5

I have been reading some questions about the Unsung Hero badge (applies to Tenacious too), and it seems this badge is a little controversial (questions arguing if it is a good thing to pursue the badge, the gamification that surrounds it e.g. the guy asks to the OP to not upvote the question, etc).

What is the point of not considering the +1 (upvote) that generally comes from the OP?

I mean, many users upvote answers when accepting them. Some users do not upvote.

If the badge's goal is to reward users who answer questions that do not get much attention, what is the difference under the point of view of getting attention when a question is accepted and has zero score, or is accepted and have 1 upvote?

Ok, the sole upvote does not always come from the OP, but surely the great majority of cases it does.

I think removing the zero score criterion on accepted answers for Unsung Hero will get more community consensus about this badge.

Tenacious (silver): Maximum of +1 score accepted answers: more than 5 and 20% of total;

Unsung Hero (gold): Maximum of +1 score accepted answers: more than 10 and 25% of total.

Please, give your opinion.

  • If the upvote from the OP is considered, then it is possible to be a slightly more-sung hero. – Aza Jul 31 '13 at 0:57
  • I can't believe there are over 5,000 users with the Unsung Hero badge on SO. I expected a much smaller number. We have ZERO on dba.SE. – Aaron Bertrand Jul 31 '13 at 1:10
6

Original answer

I agree with the sentiment of this discussion. For a while now I've been thinking that the advice to "upvote and accept" runs counter to Tenacious and Unsung Hero.

One possibility would be to ignore only an OP upvote when awarding these badges. This would correctly disqualify answers upvoted by some random party, but would leak some information about the identity of the sole upvoter (the OP), which may be a bad thing. (This would also make it slightly more difficult to correctly word the badge descriptions; how about

(gold) Unsung Hero: At least 10 accepted answers, of which 25% have not been upvoted (except by the asker).

?)


Response to AsheeshR and jmac

The above users argue that:

  1. Ignoring OP upvotes in badge calculation would impede progress measurement.
  2. If measurement were impossible, people would be less motivated to get the badge.
  3. If measurement were possible, lone OP upvotes would be de-anonymised, which would be bad.

On the tracking side

I don't have much data either, but I don't believe that the de-anonymisation would be much of a net negative. It is unlikely that all that many of the questions counting towards your badge were asked by the same person, making it difficult to collect aggregate statistics; also "otherwise-zero-score-answer-upvote rate" probably isn't a priority to stick on a usercard like accept rate was (wrongly) judged to be, and it's difficult to determine by other means.

On the non-tracking side

Is tracking progress so critical? I was awarded Tenacious just as a side-effect of answering questions and helping people. If anything, not knowing how close (or far away) they are from Unsung Hero will just lead people to try and post more answers in lonely sections. (That's my theory, anyway. Feel free to produce data that shoots it down.)

tl;dr

Tracking progress probably won't be that bad; not tracking progress might not be that bad. I vote to track.


Second response to jmac

In the comments below this answer jmac points out that this proposed change would dilute the value of the badges by making them quite a bit easier to get. This is a more serious problem for the proposal, which I don't have a solution for at this time.

  • Thanks for replying (+1). That would be very positive (IMO). Not sure only, if the routines to accomplish this criteria would be easy to implement. – Andre Silva Jul 31 '13 at 1:15
  • @AndreSilva Votes must be identifiable at some level; the system knows to revoke the right ones when there's serial voting. Shouldn't be too hard to remove the upvote by $OP_USER when checking for Unsung Hero. – michaelb958 Jul 31 '13 at 1:44
  • Ok, perfect then. – Andre Silva Jul 31 '13 at 1:52
  • Disagree for the reasons AsheeshR eloquently stated – jmac Jul 31 '13 at 2:59
  • @jmac I have expanded my answer to respond; see how it looks now. – michaelb958 Jul 31 '13 at 4:31
  • It looks like 34/42 of my accepted answers with upvotes were upvoted by the OP. This would change my percentage of posts for Unsung hero from it's current 31% to something like 87%. – jmac Jul 31 '13 at 4:57
  • @jmac Nice destroying my "difficult to determine by other means". :) More seriously, though, that is what the proposal is all about. How many people would go from 6% to 28%? – michaelb958 Jul 31 '13 at 4:59
  • The question is where do we draw the line? This is like removing the 25%, or decreasing it from 25% to 10% or somesuch. Should this be one of those situations where moonlighting in an unpopular tag occasionally gets you the badge? – jmac Jul 31 '13 at 5:06
  • @jmac I'm not seeing the mathematical/statistical basis for that. You'd still need, at minimum, 10 (11?) answers either at zero score, or at one score and with an OP upvote. – michaelb958 Jul 31 '13 at 5:15
  • I spend 100% of my time in unpopular tags. 31% of my accepted answers are 0-score. If we changed this counting method, that number would leap from being barely about the Unsung Hero threshold of 25%, to being dramatically over it (85%). This is the equivalent of keeping the scoring the same (zero answers only) and dropping the % of total answers from 25% down to 10% (give or take) in the sense of how much more accessible the badge would be. – jmac Jul 31 '13 at 5:42
  • The point is that making this change will make the badge dramatically easier to get in the same way lowering the % of your total accepted questions from 25% to 10% would. It would dilute the "value" of the badge by making it easier to get/game for those who felt the need, without increasing the contributions of anyone. It is just redrawing the line for the badge so more people can have it. – jmac Jul 31 '13 at 5:43
  • See update. cc @jmac – asheeshr Jul 31 '13 at 7:26
  • @jmac, why are you saying to shrink the percentage to 10%?. What I'm proposing (asking for opinions) is: today you have 132 answers, 30 answers accepted and with +1 score, 19 accepted answers with zero score. In the current criterion, one would not be a Unsung (14.39% of zero score accepted answers). In the new scenario the percentage would be 37% (49/132), and one would get the badge. It would increase the number of badges awarded (but not deliberately). I am stating that to consider the OP's upvote is legitimate and fair (the question does not get more attention if it has one upvote). – Andre Silva Jul 31 '13 at 13:16
  • @AndreSilva Unsung Hero is counted as 25% of accepted answers of which I have 61 or so. Half of my answers are unaccepted, not because they are poor, but because many people just hit and run question. That's why there's the badge. Perhaps before suggesting, you look at the query I posted and understand the badge criteria? The 10% is saying that I would go from 31% to 85% or so with just this change. That's an absurd increase in the likelihood of getting the badge even if I posted in popular questions. – jmac Jul 31 '13 at 13:34
  • 1
    let us continue this discussion in chat – jmac Jul 31 '13 at 23:55
4

Unpopular tags have really bad track records with getting answers accepted, let alone accepted with upvotes. The question volume is low, there are many hit-and-run questions in general, and very little in the way of alternative resources on the net.

Having valuable information not available on other resources located on Stack Overflow greatly increases the value of SO as a resource. People looking for the answers to questions will find them on SO high up in Google searches, and that will grow the knowledge base on SO and hopefully start the ball rolling toward creating a community of experts (or at least enthusiasts).

I joined to get guidance on those unpopular tags, hoping someone would know. Instead I ended up answering my own question and providing 132 answers to questions from other people.

While I don't care much about reputation or badges, it still felt nice to have a nice shiny gold badge awarded for what felt like (and still does feel like...) pretty lonely work. I think the badge works fine as-is and gives those of us fighting our shadows a twisted satisfaction when we rack up yet another 0-score accepted answer.


Some stats. I have 61 accepted answers according to this query. Of those 61 accepted answers, 42 have upvotes, and 34 seem to have been upvoted by the original poster (they had an upvote around the same time the post was marked as answered). Still trying to figure out how to use the to create a query to get the reputations of people whose questions I have answered, but I am finding myself incapable.

If this query were implemented, I would go from being 31% unsung to being over 85% unsung. I think this would dramatically change the concept of the badge from promoting questions from users without the reputation to upvote to incentivizing answers in unpopular tags.

  • 1
    "gives those of us fighting our shadows a twisted satisfaction when we rack up yet another 0-score accepted answer" well put – apaul Jul 31 '13 at 2:21
  • Thank you @jmac. Congratulations for being an Unsung-Hero. Let's suppose some OPs had upvoted your answers, but the questions still would be low traffic (e.g. because the unpopular tag as you cited), Would not still be a lonely work? Regardless the OP upvote you would continue to be a true Unsung-Hero. – Andre Silva Jul 31 '13 at 2:21
  • This is an accurate explanation for why the badge exists +1 – asheeshr Jul 31 '13 at 2:41
  • @AndreSilva Many of the OPs never come back to StackOverflow. Here is the reputation of the last 10 users I answered (minus upvotes for the question asked): 101, 20, 1, 110, 1, 2737, 7, 138, 94, 4892. So 20% of my last 10 questions were asked by 100% new users. I would write a query to see the reputation of askers whose questions I've answered, but I haven't the foggiest idea how. – jmac Jul 31 '13 at 2:42
  • I like upvotes by the OP because I am not a professional programmer. I dabble. An upvote tells me it worked/is correct, which is nice. If you look at my answers, you will notice that I usually test my code (provide a working sample), show the output in the post, and put real effort in to making sure it solves the issue. Seeing that someone appreciates that (with an accepted answer or upvote) is incredibly nice. As is, 44 out of my 132 answers are 0 score, no accepted answer. After putting in effort, ANY attention is nice and makes it less lonely. – jmac Jul 31 '13 at 2:48
  • Ok, not 100% of new users, right?. The point here is that considering the upvote that comes from the OP toward the badge brings no harm to the current status (IMO). However, I think the argument we are discussing would strengthen the attitude the badge tries to encourage (less gamification and more consensus by community). But the way this is going, it has been as much controversial as the other Unsung-Hero threads. – Andre Silva Jul 31 '13 at 2:50
  • @jmac, no objection (agree). – Andre Silva Jul 31 '13 at 2:53
  • 1
    @AndreSilva I'm saying that it isn't 100% new users who can't vote, and yet I still earned the badge without any hassle whatsoever. So it seems like your post is more, "I want to expand the number of people to get the badge" rather than realizing that those of us who earned the badge did so with little effort because of the nature of actually contributing to unpopular tags. I don't care if the OP's vote is included or excluded, I care more that it sounds like an attempt to make it easier to game a badge. – jmac Jul 31 '13 at 2:54
  • Yes, I think more people would earn the Unsung, but not so much more. I am just saying that if you answer low traffic questions, you can have your answer accepted and upvoted by the OP and no one else. And one would still merit the badge. That's all. – Andre Silva Jul 31 '13 at 3:08
  • @AndreSilva But the question becomes, "Why do we want more people to earn the badge?" If this will somehow give new users/unpopular tags more exposure, great! But if it just gives current users badges for doing absolutely nothing different from what they're doing now, it is just an elaborate attempt to game a badge that's tough to get as-is (because it requires rather thankless work long-term). – jmac Jul 31 '13 at 6:26
  • @jmac, the question is about improving the objective for what the badge was created for (one might disagree of the proposal, though). The arguments are written in question body. I am saying why can't the answer be considered "unsung" if the OP upvotes it (I am saying there is a very soft difference between zero accepted answer and +1 accepted answer (again, one might disagree). – Andre Silva Jul 31 '13 at 12:47
3

I would guess that the badge is set up to encourage people to take the time to help new users who don't yet have the reputation to up-vote.

From the original proposal, Badge suggestion: Unsung Hero (5 accepted answers with no votes):

Why do this, you ask? Well, one of the much-belabored issues of the SO model is that rep is more easily accumulated by facile participation in shallow, accessible, popular topics, while successful handling of difficult, high-expertise-required issues often enough goes unrewarded. This badge attempts to act against that tendency, interpreting someone who writes a lot of accepted-but-not-voted-on answers as actively helping people solve "unpopular" problems — likely, in large part, problems from new, low-rep users who lack even the power to upvote a good answer to their question, and so may go ignored by more rep-oriented users.

I think we should leave it be. New users need answers more than we need easy badge opportunities.

If the answer was upvoted and accepted you received the 15 rep from the accept and another 10 rep from the upvote, that's a grand total of 25 rep, your efforts have been rewarded, move on.

The spirit of UNSUNG Hero, notice the emphasis on unsung, is to offer some recognition to those who labor without the extra incentive.

  • Thanks for the answer. The requirement for upvoting is just 15 points of reputation. It is a very small window. Usually if the user is seeking to engage in the community he/she will be able to upvote very soon. Users who do not reach 15 points, generally asks and go away (without accepting). – Andre Silva Jul 31 '13 at 1:42
  • 2
    @AndreSilva that's exactly my point... If people look at brand new users as if they're not likely to upvote or accept whats the "motivation" to answer their questions? That's where the Unsung Hero badge provides "motivation" – apaul Jul 31 '13 at 1:49
  • Ok, but how one would feel motivated if they already expected to not have the answer accepted?. I can be wrong, but for what I have seen the zero score accepted answers do not come from <15 rep users. And, even if they come why would be a bad thing to consider the upvote that comes from the OP? – Andre Silva Jul 31 '13 at 1:56
  • @AndreSilva If the answer was upvoted and accepted you received the 15 rep from the accept and another 10 rep from the upvote, that's a grand total of 25 rep, your efforts have been rewarded, move on. The spirit of UNSUNG hero, notice the emphasis on unsung, is to offer some recognition to those who labor without the extra incentive. – apaul Jul 31 '13 at 2:03
3

Jmac's answer provides an accurate explanation for why the Unsung Hero batch exists in the way that it does.

One major aspect that you are missing is data. How do you know that the majority of askers upvote answers? Voting is anonymous and hence this cannot be accurately measured by any user (only devs can). I, for one, very rarely accept and upvote the same answer.

What is the point to not consider the +1 (upvote) that generally comes from the OP?

  • A lot of people are interested in measuring their progress towards badges. Currently, this can easily be measured by observation, and also through SEDE. If asker upvotes were to be not counted then the ability to measure progress would be lost (why? see next point).

  • If progress measurement were possible then it would be very easy to figure out whether the OP voted you up or not. This is a direct break in the anonymity of voting. If anonymity was lost, even for just asker upvotes, we would next have comments all over SO asking and demanding for upvotes. This will just be even more troublesome for new users.

  • Users may start ignoring the questions from users whom they may know to not upvote their answers. We would have the same problems that were being caused by the display of the accept rate.

  • If progress measurement is not available, then there are two side effects. One, a loss of possible motivation towards acquiring the badge as mentioned in comments. The other bigger issue is the surge in meta posts asking why they havent earned a badge, what are the criteria, i should have the badge, etc. This can in itself become a problem.


It has been mentioned that tracking statistics will be tough. I disagree.

  • Per question stats can be easily tracked. Look at the progress percent before answering. Then have a look after the OP responds to your answer. If no change, that means the OP hasn't voted and will probably not in the near future.

  • As far as aggregate statistics are concerned, if you happen to be answering questions in a tag where you are not expecting/getting votes from people other than the OP or more than 1 vote on answers, then there is a very high probability that very few users participate in that tag. What this means is that you will in all likelihood start recognizing users pretty quickly, and so will others. Communities tend to form around such tags frequently and users not voting will get identified pretty quickly through comment threads.

  • Thanks for the answer. First part, I did not say the majority of askers upvote the question ( I said it is random). What I said is that, when there is just one upvote on the answer, the most cases this upvote comes from the OP. – Andre Silva Jul 31 '13 at 2:58
  • @AndreSilva I would tend to disagree with that. A lot of my accepted 0-vote questions get an upvote months after I made it (months after the answer was accepted). I could go through my reputation history to confirm if you'd like, what the ratio is of accepted answer+upvote at the same time, vs. at different times. – jmac Jul 31 '13 at 3:00
  • Not necessary, I have no further comments on your answer. And you are right, I do not have data (low rep). I am just guessing, and I can be wrong. – Andre Silva Jul 31 '13 at 3:02
  • 1
    I disagree with the premise of your argument: losing the ability to track badge progress is not reason enough to discard this suggestion. You're correct that progress tracking would have to be disabled, but so what? That seems like a perfectly reasonable tradeoff to make the badge actually attainable and add consistency to the kinds of user behavior encouraged. – Esoteric Screen Name Jul 31 '13 at 3:08
  • 1
    @EsotericScreenName The loss of the ability to track progress would dissuade people from pursuing it, which would tend to go against the very reason the badge was created in the first place, no? – jmac Jul 31 '13 at 3:56
  • 1
    @jmac I don't know. I just kind of got an unexpected Tenacious badge one day. – michaelb958 Jul 31 '13 at 4:01
  • @michaelb958 That is a good point -- I didn't track my progress either and was more blindsided by the badge, but I don't think everyone necessarily shares my approach (or yours) and some people would want the tracking... – jmac Jul 31 '13 at 4:09
  • Please, attempt the question refers to "maximum" of +1 score. If the answer has zero score, but it is accepted it is ok too. Accepting the question continues to be the main issue. It happens the OP will not continue to be asked "please, accept my answer but do not upvote it". – Andre Silva Jul 31 '13 at 14:06
2

Vote up is a privilege requiring 15 rep. If the asker of the question has less, they simply can't upvote.

In low traffic tags (where upvotes are infrequent), askers may have trouble getting there. One may ask two good questions (those who put effort into "doing their homework" prior to asking, tend to have less questions), get at most two upvotes from answerer, totaling at +10, and still be not there.

  • Per my recollection, this was the case eg in tag in about 2009-2010, even reasonably well written questions weren't getting much upvotes, simply because nobody else was interested besides asker.

From this perspective Unsung Hero motivates answering questions from low rep users who can't yet upvote (they can accept though, and this is supposed to somehow indicate answer quality).


It is also worth noting that in low traffic tags, newcomers may also lack experience / examples of how to use Stack Overflow, so even after acquiring required reputation, they might still not know when, how and why to upvote.

  • IMO, the same way new users can lack experience to upvote, they can lack to accept an answer. I agree unsung motivates people answering questions from new users, but it also motivates answering on threads with low traffic/attention from all community. Thanks for replying. – Andre Silva Jul 31 '13 at 13:22

You must log in to answer this question.

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