30

For the 400 votes tag badge, does it take into consideration answers marked as the accepted answer? The description indicates it does not.

I would think an accepted answer would be worth more, or at least the same amount, towards getting a tag badge than just a simple upvote? Lots of users mark answers as accepted and do not bother giving an upvote since the 15 points is like and upvote and a tip.

So it's possible for a user to have 400 accepted answers with no upvotes and not get a badge where a user with 400 answers (which could be right or wrong) with a single upvote and no accepted answers ever would get the badge. Just seems a little odd or does accounting for accepted answers make the badge calculation too complicated / slow?

Edit: Questions you ask and submit an answer to shouldn't be counted towards the total for accepted answers.

Edit: It just seems to me that the most important thing in the system is an accepted answer yet there are no badges for achieving more than one accepted answer. If up votes were more important, and therefore far more worthy of a badge, then why do the top voted answers not appear above the accepted answer? When it come to badges, accepted answers are second class citizens as compared to a possibly random upvote.

I have seen many upvotes on completely wrong answers but rarely do I see an accepted answer that was completely wrong.

4
  • 4
    The concept is not "correct", it is "accepted". There is a big difference between the two.
    – nb69307
    Apr 15, 2010 at 22:50
  • 6
    An upvote doesn't mean the answer is "correct" but it counts
    – Germ
    Apr 15, 2010 at 22:55
  • Did a search and replace for correct to accepted :P
    – Kelsey
    Apr 15, 2010 at 22:57
  • I think that this is a valid argument; but the site owners/admins seem to want to make badges harder to achieve (no proof but it would seem logical) because some people have got hundreds or thousands of badges. So I think any measure that would make badges easier to achieve will probably not be popular with the site owners/admins as it would auto-award a ton of new badges.
    – amelvin
    Jan 12, 2014 at 14:05

5 Answers 5

5

Another reason why accepted answers shouldn't be counted for tag badge totals is because users have the ability to give them to themselves. You can accept your own answers, but you can't upvote your own answers, so counting accepted answers for the tag badge totals opens up an avenue for people to game the system.

5
  • 12
    True, although the system is obviously aware of the distinction, since you don't get rep for self-accepts either. So, while I'm on your side here, this particular issue could be ironed out.
    – Aarobot
    Apr 15, 2010 at 22:57
  • @Aarobot: You're correct, the issue could probably be ironed out, but that would likely make it more difficult to implement and thus even less likely to ever happen. Apr 15, 2010 at 23:00
  • 2
    Questions you ask and submit an answer to shouldn't be counted towards int the total for accepted answers. I will add this to my question.
    – Kelsey
    Apr 15, 2010 at 23:02
  • 1
    Surely this can be solved in the same way that no rep is gained in self-answering
    – amelvin
    Jan 9, 2014 at 9:53
  • Accounting for this is absolutely no rocket science. If the answer is accepted, add one, and also subtract 1 if at the same time the OP is the author or is among upvoters. Aug 29, 2018 at 15:49
2

Sorry about answering an old question - better than duplicating!

As the long-tail of questions spreads into more and more tags there are lots of tag areas with little traffic and even fewer upvotes - so if you want to get a badge (rather than some rep) in a large number of tags it is very difficult with the current voting structure.

One badge I've been trying to work towards is the 'Umbraco' badge - which currently has 1,421 questions, but doesn't attract an awful lot of votes. And so only has three people with the 100 upvote 'bronze' badge. 80% of questions asked with this tag have an accepted answer but more than 50% of answers have no upvotes. Surely after nearly 1,500 questions only three badges awarded suggests that the algorithm needs a tweak?

Also there seem to be fewer upvotes generally (I know its been mentioned elsewhere) - but of my 42 accepted answers up to December 2011 only 7 (17%) have no upvotes - in the last 18 months 8 of 21 accepted answers (38%) have no upvotes.

So to balance the decline in upvotes I think an accepted answer should count as a tag credit - as long as you are not answering your own question. I also think that new users should be able to upvite immediately - even if it confers no reputation until they reach a certain reputation themselves.

2
  • "accepted answers are a fine social contract, but not a good data point for question or answer quality" (Jeff Atwood)
    – gnat
    Feb 22, 2018 at 15:59
  • 2
    A bad answer to a popular question often generates more rep than a good answer to a bad question, simply as a function of traffic - so I don't agree that answer upvotes are any better as a data point for answer quality than accepted answers.
    – amelvin
    Jun 6, 2018 at 7:43
1

I actually fail to see why. The sole purpose of the tag scope is to measure a person's contribution to the tag.

  • Both accepts and upvotes mark the answer as "useful" for another person -- and "usefulness to other people" is the main (if not the sole) criterion by which SE's system awards reputation points (thus measures a user's contribution)
    • in fact, acceptance even more so: it marks the answer as "the most helpful and useful" rather than just "helpful"
    • So they both mean that the answer is a useful contribution to the site -- and, transitively, -- to the tag
  • They are both equally personal and subjective
    • So are both equally prone to marking an "incorrect" answer, being misguided and such
    • Yes, aggregate votes show community consensus, but each individual vote is a personal opinion. Acceptance is the same personal opinion, just expressed by another mechanic.
      • E.g. would you think that votes by the OP shouldn't count if there was no separate acceptable mechanic?

To add insult to injury, the system currently happily counts an upvote by an OP. Which both disproves those who claim that the current design's intention is to disqualify OPs from judging answers to their questions, and unfairly discriminates those answering questions by novices (something that the system designers explicitly claim to actively encourage).

  • (Others suggest that the discrimination also hits answers in unpopular tags. So, in addition to being bad on its own, it also exacerbates the problem of creating tag synonyms since those tags have few users eligible for voting to begin with.)
-1

Accepted answer != correct answer.

This may have changed somewhat due to the new 15-minute minimum time imposed on acceptances, but before that, it was very common for people to just accept the first answer that was posted, often within seconds (apparently without even reading it).

So no, I wouldn't say that acceptances are worth more than upvotes. Upvotes have been given by the community; acceptances are only given by the person who asked the question. And although it is possible to get specialist badges simply by answering hundreds upon hundreds of questions and getting a bunch of single-serve accept/upvote combos from the question starters, I think the badge is really meant to be for giving consistently good answers as determined by multiple upvotes.

3
  • Some questions not about a technical problem. In the case Accepted answer say nothing. But it must be extriemly seldom if the answer is "accept" but without upvotes and it is not really solve the problem. In the most cases it is an answer on a technical question from a person who are new on the stackoverflow and the tags of the question are not popular enough, so the question and the answer read not many people.
    – Oleg
    Sep 7, 2010 at 12:46
  • 1
    Upvoted answer also != correct answer. A low-traffic tagged question might only get a dozen views in a day--if it's accepted by the OP, but not upvoted, I guess I don't believe I shouldn't get the same consideration as when it's upvoted. Nov 26, 2011 at 3:14
  • "Upvotes have been given by the community" -- patently false. You imagine "the community" as some amorphous cloud that is infinitely wise and can't be wrong. In actuality, only the aggregate score gives some kind of "consensus". Individual votes that comprise it are personal opinions, just like the acceptance mark is. Aug 29, 2018 at 15:53
-1

I'm going to steal my answer from another question:

I don't think accepted answers should contribute to tag badge counting. Having an accepted answer only means the OP thinks your answer will work for what s/he is doing, and the OP has a vested interest in getting an answer that works for him/her. What truly counts, IMO, is the people who do not have a vested interest in the outcome of the question, the people who simply think the answer is clear, concise, and correct.

EDIT: Response to this:

So it's possible for a user to have 400 accepted answers with no upvotes and not get a badge where a user with 400 answers with a single upvote and no accepted answers ever would get the badge.

Yes, it is possible, but it is extremely unlikely. Further more, the reward would be that the person with 400 accepted answers would get 2000 more rep than the person with 400 upvotes and no accepted answers.

3
  • 1
    It seem that upvotes/accepted is extremely depend on the popularity of tags from the question. Compare this on stackoverflow.com/tags/c%23/stats, stackoverflow.com/tags/jquery/stats, stackoverflow.com/tags/jqgrid/stats. In more seldom tags there are a lot of question from users who ask only one question. They use either no accept and voting at all and write just "thanks! it work now!" in the comment or accept the answer without upvotes. In my experience the most "accept" answers really soled the problem. So calculation of "accept" would be fair.
    – Oleg
    Sep 7, 2010 at 12:47
  • it must be extriemly seldom if the answer is "accept" but without upvotes and it is not really solve the problem. In the most cases it is an answer on a technical question from a person who are new on the stackoverflow and the tags of the question are not popular enough, so the question and the answer read not many people.
    – Oleg
    Sep 7, 2010 at 12:48
  • 1
    Almost 7,000 people on SO have been awarded 'unsung hero' - those people alone must have had an accepted answer on well over 100,000 questions that scored zero upvotes (an absolute minimum of 11 answers like that for each person - but probably higher). Zero score accepted answers are far more likely in 2014 than in 2009 - and the SO algorithms should shift to acknowledge this.
    – amelvin
    Jan 12, 2014 at 14:41

You must log in to answer this question.

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