 10:16
I like it, but I don't agree with your musing on "more answers with less upvoting". That just seems to encourage spamming a bunch of low-quality answers across old questions. –  mattdm May 10 at 3:56
@mattdm, I agree that's a very important point to consider. I'm relatively new in StackExchange so I don't have an answer, I just tried to present some data. Whoever makes the final call on this proposal surely knows better. :) –  Nemo May 10 at 19:24

I don't think a "gold necromancer" would bring any additional value to the table.

Necromancer exists to encourage people to continue answering old questions instead of focusing entirely on new questions. There's not a lot of difference between 60-day-old questions and more-than-60-day-old questions, so the new badge would be filling the exact same niche as the current one.

You could increase the score requirement, but as commenters have pointed out, there are already badges for high-scoring answers.

"Gold necromancer" wouldn't complete the "Revival/Necromancer/... pattern," either, because it's not a pattern. Revival is only awarded to the "first answer scoring 2 or more" (emphasis mine), so it serves a different purpose than Necromancer: reducing the number of unanswered/poorly answered questions.

One criterion that might make sense for a new badge is earning an acceptance on an old question. However, the OP may no longer be around, and a decent answer would get upvotes in addition to the checkmark, so this wouldn't be too useful in practice. It wouldn't deserve to be gold, either.

share|improve this answer

You must log in to answer this question.

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