It has come to my attention that the amount of reputation awarded for questions can be adjusted on a per-site basis. For example, Stack Apps awards +10 reputation for question upvotes. As you probably already know, we currently award +5 reputation, as do most Stack Exchange sites, including Stack Overflow.

The lower reputation amount makes sense in the context of a Q&A format since it takes less effort to ask a question than it does to answer it. However, we're a little different in that quality challenges can be far more difficult to create than quality answers.

Thus I propose that we raise the amount of reputation we award on questions from +5 to +10.

This was proposed 4 years ago and was deemed to be a good idea but no action was taken. It was also addressed on Meta.SE and was met with hesitation there. I think it would be worth a discussion now on our site, particularly since it has become apparent that the modification is possible.

Note that the change would apply retroactively; reputation would be recalculated for all users who have ever posted a question on this site. That is, in the system, it will be like the user earned the reputation the day the upvote was given, even if it was in the past. The 200 reputation per day cap would still apply on those days though. For some users, this may result in many thousands of reputation points gained.


Note: Due to near unanimous support for this proposal, I've converted this from a community discussion to a formal feature request that will hopefully be addressed by Stack Exchange.

  • 29
    \$\begingroup\$ The second to last paragraph is particularly important to note. Specifically, rep recalcs could be on the order of many thousands. \$\endgroup\$ – Doorknob Dec 7 '15 at 20:43
  • 5
    \$\begingroup\$ This seems very helpful - there's less to be "gained" from posting questions as of now, so there's far fewer questions than answers. +1 because this would definitely increase the incentive to post questions. \$\endgroup\$ – Addison Crump Dec 7 '15 at 20:53
  • 52
    \$\begingroup\$ Calvin's is crying from happiness \$\endgroup\$ – Optimizer Dec 7 '15 at 20:55
  • 28
    \$\begingroup\$ Massive support. Writing good questions is just as hard as writing answers, if not harder. The rep gained should be reflective of that. \$\endgroup\$ – Mego Dec 7 '15 at 20:57
  • 1
    \$\begingroup\$ @FlagAsSpam having far more answers than challenges is a good thing, especially here where every new language can add something interesting to a challenge, but rewarding the hard task of writing a good challenge is the important part. \$\endgroup\$ – Martin Ender Dec 7 '15 at 21:02
  • 6
    \$\begingroup\$ @FlagAsSpam Unhealthy isn't the right word. It's a good question/day rate but it's below the threshold required for graduation. Unhealthy would imply that we would be in danger of being shut down, which we aren't. \$\endgroup\$ – Alex A. Dec 7 '15 at 21:04
  • 10
    \$\begingroup\$ Would this mean I finally get my Mortarboard badge for the day I got +195 rep off of one question? :D \$\endgroup\$ – ETHproductions Dec 8 '15 at 0:33
  • 5
    \$\begingroup\$ I don't disagree with the proposal. But the argument about it taking more time to write questions is not very convincing. The amount of rep you typically get for a post on any SE site I have seen has nothing to do with the amount of effort. On this site, I have spent several days on some performance challenges, and they normally get very few votes. On the other hand, easy answers to trivial challenges get the most votes. So for answers, I would state that the amount of rep gained has an inverse correlation to the amount of time spent on it. At best, rep is unrelated to effort. \$\endgroup\$ – Reto Koradi Dec 8 '15 at 0:38
  • 4
    \$\begingroup\$ @Geobits I left a comment replying to Shog's dismissal of the comparison, because the Q&A argument applies even less to PPCG than StackApps. Anyway, I think the idea of this post was to gather more up-to-date opinions on this, because the original post was from 3 days after this site launched, so despite the large number of upvotes it might not be too convincing to make a change based on that. \$\endgroup\$ – Martin Ender Dec 8 '15 at 14:20
  • 3
    \$\begingroup\$ I'm believe I'm the user who has the most to gain from this and honestly I have no strong feelings either way. I'm happy with the current system, though I suppose I can't complain about a rep windfall. In my experience a decent question with decent answers can readily get double the votes of the top answer, so the rep gain is about the same. I guess the crux is that most people view writing questions as harder than writing answers. I'm apparently the exception there so perhaps I'll just stay out of this conversation. \$\endgroup\$ – Calvin's Hobbies Dec 18 '15 at 2:01
  • 6
    \$\begingroup\$ Nevermind, I got my Mortarboard badge. You can all go home now. ;) \$\endgroup\$ – ETHproductions Dec 18 '15 at 3:30
  • 3
    \$\begingroup\$ Can we simulate the amount of rep gain if this is implemented? Then we can see how much this actually affects the reps of users, and have further evidence (or counter) to this proposal. \$\endgroup\$ – justhalf Dec 31 '15 at 6:50
  • 5
    \$\begingroup\$ @tuskiomi According to Grace Note, our Stack Exchange community manager, this is unlikely to happen without a network-wide change, which seems similarly unlikely. \$\endgroup\$ – Alex A. Nov 10 '16 at 21:43
  • 2
    \$\begingroup\$ @programmer5000 I just contacted SE \$\endgroup\$ – Christopher May 15 '17 at 23:41
  • 2
    \$\begingroup\$ So this was unveiled by SE without any user input. meta.stackoverflow.com/questions/391250/… \$\endgroup\$ – qwr Nov 14 '19 at 20:49

11 Answers 11



I say that we should go ahead with this change.

Apart from simply correctly appreciating the massive amount of effort put into writing good challenges, this would also be a very nice added incentive to start writing questions, especially for existing users who now want to earn more rep. :)

  • 8
    \$\begingroup\$ Prepare for the wave of up votes. :P I don't see why anyone would disagree (yet, there's bound to be someone). \$\endgroup\$ – Addison Crump Dec 7 '15 at 21:12


PPCG is different from most other sites on Stack Exchange in that it is not a Q+A site; in fact, many questions that ask for help (other than "how to golf in ____") are marked as being not on topic. The thing that makes PPCG unique is that the interaction between question asker and question answerer is important. PPCG is not a site where people ask questions looking for help (mostly), but a site where the question askers want to challenge the answerers to solve their problem.

When I write questions, I don't write them because I want help solving them. I ask them because I want to challenge others to solve them, and so I can see the interesting things you can do with all sorts of programming languages, from the classic languages of Python, Java and PHP to the golfing languages of Pyth, CJam, and Seriously to the esoteric and unusual languages of Brainfuck, ><>, and Labyrinth, among many many many others. Questions are designed to be high-quality so they attract high-quality answers. In my opinion, good questions should have reputation systems equivalent to that of answers.

  • 6
    \$\begingroup\$ I like the elaboration. \$\endgroup\$ – xnor Dec 10 '15 at 17:08
  • 2
    \$\begingroup\$ I feel that adding further reasons for agreeing with an existing opinion is an important part of a discussion, and this answer may influence how people vote on other answers. \$\endgroup\$ – trichoplax Dec 18 '15 at 18:18

Yes, but only if downvotes scale to -4

Peter Taylor brought up a very good point in a comment: There are many questions which would change from being worth negative rep to positive rep should this change be implemented without scaling downvotes accordingly.

Currently, downvotes on both questions and answers are worth -2 reputation. I propose that downvotes on questions scale to -4 rep, with answer downvotes remaining unchanged.

This is a simple doubling of loss to match the doubling of gain. While it may seem harsh, it has the benefit of not awarding undeserved reputation if the question reputation value were to increase. It also makes it so it doesn't require 5 downvotes to counter a single upvote in terms of reputation, which would be the case if the -2 was unchanged.

  • 21
    \$\begingroup\$ I don't disagree that downvotes should scale accordingly, but I wouldn't consider it a deal breaker. The benefits of raising the upvote rep outweigh the downsides of altering the proportion. \$\endgroup\$ – Dennis Dec 9 '15 at 3:17
  • 28
    \$\begingroup\$ I disagree with the notion that downvotes are for countering the reputation gained from upvotes. Downvotes are for filtering out bad content. This change also seems to support what was happening before: punishing people for asking questions instead of writing answers. Answers don't carry this additional risk of more reputation loss, and actually instead already discourage downvotes more than questions by taking rep from the downvoter. On a somewhat unrelated note, I don't really think "many" accurately describes that about 1% of questions will change net reputation worth. \$\endgroup\$ – FryAmTheEggman Dec 9 '15 at 17:37
  • 9
    \$\begingroup\$ I don't really understand. I feel like this would be a more complete explanation: "There are many questions which would change from being worth negative rep to positive rep should this change be implemented without scaling downvotes accordingly, and this is a bad thing because [reason(s)]." \$\endgroup\$ – Rainbolt Dec 10 '15 at 20:08
  • 3
    \$\begingroup\$ @Rainbolt, the obvious reason is that increasing rep given for upvotes and leaving downvotes untouched increases the incentive to write good questions and also increases the incentive to write bad questions. People upvote all kinds of crap on this site, and, with rare exceptions which get piled into the ground, it's quite unusual for a question to get more than a couple of downvotes. \$\endgroup\$ – Peter Taylor Dec 14 '15 at 12:29
  • 3
    \$\begingroup\$ Why not also double the amount of reputation lost for downvotes on answers? Why should a downvote be more punishing for questions than for answers? \$\endgroup\$ – Rainbolt Dec 14 '15 at 14:10
  • \$\begingroup\$ @Rainbolt I suppose you could argue that bad questions are more detrimental to the site's overall quality than bad answers, but I don't think I've actually seen Peter saying downvotes on answers shouldn't be doubled. ;) \$\endgroup\$ – Martin Ender Dec 16 '15 at 12:34
  • \$\begingroup\$ @MartinBüttner I agree; I think bad questions are more detrimental than bad answers. I also think good questions more beneficial than good answers. If downvotes on a question count extra, then I think upvotes should as well. \$\endgroup\$ – Rainbolt Dec 16 '15 at 14:06
  • \$\begingroup\$ @MartinBüttner, my immediate reaction was along the lines of "Hmm, that's certainly an idea worth thinking about". But I forgot that it was on my list of things to think about. \$\endgroup\$ – Peter Taylor Dec 17 '15 at 15:00
  • 4
    \$\begingroup\$ Update nearly a month later: I'm actually not so sure I agree with this, despite being the one who posted it. See @FryAmTheEggman's comment for an excellent counter. Both the moderators and the community can step up an effort to weed out bad content with closing, deletion, flags, etc. Especially since this change would increase in the number of users with trusted user privileges. There are of course downsides, as Peter has mentioned, but as a whole I think this change would be an incredible benefit to the community regardless of said downsides and regardless of any reputation loss changes. \$\endgroup\$ – Alex A. Dec 31 '15 at 16:18

I'm late to this party, so I probably won't get many votes.

Upvotes +15, Downvotes -5

For each one challenge, approximately 9.9 solutions are written. It seems obvious to me that challenges are far more valuable to the site than solutions. Make upvotes count for +15.

Fifteen reputation for an upvote will reward poorly written challenges. We have a Sandbox dedicated to challenges. I don't see why anyone would forego a free critique, but if you do, then you really have no excuse. Make downvotes count for -5.

The result is that challenges that can maintain a ratio of 1 upvote for every 3 downvotes will break even. That seems reasonable to me.

  • 3
    \$\begingroup\$ I think this is a good balance. It makes sense that good questions should be rewarded. I currently upvote almost 60% of challenges, and if upvotes give +15, I'll be able to be more selective with my votes, which would benefit the site. \$\endgroup\$ – lirtosiast Dec 18 '15 at 0:28
  • 9
    \$\begingroup\$ +1 for the +15, -1 for the -5. The sandbox isn't perfect. \$\endgroup\$ – Ypnypn Dec 18 '15 at 0:35
  • 4
    \$\begingroup\$ @Ypnypn The sandbox isn't perfect, but at the very least, I would expect the average PPCG challenge to be of higher quality than the average StackOverflow question simply because we have created a way to make it so. \$\endgroup\$ – Rainbolt Dec 18 '15 at 14:05
  • 18
    \$\begingroup\$ +15 is definitely too much. \$\endgroup\$ – nicael Dec 25 '15 at 15:39
  • \$\begingroup\$ @nicael Your comment got exactly 15 upvotes \$\endgroup\$ – FireCubez Nov 6 '18 at 18:56
  • \$\begingroup\$ @FireCubez nice! almost three years passed, after all \$\endgroup\$ – nicael Nov 15 '18 at 20:32
  • \$\begingroup\$ We have a Sandbox dedicated to challenges. I don't see why anyone would forego a free critique, but if you do, then you really have no excuse. Make downvotes count for -5. Then why make it +15. I'd be in support of +10/-5 \$\endgroup\$ – MilkyWay90 Jun 17 '19 at 15:13

One time bonus for Nice Questions

This probably won't be implemented. But it seems to be exactly what all you want:

  • Easy questions getting a lot of upvotes won't be affected much.
  • Questions having a lot of upvotes and downvotes won't be affected much.
  • New users writing bad questions won't be affected much before understanding this situation.
  • Questions having 10 upvotes and no downvotes would be awarded more.
  • Writing many good questions is preferred to writing one extremely popular question, so it really encourages writing questions, instead of posting the idea you had anyway or just being lucky.
  • It wouldn't be much worse to upvote the question for the good answers in it.

If changing reputations for upvote / downvote isn't the perfect solution, I'd think it is understandable not making any changes only to revert the changes after problems happen.

  • 13
    \$\begingroup\$ While I think it's an interesting idea, I'd be surprised if it was actually possible. I think the system is set up to only reward badges for milestones (such as +10 on a question) and only reward reputation on a per-vote basis. \$\endgroup\$ – Alex A. Dec 21 '15 at 0:39

As of yesterday, November 13th 2019, this is now in effect network wide:

We’re changing the reputation earned from getting a question upvote to ten points, making it equal to the reputation earned from an upvote to an answer.

  • 7
    \$\begingroup\$ Can we up it to +15 now? :P \$\endgroup\$ – Jo King Nov 14 '19 at 8:59
  • 1
    \$\begingroup\$ Or we could make them both +15? :P \$\endgroup\$ – a'_' Nov 14 '19 at 14:16
  • \$\begingroup\$ Yay! Something we wanted happened! \$\endgroup\$ – isaacg Nov 15 '19 at 17:47


But not in questions.

Tips questions don't fit the classic PPCG question and are also quite easy to write, as compared to a challenge. Just look at the general structure.

This is my general structure when I write questions here:

cheesy explanation


task description; 1 or 2 short sentences


detailed description of input rules etc


detailed description of output rules etc


  • unordered
  • list
  • of rules
  • ending in winning criterion


list of examples

cheesy justification of shortest code

And this is the general structure I have seen for questions:

Do anyone have any tips for golfing in language? Please post one tip per answer.

There is a vast difference, as you can see. So if we start offering 10 rep on all questions, we reward people for writing 30 words asking for help, we are going to end up with a mass influx of questions about obscure languages, just to get the rep.


  • +5 rep for a question upvote
  • +10 rep for every other question upvote
  • -2 rep for a question downvote

What do you think?

  • \$\begingroup\$ This sounds like a good idea to me, but I don't think it's possible without making major changes internally to the SE code. \$\endgroup\$ – ETHproductions May 22 '17 at 20:46
  • \$\begingroup\$ @ETHproductions something I've wanted since I joined was that tips pages be part of meta as they aren't actually challenges. If we migrate them here, then we can implement this but without tips getting any rep. \$\endgroup\$ – caird coinheringaahing May 22 '17 at 20:48
  • 2
    \$\begingroup\$ But then the tips themselves won't earn any rep, and also won't be as visible... \$\endgroup\$ – ETHproductions May 22 '17 at 20:58

No / Maybe

I'm guessing that this might be an unpopular opinion, but I thought I'd bring it up.

As much as I agree that a massive effort is put into writing good challenges and they should be rewarded more, this would also go the other way. The problem is generally stated here, trivial questions generally get flooded with up-votes especially with help from HNQ.

I'd say the best solution is instead of changing the reputation awarded for questions, is to switch the focus to using question votes more effectively. By this I mean, upvoting what you feel is a high-quality, good challenge, rather than a challenge simply because it's easy.

My point is, as much as I do agree high-quality challenges should be rewarded more, this would also affect trivial challenges significantly also, with a single upvote requiring 5 downvotes to counter. So my proposal is:

No, unless downvotes are worth much more

  • 6
    \$\begingroup\$ I don't think that questions which get tons of upvotes via HNQ will be affected by this change at all, because their authors will likely rep cap on those days anyway. \$\endgroup\$ – Martin Ender Dec 8 '15 at 8:26
  • 3
    \$\begingroup\$ There are 42 questions which would change from being worth negative rep to positive rep. \$\endgroup\$ – Peter Taylor Dec 8 '15 at 8:47
  • 4
    \$\begingroup\$ I do wish downvotes did more, and it would be unfortunate if they were made yet weaker, but not enough to sink the proposal for me. But in a world where devs would tweak rep values for us, could harsher question downvotes be an option? \$\endgroup\$ – xnor Dec 8 '15 at 11:08
  • 3
    \$\begingroup\$ I would upvote this if it said 'yes, but downvotes should scale accordingly'. \$\endgroup\$ – Sanchises Dec 8 '15 at 11:42
  • \$\begingroup\$ @xnor I'm of the same opinion but I've been told that harsher downvote penalties have been proposed and shot down in the past. \$\endgroup\$ – Alex A. Dec 8 '15 at 16:57
  • 1
    \$\begingroup\$ I agree that the reputation lost for downvotes should increase but I believe that attempting to change how people vote is pretty much impossible. \$\endgroup\$ – Alex A. Dec 8 '15 at 17:00
  • \$\begingroup\$ @sanchises I posted an answer to that effect. \$\endgroup\$ – Alex A. Dec 8 '15 at 22:25
  • 3
    \$\begingroup\$ @AlexA. Do you mean harsher downvote penalties have been proposed and shot down for PPCG, or for SE in general? \$\endgroup\$ – xnor Dec 8 '15 at 22:30
  • \$\begingroup\$ @PeterTaylor If you adjust it to count double downvote costs, then that number goes to 0. \$\endgroup\$ – LegionMammal978 Dec 9 '15 at 11:41
  • 3
    \$\begingroup\$ Downvotes are not just about reputation. They are a message to the author. If the author chooses to ignore downvotes then a larger reputation penalty will not necessarily change that. What is important for poor quality questions is the ability to vote to close and/or delete. \$\endgroup\$ – trichoplax Dec 18 '15 at 18:24
  • \$\begingroup\$ Boooo!!! -1 I want rep! \$\endgroup\$ – caird coinheringaahing Jun 4 '17 at 0:06

Alternative: Give more rep every time someone answers a challenge

I know I'm late to respond here so the wagon may have already left the station, but it seems to me we want to award good, enjoyable challenges, not just anything posted. An indication of a good challenge is the number of people who take part in it. So instead of making the initial rep for a challenge higher, we should award additional rep every time someone posts an answer to a challenge.

  • 23
    \$\begingroup\$ I see two problems with this. 1. That would require modifying the code rather than just a config parameter - not going to happen. 2. I think you're wrong. Having a lot of answers is often a sign of a trivial challenge, which isn't the same as a good challenge or an enjoyable challenge. \$\endgroup\$ – Peter Taylor Dec 21 '15 at 16:57
  • 1
    \$\begingroup\$ @PeterTaylor Is a trivial challenge bad? A question that brings a novice in the door is probably going to be good for the long term health of the site. \$\endgroup\$ – corsiKa Dec 24 '15 at 7:50
  • 2
    \$\begingroup\$ @corsiKlauseHoHoHo Sometimes novices get to answer the easy challenges, but other times experienced users are too fast and take every language's optimal solution before the challenge even gets on HNQ. I'm guilty of this-- I think once I answered an easy qiestion in Pyth, APL, TI-BASIC, and Matlab in the space of an hour. \$\endgroup\$ – lirtosiast Dec 27 '15 at 4:25
  • 2
    \$\begingroup\$ I don't like the idea of linking question rep to the rep from its answers. Although it seems useful for some questions, there exist questions for which I would like to upvote answers but downvote the question. I see it as important to keep the two types of post separate for voting purposes. \$\endgroup\$ – trichoplax Dec 30 '15 at 15:33

Yes, but make it cost reputation to downvote questions

I'm definitely all for the increase in reputation for upvotes on questions, and whether or whether not downvotes scale is also up for debate.

But, I think it should cost reputation to downvote questions, and should probably be -1, the same as for answers. The reason it currently does not cost any reputation to downvote is because this motivates users to vote on questions more, which doesn't make sense if this site isn't really about questions at all, like other SE Networks.

I think it should cost -1 reputation to downvote questions

  • 6
    \$\begingroup\$ I'd be surprised if it was possible given that, to my knowledge, no sites in the network detract the voters' reputation for downvoting questions. That kind of implies to me that it just isn't part of the system, though I have no hard evidence. \$\endgroup\$ – Alex A. Dec 22 '15 at 17:54
  • 2
    \$\begingroup\$ @AlexA. I'm pretty sure it used to be part of the system, but it was removed so it actually might not be possible now \$\endgroup\$ – Jojodmo Dec 22 '15 at 17:55
  • \$\begingroup\$ Jojodmo, I think only downvoting answers subtracts reputation. \$\endgroup\$ – Timtech Feb 26 '16 at 16:39
  • \$\begingroup\$ Edited to fix my vote (mistake :P) \$\endgroup\$ – Christopher Sep 14 '17 at 21:19
  • \$\begingroup\$ Wait what? That doesn't already happen? I thought it did \$\endgroup\$ – FireCubez Nov 6 '18 at 18:59
  • \$\begingroup\$ This is basically just supporting the upvotes. If a challenge is bad, let's not deter everyone from downvoting it. \$\endgroup\$ – MilkyWay90 Jun 17 '19 at 15:14

No, because questions already give more rep points than answers

Just look the number of upvotes for any question and compare to the number of upvotes for the related answers.

Here is a query showing how our top users get more reputation (on average) from a question than from an answer. Not 'much more' really, but more.


  • 4
    \$\begingroup\$ You might want to include something like a SEDE query to back this up. I can count the days I've rep capped from question upvotes on one hand. \$\endgroup\$ – Martin Ender Dec 12 '15 at 1:19
  • 18
    \$\begingroup\$ Why would earning more reputation for questions than answers be a negative? \$\endgroup\$ – Alex A. Dec 12 '15 at 4:28
  • 2
    \$\begingroup\$ I too can count it on one hand! Unfortunately, I don't need hands to count the days I've rep capped from answer upvotes... \$\endgroup\$ – Stewie Griffin Dec 14 '15 at 8:04
  • \$\begingroup\$ @AlexA.we need a balance. Else, why not 1000 points for each upvote? \$\endgroup\$ – edc65 Dec 14 '15 at 8:28
  • \$\begingroup\$ I'm not sure what balance you mean, but we already have 10 times more answers than questions. I think if anything, encouraging more questions would help balance things. \$\endgroup\$ – Alex A. Dec 14 '15 at 19:06
  • \$\begingroup\$ @AlexA. as little as 10? I hope we have some bad question with few or no answer, and much more than 10 answer for question for the other, good ones \$\endgroup\$ – edc65 Dec 14 '15 at 20:20
  • \$\begingroup\$ Yes, that is indeed the case. \$\endgroup\$ – Alex A. Dec 14 '15 at 20:47
  • \$\begingroup\$ @MartinBüttner query added \$\endgroup\$ – edc65 Dec 15 '15 at 8:48
  • 7
    \$\begingroup\$ @edc65 I don't think you can just compare average rep per post. It seems that a question gives on average about 1.5 times as much rep as an answer... but I think on average a high-quality question takes much more than 1.5 times the effort of a decent answer. (as evidenced by the question/answer ratios of the top users, except CH). \$\endgroup\$ – Martin Ender Dec 15 '15 at 8:56
  • 1
    \$\begingroup\$ I hope I am not the only one who appreciates the irony that the only "Absolutely No" answer is from a user whose entire rep is from answers. \$\endgroup\$ – Mego Dec 17 '15 at 10:40
  • \$\begingroup\$ @Mego yes, I would not gain anything from this change, That's not my only motivation thogh \$\endgroup\$ – edc65 Dec 17 '15 at 10:44
  • 1
    \$\begingroup\$ @edc65 Oh, no, I did not assume that was your motivation. I just found it vaguely amusing that you don't have any rep from questions. \$\endgroup\$ – Mego Dec 17 '15 at 10:45
  • \$\begingroup\$ I looked myself up on that table. My average question score is 94 and my average answer score is 97, so apparently I'm a (slight) counterexample. :P \$\endgroup\$ – Joe Z. Dec 28 '15 at 3:10

You must log in to answer this question.

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