Timeline for The Stack Exchange reputation system: What's working? What's not?
Current License: CC BY-SA 4.0
34 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Mar 17, 2023 at 15:53 | comment | added | user400654 | @IñakiViggers No, it's not replaced at all, it's entirely lost. that's the entire point. If you don't accept anonymous feedback that feedback doesn't exist. It doesn't magically result in more comments. | |
| Mar 17, 2023 at 15:51 | comment | added | Iñaki Viggers | That metric is replaced with something more useful: articulated remarks that prompt others to assess issues which hitherto might have gone unnoticed. | |
| Mar 17, 2023 at 15:45 | comment | added | user400654 | @IñakiViggers Yes, however we lose a valuable metric that is used to determine whether or not a post is actually useful, if we just ignore any feedback from users who don't wish to enter into a confrontational interaction with someone else on the internet. | |
| Mar 17, 2023 at 15:43 | comment | added | Iñaki Viggers | "Limiting that kind of feedback to users leaving comments is a barrier that stops most people from leaving feedback at all." That is exactly what should happen. Feedback cannot lead to improvement of a post if the downvoter doesn't bother to articulate the reasons for downvoting, which sometimes reduce to animosity toward the author. | |
| Mar 14, 2023 at 19:47 | comment | added | jpyams | I remember when I got a bunch of downvotes on an answer back in 2015. I was still a new contributor and it did hurt a bit. But reviewing my answer, it was pretty stupid and I hadn't put two seconds of thought or research into it. The system worked, and from then on I actually thoroughly thought through any question, answer, or edit. | |
| Mar 14, 2023 at 5:33 | comment | added | user400654 | Yeah I'm not a fan of suggested edits getting rep much less any other moderation action. | |
| Mar 14, 2023 at 2:52 | comment | added | Louis Go | Moderation and reputaion should be separate. However Encouraging duplicate close is double edged.. It might inflict much more frustration for false vote to closed. | |
| Mar 13, 2023 at 15:26 | comment | added | JamieB | @KevinB I'm certainly not arguing that all VTCs are wrong. But many are, which brings us back to the topic at hand: "high reputation" does not always mean "good at moderating". When you see questions closed with a bunch of upvotes and an answer by the local moderator, then you have to start questioning where the reputation system has left the rails and put the wrong people in charge of sealing off content. | |
| Mar 13, 2023 at 15:25 | comment | added | user400654 | @JamieB Perhaps because it's off topic and should be closed? Just because you have an opinion that it isn't off topic, doesn't mean your opinion is any more right than theirs. Closing questions is a collaborative effort; if on your stack there's a disagreement about what is and isn't off topic, that needs to be dealt with on the per-site meta. | |
| Mar 13, 2023 at 15:24 | comment | added | JamieB | @vlad Ah, interesting. That really makes it extra baffling why people go to VTC when what they should be doing is downvoting. The votes to reopen these wrongful closes gets tedious at times. | |
| Mar 13, 2023 at 15:04 | comment | added | VLAZ | @JamieB downvoting questions does not cost rep. Only downvoting answers does. So, your observation is off. | |
| Mar 13, 2023 at 15:04 | comment | added | user400654 | @JamieB eh, i think you're jumping through a few unnecessary assumptions there. question downvotes are free, so a close vote instead of a downvote makes no sense there, in terms of it being a reputation issue. | |
| Mar 13, 2023 at 15:00 | comment | added | JamieB | You mention rep encouraging answering rather than closing, but what really irks me is how reputation encourages closing rather than downvoting. I have seen good questions closed, despite numerous upvotes, simply because 5 people decided they couldn't answer it, therefore it was bad, but a downvote costs them rep whereas a VTC is free. | |
| Mar 13, 2023 at 1:44 | comment | added | Enfield Li | @John Bollinger You should totally post an answer about this penalty stuff. So broken | |
| Mar 11, 2023 at 4:19 | comment | added | starball Mod | "rather, remove the negative rep of receiving and giving downvotes" whoa there. I almost missed that while skimming. | |
| Mar 10, 2023 at 16:04 | comment | added | John Bollinger | Dunno whether it helps any, but I would also be ok with reframing the rep gain for post scores so that instead of +rep for upvotes and -rep for downvotes, we had +rep for net positive scores. The effect would be similar to balanced scoring for individual votes, but conceptualizing it differently may be useful to some. Also, in the form I just stated it, that would remove any penalty for net negative scores. | |
| Mar 10, 2023 at 15:49 | comment | added | user400654 | I've seen too many people say they don't downvote because it removes rep from the user too many times to want to increase the penalty. The penalty is a punishment when we should instead be encouraging the user to improve. Penalties should really only come into play once it has been determined that the user isn't improving. Seeing a -1 on their question score is plenty enough to get users who would be impacted by a loss of rep to want to do something about it. Those who don't care won't be influenced by anything we do. | |
| Mar 10, 2023 at 15:45 | comment | added | NotThatGuy | @JohnBollinger I've seen my fair share of people commenting and explicitly admitting to upvoting something because someone else downvoted it. I've seen downvotes have very little positive effects (in case of poor quality, the post usually ends up getting deleted anyway) and cause plenty of frustration with the site and users on the site. Making upvotes and downvotes symmetric will make all of that (and votes for bad reasons) worse. It may also aggravate the problem of posts having too many upvotes and too few downvotes, because users may be less likely to downvote. | |
| Mar 10, 2023 at 15:21 | comment | added | John Bollinger | @NotThatGuy, I am well aware of downvotes occurring for petty and / or arbitrary and / or just bad reasons. I have received my share. I have seen even more posts getting upvotes despite being poor quality or downright wrong (and I've received one or two of those, too). I think the DV / UV asymmetry is more problematic. | |
| Mar 10, 2023 at 10:32 | comment | added | NotThatGuy | @KevinB "The primary penalty of downvotes at the moment is the user loses the ability to post entirely if they do not improve" - for new users, perhaps. Downvotes may also incentivise users to put more effort into what they post, and to improve their posts (in theory, possibly, although in practice I've found downvotes to do little more than breed resentment and frustration). | |
| Mar 10, 2023 at 10:21 | comment | added | NotThatGuy | @JohnBollinger "A downvote should fully cancel an upvote" - good idea in theory, but it's problematic in practice. People downvote for all sorts of petty reasons. Getting downvoted for seemingly no reason already sucks, that'll make it suck more. The unequal reputation probably isn't ideal, but most users are human, so we should be considerate of that. If downvotes were more limited, this may be a better idea. The bigger issue, perhaps, is that one can have a negative scoring post that still gives you reputation (this might be easy enough to fix by just ... not having it give you reputation). | |
| Mar 10, 2023 at 2:26 | comment | added | skomisa | @Milliways See Why do you lose reputation for downvoting answers? for some explanations. I have no strong feelings either way, but I've always been more puzzled by the inconsistency that downvoting answers is penalized, but downvoting questions is not. | |
| Mar 10, 2023 at 0:48 | comment | added | Milliways | I fail to understand WHY downvoting a poor answer penalises those who downvote. This is a disincentive to flag poor answers. | |
| Mar 10, 2023 at 0:41 | comment | added | Milliways | On the raspberrypi.stackexchange.com site the +10 -2 causes anomalies. There is a serial upvoter (practically every question gets upvoted, even those with multiple close votes). I don't particularly care up reputation but this defeats the need for users to have reputation to Comment. | |
| Mar 9, 2023 at 22:41 | comment | added | Andras Deak -- Слава Україні | I agree that +10 vs -2 for up/down votes is a huge issue. There's a looot of users who leave a stream of crap, but if you get +2/-2 on a post you're in the net positive. Considering that it takes almost zero effort to get enough reputation to upvote, and that clueless masses will upvote literally anything that has an up arrow next to it, this is a pretty bad premise. | |
| Mar 9, 2023 at 22:18 | comment | added | Mark | "What's broken about it, particularly on Stack Overflow, is it encourages users to answer posts that should be closed as a duplicate or off topic rather than voting to close because one is rewarded and the other isn't" - I completely agree with this. | |
| Mar 9, 2023 at 21:28 | comment | added | user400654 | The primary penalty of downvotes at the moment is the user loses the ability to post entirely if they do not improve. I think that's penalty enough, generally | |
| Mar 9, 2023 at 21:25 | comment | added | John Bollinger | I wasn't happy about the rep cost of receiving a downvote being reduced from 10 to 2. A downvote should fully cancel an upvote. I'm even more not happy with the idea of removing the rep penalty altogether. There's nothing wrong with penalizing people for posting bad content, especially when the penalty can easily be canceled by deleting the content. On the flip side, I'm fine with paying rep to cast a downvote, but I'd also be ok with removing that cost, I think. Consider that the rep cost for downvotes offers some protection against wanton downvoting. | |
| Mar 9, 2023 at 21:22 | comment | added | T.E.D. | So no, last I checked, SO had the exact opposite problem to the one related in this question, IMHO as a side effect of it having orders of magnitude more users than the reputation system's designers anticipated. | |
| Mar 9, 2023 at 21:19 | comment | added | T.E.D. | I don't hang out on SO as much as I used to, but back when I did (and it was using the same rep system as today), the userbase was far far too quick to close questions as dups. All it takes is a few users who don't quite understand the question, but think they do, and SO has thousands of users. I once had the fun nightmare scenario of suddenly starting to get downvotes on a well-recieved answer years later because it had wrongly been closed as a dup, then merged with the supposed (but actually unrelated) "dup", where my good on topic answer was made retroactively completely off-topic. | |
| Mar 9, 2023 at 20:49 | comment | added | user400654 | ideally, this would be combined with some of the higher reputation privileges moving to an alternative method for attaining them, see the "abilities" or "roles" suggestion in Spevacus's answer. the point of all this would be to reduce the gamification slightly while removing barriers and the perceived negatives of moderation actions. going to +3 would be a bit extreme, I'm not certain it'd need to actually change from +10, but I'd definitely want to see the -2 and -1 removed. | |
| Mar 9, 2023 at 20:42 | comment | added | wimi | "reduce the incentives of accept votes and upvotes" how can you do that, when those are the only sustainable way to gain reputation? If we change the "+10" to "+3", all scores will be about a third of what they are now and the incentive will be the same. (Bounties will of course also be reduced by the same amount because the capital for bounties also comes from upvotes and accepts). | |
| Mar 9, 2023 at 17:20 | history | edited | Dharman | CC BY-SA 4.0 |
edited body
|
| Mar 9, 2023 at 15:56 | history | answered | user400654 | CC BY-SA 4.0 |