I'm aiming for a mechanism that can be used to highlight high-quality answers that have garnered few votes. I think this would provide a valuable UX element for users to become more engaged, and on-boarded in community practices.

It would essentially be a badge-recognition of answers that heavily use references in the format of this meta.SE feature request

Thus, this may not be a functional addition to the engine until the above reference/citation feature request is fulfilled.

An example of such an answer

References to citations are a critical part of discourse. Academic areas like History and Chemistry rely upon references / citations more heavily. To give a succinct-comparison from another academic field, it has been said that "citations are the source code of Physics.SE"

Should we allow the engine to recognize good citation / reference practices via:

  • a badge after a fixed number of answers, voted to N-score, each with at least X citations
  • a badge when total citations in all answers exceed Y
  • when a moderator awards Q or A with subjectively-good ('policy') citations.
  • etc.

By 'counting citations', I mean "counting the number of reference link that are included in the body of the question or answer".

  • 9
    How would you automate detection of citations? Not all links are citations, and not all citations are links. Relying on moderators to make decisions is an invitation to drama (as well as being extra work). Apr 15, 2013 at 18:26
  • 1
    too easy to grind Apr 15, 2013 at 18:27
  • 1
    I suppose you could have some sort of [citation][/citation] tagging system, but is that worth implemetning for a few badges? Apr 15, 2013 at 18:29
  • 3
    @BlahBlahGrabblesnackers maybe not just for the badges
    – Rory
    Apr 15, 2013 at 18:31
  • @BlahBlahGrabblesnackers I'd prefer a mechanism that can be used to highlight high-quality answers that have garnered few votes. Apr 15, 2013 at 18:44
  • @MonicaCellio the ref. given by Rory is the real meat of my idea. I've updated the question to reflect that. Apr 15, 2013 at 18:54

2 Answers 2


As another answer says, this is what voting is for. In addition, anybody who feels so moved can reward a high-quality answer with a bounty. In addition, you can use chat to highlight under-valued (or over-valued :-) ) answers; I've seen the chat approach work on Biblical Hermeneutics (a tiny site) and Mi Yodeya (not tiny but not SO :-) ), so presumably it would work elsewhere too if you have any sort of community cohesiveness already. (And if you don't, you have a bigger problem than not having a badge for citations!)

If you want to recognize these posts in ways other than reputation, consider creating a "citation hall of fame" post on your meta and linking answers that you want to highlight there. Of course, you will have to come up with criteria for doing so. Since this is a per-site value, your meta is the place to work this out.

  • While this is a fair way to deliver recognition, it lacks the search-ability of a badge that is delivered to a well-cited (referenced) answer. Referencing is a strong addition to a Q&A, and part of the modus operandi of SE is to exemplify good community behaviors so that new users can be encouraged to actions that improve site quality. Badge-rewarding such answers could be used for that purpose. Apr 15, 2013 at 20:57
  • So if there were a way to search for bountied answers, that would meet the need but not require AI or subjectivity (read drama) for badge awards? It could be handy to find bounty awards (independently of this use case); consider a feature request. Apr 15, 2013 at 21:04
  • Yes - I propose that would be an interesting feature, also Apr 15, 2013 at 22:16

I'm aiming for a mechanism that can be used to highlight high-quality answers that have garnered few votes.

This is antithetical to the entire premise of all SE sites.

If citations are particularly important on Physics.SE then that will be reflected in voting. Badges should be uniform across all SE sites, so that working code on SO, citations on Physics.SE (allegedly), recipes on cooking.SE (I flippantly guess), or whatever suits the domain garners upvotes, and is rewarded through rep and through badges like Nice Answer, Enlightened, etc.

  • No, it's not. There are good answers that are only interesting to a few people. The answers are not low-quality despite receiving little attention. History.SE sees these. Apr 15, 2013 at 19:20
  • 2
    @NewAlexandria Okay, fair. But my point is that until computers are actually intelligent, automating expertise is futile.
    – djechlin
    Apr 15, 2013 at 19:24
  • 1
    How's a mechanism that can be used to highlight high-quality answers that have garnered few votes antithetical to the entire premise of all SE sites? What are "reward existing answer" bounties supposed to be about then?
    – yannis
    Apr 15, 2013 at 20:02
  • @Yannis the premise of SE is that the community can use its collective expertise to drive what is a high quality answer. Once this can be automated we can get rid of SE. And I don't know what those bounties are about or if anyone has ever started a bounty for that purpose. AFAICT it's just a way to launder reputation.
    – djechlin
    Apr 15, 2013 at 20:17
  • 1
    @djechlin I've started several bounties to reward outstanding answers, are you accusing me of reputation laundering? "reward existing answer" is not something I made up, btw, it's one of the predefined reasons for bounties, it's an option the system gives you. I can see your points against an automated system, but saying that the whole idea of highlighting high-quality answers is somehow "antithetical to the entire premise of all SE sites" is plain wrong.
    – yannis
    Apr 15, 2013 at 20:23
  • @Yannis so can you please answer your question, "what are 'reward existing answer' bounties supposed to be about then?"
    – djechlin
    Apr 15, 2013 at 20:26
  • @djechlin Isn't it self explanatory? Give a little of your rep to reward an existing answer. And after you award the bounty, the rep amount is visible for all to see that someone thought the answer was worth some additional rep (thus highlighting it).
    – yannis
    Apr 15, 2013 at 20:26
  • @Yannis I don't know in what situations someone would be compelled to do that or if whatever reason that reason was added for still pertains. In any case that falls under the community regulating itself.
    – djechlin
    Apr 15, 2013 at 20:28
  • @djechlin Ok then. Also, I do agree with you that an automated system wouldn't be the best idea.
    – yannis
    Apr 15, 2013 at 20:29
  • @Yannis back to the link Rory provided, above, I think that an automated system is entirely feasible, once a proper citation mechanism is in-place. Apr 15, 2013 at 20:59
  • @djechlin Once A proper citation system is in place, the engine already allows for complex rules, per the Winter Hat event in 2012. Pardon, but I think you're being a bit dramatic to suggest that this is an attempt to "automate intelligence" Apr 15, 2013 at 21:03

You must log in to answer this question.

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