27
$\begingroup$

I am newer to the SE community so forgive any naivete. I suspect this is a recurring issue, but I regularly come across users whose reputation continually increases despite many incorrect, or incomplete, pseudo answers (answers that dont contain references, list equations without explanation, a single picture, etc.).

By observation, this type of poster gains reputation by answering a lot of questions poorly. In many cases, their answers are not overtly incorrect (deserving downvotes) but seem to target lower quality (homework type) questions, hoping to receive a statistical percentage of upvotes from incomplete, pseudo answers. I actively downvote incorrect answers when I feel qualified to do so- While downvotes do discredit these answers, they hardly discredit the poster or deter incomplete, opportunistic answers.

For example, an [unnamed-user] answered 67 questions in 14 days, earning +27 votes and -25 votes (net +2 votes)- this results in a net +550 reputation.

This practice exploits the fact that a single upvote offsets 5 downvotes! In my opinion, it undermines the system and deflates the reputation of legitimate users. In other words, (from the perspective of a new user) their answers carry more weight, as reputation should reflect the communities endorsement. I argue that:

  1. Their reputation carries authority that misinforms
  2. Opportunistic answers perpetuate low quality questions
  3. They might gain authority to 'vote to close', etc.

Is this problematic? How is this currently addressed?

Edit:

I should note that the [unnamed-user] was placed on temporary suspension (presumably independently of this Meta post). In this light, and combined with dissenting opinions of long term users with valued perspective, I am convinced that the current system naturally responds to abuse.

That said, I'd like to clarify my position regarding the spread of misinformation with respect to reputation and voting.

The problem with that is that science is not a democracy. You shouldn't judge an answer by the number of upvotes or downvotes. That's letting other people do your thinking for you, and for all you know they're colluding and gaming the system. You should judge the answer for yourself, and follow up on the evidence and references and explanation

In my opinion, this (and similar) sentiments misstate the larger problem. Certainly, science is not a democracy! However, when an OP asks a question (good or bad), they don't know the answer. Incomplete answers (correct or incorrect), undermine the OP's (and the community's) ability to learn and objectively evaluate the answer. In effect, incomplete answers place additional weight on the democratic process (via reputation and consensus). In that sense, they 'misinform' simply because they are not adequately explained.
Similarly, if the OP does not feel qualified to evaluate an answer (often the reason the question was asked), they will likely defer to reputation. Reputation carries weight, especially for trivial questions with accepted physics: Whose answer would you bet on, @JohnRennie or @OneRepPoint?

We might discourage this practice or expose it by:

  • Increase downvote reputation losses by some agreed amount?
  • Introduce a visible statistic (such as Upvotes/Downvotes, Votes/Answer, etc)?
  • Negate reputation awarded when a question is closed, etc.(to prevent opportunistic answers and reduce incentives to answer homework like questions)?

These are large system changes and as @dmckee notes, NOT likely to happen.

What changes are possible within Physics.SE?

$\endgroup$
9
  • 5
    $\begingroup$ Upvoting because this is a good question to bring up, not as an indicator I think something needs to be done about it. $\endgroup$
    – David Z
    Nov 17 '15 at 4:34
  • $\begingroup$ This query can be a useful metric. I can't see an easy (or computationally cheap!) way to integrate it into the site, though. $\endgroup$
    – Emilio Pisanty
    Nov 17 '15 at 17:41
  • 5
    $\begingroup$ My two cents: I think if users stay on the site long enough, and are serious about learning physics, they learn to basically ignore the rep count and look instead at the answer given to their question and/or look at the profile for the quality of questions asked and answered previously by the user answering the question. $\endgroup$
    – user81619
    Nov 18 '15 at 19:45
  • 1
    $\begingroup$ This question might apply to the entire SE as a whole. Nevertheless, down-votes are often discouraging to new users, especially if they take too much rep away, which would be harmful to the community in the long run. $\endgroup$
    – AvZ
    Nov 19 '15 at 16:04
  • 1
    $\begingroup$ @AvZ, loosing users with thin skins is harmful to the community in the long run? That's a judgement (incorrect on my view), not a fact. $\endgroup$
    – Alfred Centauri
    Nov 21 '15 at 1:32
  • 1
    $\begingroup$ @AlfredCentauri When trying something new, almost everyone has a little thin skin. It's later when they see it's merits clearly, that they don't get too discouraged. $\endgroup$
    – AvZ
    Nov 21 '15 at 14:54
  • 3
    $\begingroup$ Another query of relevance Users by percent of answers that have negative score. $\endgroup$
    – Emilio Pisanty
    Nov 21 '15 at 20:15
  • 1
    $\begingroup$ The problem with that is that science is not a democracy. You shouldn't judge an answer by the number of upvotes or downvotes. That's letting other people do your thinking for you, and for all you know they're colluding and gaming the system. You should judge the answer for yourself, and follow up on the evidence and references and explanation. $\endgroup$
    – John Duffield
    Nov 22 '15 at 12:07
  • 1
    $\begingroup$ Not a big deal, but @theNamesCross, could you be a little conservative with how often you edit? $\endgroup$
    – David Z
    Nov 26 '15 at 6:14
14
$\begingroup$

This request has a long history going back to the early days of Stack Overflow. You'll notice that question is marked as declined. It was actually for months before being declined and instead the question up-vote value was reduced to 5 (from the original value of 10 just like answers).

I doubt the team will be willing to revisit this decision unless we can show that it represents a real problem (i.e. that a significant number of problem users are gaining access to powerful tools). But I sympathize with your desire.

$\endgroup$
7
  • $\begingroup$ Thanks, Good answer. These posts did not appear in the Physics Meta search. I was under the impression different communities governed themselves (so that Physics.SE could adopt this even if other communities did not)? $\endgroup$
    – OnStrike
    Nov 16 '15 at 23:00
  • 1
    $\begingroup$ Self-governance is one things, changes to the engine that runs Stack Exchange sites are another. I suppose that in principle each site could have its own vote weights, but that is not currently implemented. I really think that step one is figuring out just how much of a problem this is. $\endgroup$
    – dmckee --- ex-moderator kitten
    Nov 16 '15 at 23:04
  • $\begingroup$ SItes govern themselves by consensus building (mostly here on meta), but we can only change the things for which there are knobs. Ordinary users can see a lot of the knobs, and moderators can turn them. Which we're suppose to do according to what y'all tell us. Other things have knobs that can only be turned by the team, and still others would require programming changes. As you go up the scale the burden on us to get things changed increases. $\endgroup$
    – dmckee --- ex-moderator kitten
    Nov 16 '15 at 23:15
  • $\begingroup$ @theNamesCross So in practice, it is not possible for only one site to adopt the change you're proposing here. $\endgroup$
    – David Z
    Nov 17 '15 at 4:03
  • $\begingroup$ @DavidZ Yeah, I understand that now- in that sense, this correctly answers my question. I modified my question to reflect this reality and effectively rephrase the question. Should I accept this as the answer or modify the title to address this? $\endgroup$
    – OnStrike
    Nov 17 '15 at 4:09
  • $\begingroup$ I think it's a useful discussion, and there is no need to single that it's in some sense over if there are still people who want to chip in. And thanks for the editorial suggestion--exactly right. $\endgroup$
    – dmckee --- ex-moderator kitten
    Nov 17 '15 at 4:11
  • $\begingroup$ @theNamesCross As a general rule, it's probably better to accept an answer that addresses your question, and ask a followup question if there is one to be asked. $\endgroup$
    – David Z
    Nov 17 '15 at 4:33
7
$\begingroup$

After about three years participating more or less regularly here, I've seen a number of what one might judge as disreputable users come and go so, on my view, the system is working (in the mean) to discourage such users.

Like dmckee, I sympathize with your concern but honestly, I'm not aware of any regular contributor here that is concerned with deflation of their reputation by the occasional opportunistic or disreputable user.

Indeed, I suspect that the greatest concern is with the quality of questions of late rather than the quality of answers. But that's a different topic.

$\endgroup$
9
  • 14
    $\begingroup$ I'm certainly not concerned with deflation of my own reputation. But I am concerned that the few inflated reputation users can go and give misinformation, the questioners will leave thinking they got info from reputable sources, and those of us who know better might not catch what happened. $\endgroup$
    – user10851
    Nov 16 '15 at 23:13
  • 4
    $\begingroup$ Also, we should be mindful of users who are less addicted to this site than us. Perhaps <3k rep users are disheartened that they have the same rep as people who get as many downvotes as upvotes. I don't know if this is the case, but it's a possibility. $\endgroup$
    – user10851
    Nov 16 '15 at 23:15
  • 3
    $\begingroup$ I think the misinformation issue is the biggie, and I don't know of a solution inside the current framework except voting down the half-assed work and writing better answers. $\endgroup$
    – dmckee --- ex-moderator kitten
    Nov 16 '15 at 23:17
  • 1
    $\begingroup$ @ChrisWhite Your comments echo my larger concern. Speaking for myself, Im not disheartened at their reputation but worried their reputation carries authority that misinforms. Id be concerned if those users gained the ability to close questions etc. $\endgroup$
    – OnStrike
    Nov 16 '15 at 23:23
  • $\begingroup$ @ChrisWhite, for this system, such as it is, I don't believe that there is a way to eliminate the possibility that questioners will leave thinking they got info from reputable sources. In fact, I'm more concerned that efforts to effectively eliminate said possibility will, on the whole, do far more harm than good. I understand that some will disagree but that will not alter my judgement on this one whit. $\endgroup$
    – Alfred Centauri
    Nov 16 '15 at 23:31
  • 3
    $\begingroup$ While not really the topic here, I think "Indeed, I suspect that the greatest concern is with the quality of questions of late rather than the quality of answers" is indeed something more troubling than a few people having a bit of "undeserved" reputation. $\endgroup$
    – ACuriousMind Mod
    Nov 17 '15 at 2:59
  • $\begingroup$ To be fair, I'm not sure reputation is a deciding factor when it comes to misinformation. If John Rennie posted "The world is flat and everything orbits the Earth," I don't know that too many people would look at the reputation and say "Oh, he must be right!" I think that reputation is a factor, but only a factor in deciding between a bad answer and a marginal one, or a good and a great, or between two answers that are different but in subtle ways. I am not sure that there are reputable users posting utter rubbish. But I could also just avoid those topics and not see them... $\endgroup$
    – tpg2114
    Nov 18 '15 at 3:49
  • 3
    $\begingroup$ @tpg2114: you say that, but I've posted answers that I only subsequently realised were rubbish, and by the time I've realised this those answers have attracted upvotes. I suspect that many voters are less discriminating than we might wish. Fortunately we have enough discriminating site members ready to hand out a savaging for sub-standard answers :-) $\endgroup$
    – John Rennie
    Nov 18 '15 at 7:29
  • $\begingroup$ @tpg2114 : the problem comes when John Rennie says the Earth is flat and gets a hundred upvotes, and some guy says it isn't and gets a hundred downvotes. Then some naïve guy like theNamesCross says we should increase the weight of the downvotes to deter such misinformation. IMHO there's an underlying issue here: science is not a democracy. Can you imagine Planck saying to Einstein: "Sorry Albert, I can't print your paper, it's been downvoted". $\endgroup$
    – John Duffield
    Nov 21 '15 at 10:20
4
$\begingroup$

I agree with all the above. However, would a simple rule like the one active on Physics Overflow solve (or, at least, limit) the issue? Namely, I refer to the fact of always showing the total amount of up and down votes, instead of summing them up evening out all the contributions.

$\endgroup$
8
  • 2
    $\begingroup$ Apparently, implementing showing up- and downvotes would be too much of a strain on the database, which is why vote breakdown for individual posts are only visible after 1k rep to limit the amount of users this information needs to be displayed to. $\endgroup$
    – ACuriousMind Mod
    Nov 17 '15 at 14:06
  • 3
    $\begingroup$ Just a note: PhysicsOverflow does both - always show vote counts, and count up and down votes separately. $\endgroup$
    – Abhimanyu Pallavi Sudhir
    Nov 18 '15 at 15:15
  • 1
    $\begingroup$ How would showing both sets of votes to everyone, instead of net votes, alleviate the concerns? Seems to me to be unrelated. $\endgroup$
    – Kyle Kanos
    Nov 21 '15 at 23:10
  • 4
    $\begingroup$ It does alleviate the concerns as, despite the net score and the positive marks, downvotes show (to some extends) that the answer contains some mistakes or some imperfections, making it a pseudo-answer rather than a complete one. Without doing so, the (partially erroneous) answers go on top of the lists anyway and users who don't have the full qualifications to spot the mistake are led to believe that is a fully correct answer (and this happens a lot of times). $\endgroup$
    – gented
    Nov 22 '15 at 18:59
  • 2
    $\begingroup$ Oh for goodness sakes, down votes don't necessarily indicate that the answer contains some mistakes for the same reason that up votes don't necessarily indicate that the answer is perfectly correct. Some will down vote just because others have and some will up vote just because others have. I suspect that implementing this will not significantly address the claimed concern though it might give you some false sense of satisfaction that it has. $\endgroup$
    – Alfred Centauri
    Nov 23 '15 at 2:22
  • 4
    $\begingroup$ Well, if we start from the assumption that up and down votes occur as a consequence of other users having up and downvoted we may as well close the business and leave. I do agree that this happens and it will do unless we implement some sort of correction. $\endgroup$
    – gented
    Nov 23 '15 at 8:45
  • 3
    $\begingroup$ Gennaro, don't be silly; we don't start with that assumption since it isn't the case. Reasonable assumptions are (1) the quality of up and down votes will vary and (2) showing the separate up and down vote totals, while providing additional information to a reader of the answer, will not address (1). If an answer is controversial, the controversy will almost always show up in the comments. If, as you claim, top-rated pseudo-answers are a frequent occurrence here (and you've provided no evidence for that), the mods and regulars here have some work to do. $\endgroup$
    – Alfred Centauri
    Nov 23 '15 at 13:11
  • 1
    $\begingroup$ I just wanted to give my two pennies worth, I'm not claiming that it will solve the problem once and for all (although I still believe that it does address (1)); I am however in line with all the answers and comments above and understand it is more complicated than it seems. $\endgroup$
    – gented
    Nov 23 '15 at 13:46
-6
$\begingroup$

I am newer to the SE community so forgive any naivete.

I'm afraid you are being rather naïve here.

I suspect this is a recurring issue, but I regularly come across users whose reputation continually increases despite many incorrect, or incomplete, pseudo answers.

The biggest recurring issue is homework questions and naïve questions drowning out "the good stuff". Another issue is downvoting, see the various threads including we seem to be over-eager to downvote.

By observation, this type of poster gains reputation by answering a lot of questions poorly.

There are a lot of poor answers out there. But if you think an answer is poor, you could provide a better one yourself. It isn't enough just to downvote, particularly if you don't comment either. You've been here for 18 months and you've only answered 9 questions.

I actively downvote when I feel qualified to do so. While downvotes do discredit these answers, they hardly discredit the poster or deter poor, opportunistic answers.

We really need some examples of such poor opportunistic answers, and then we need to pick them apart. Otherwise there's no meat to your argument.

For example, an [unnamed-user] answered 67 questions in 14 days, earning +27 votes and -25 votes (net +2 votes)- this results in a net +550 reputation.

Let's have a look at some of those answers. They might be wonderful, with references to rock-solid physics, and yet they might have a lot of unexplained downvotes.

This practice exploits the fact that a single upvote offsets 5 downvotes! In my opinion, it undermines the system and deflates the reputation of legitimate users.

Again, without supporting evidence your argument has no weight. For all you know those downvotes are undermining the system.

In other words, (from the perspective of a new user) their answers carry more weight, as reputation should reflect the communities endorsement. I argue that: 1. Their reputation carries authority that misinforms 2. Opportunistic answers perpetuate low quality questions 3. They might gain authority to 'vote to close', etc.

Again, your argument is weightless without evidence. This is a science website, and the science is settled by evidence, not opinion.

I should note that the [unnamed-user] was placed on temporary suspension (presumably independently of this Meta post). In this light, and combined with dissenting opinions of long term users with valued perspective, I am convinced that the current system naturally responds to abuse.

Again, what's your evidence? Do you know what this user was suspended for?

While I think incomplete answers are a detriment to the interests of Physics.SE (the mutual trade of inquiry, dissent, and value, required to build a comprehensive physics knowledge base), I recognize that current mechanisms do filter good questions, answers, and users over time.

Look carefully and you'll perhaps notice that there are expert users who are ex users. The current mechanisms have filtered them over time. Luboš Motl is hardly ever here, nor is Ben Crowell or Carl Brannen.

That said, I'd like to clarify my position regarding the spread of misinformation with respect to reputation and voting. The problem with that is that science is not a democracy. You shouldn't judge an answer by the number of upvotes or downvotes. That's letting other people do your thinking for you, and for all you know they're colluding and gaming the system. You should judge the answer for yourself, and follow up on the evidence and references and explanation In my opinion, this (and similar) sentiments misstate the larger problem. Certainly, science is not a democracy! However, when an OP asks a question (good or bad), they don't know the answer! When an [unnamed-user] provides an incomplete answer (correct or incorrect), they undermine the OP's (future OP's, and the community's) ability to learn and objectively decide if the answer is correct.

There's no undermining if the answer includes references to peer-reviewed papers and hard scientific evidence. The OP is free to follow up those references to objectively decide if the answer is correct. However there is undermining if the answer receives a slew of downvotes. That will discourage the OP from even looking at the answer. See this answer for example. You can follow up on the references to Einstein/Shapiro/etc to check that I'm not talking out of my hat. And yet there's 10 downvotes, with comments from only one downvoter, who has obviously not even read the answer, let alone followed up on the references.

In effect, incomplete answers (incapable of objective scrutiny) place additional weight on the democratic process (via reputation and consensus), unopposed by scientific inquiry.

Of course they do. But here the scientific enquiry is in following up those references and others. Not on simply accepting what somebody else tells you.

We might discourage this practice or expose it by: * Increase downvote reputation losses by some agreed amount? * Introduce a visible statistic (such as Upvotes/Downvotes, Votes/Answer, etc)? * Negate reputation awarded when a question is closed, etc.(to prevent opportunistic answers and reduce incentives to answer homework like questions)? These are large system changes and as dmckee notes, NOT likely to happen.
What changes are possible within Physics.SE?

We'll see. But if you want to call for changes, you really need to support your case with evidence and references.

$\endgroup$
7
  • 1
    $\begingroup$ The whole point: Incomplete answers are answers that dont contain references, list equations without explanation, a single picture, etc.- How should they be evaluated by the OP? $\endgroup$
    – OnStrike
    Nov 24 '15 at 22:50
  • $\begingroup$ @theNamesCross : I'm not clear on your edits. It looks like you've removed "incapable of objective scrutiny" when the references in the answer are a crucial to this discussion. And you still have no examples. As for Whose answer I'd bet on, I'd look at the content and references. JohnRennie has a high reputation and is generally one of the good guys, but if he's saying the Sun goes round the Earth, I'll bet on the other guy. $\endgroup$
    – John Duffield
    Nov 25 '15 at 8:28
  • 5
    $\begingroup$ Playing devil's advocate here, but the planets orbit the barycenter of the solar system, which is located outside the sun's radius, so betting on the guy who says earth goes round the sun is technically wrong. $\endgroup$
    – Kyle Kanos
    Nov 25 '15 at 19:18
  • $\begingroup$ His name was Copernicus. And the Copernican revolution took a hundred years. $\endgroup$
    – John Duffield
    Nov 25 '15 at 20:04
  • $\begingroup$ Hmmmn. No upvotes for making the point about references and evidence I see. Just downvotes, a change to the question, and the deletion of theNamesCross's comment about editing his answer. $\endgroup$
    – John Duffield
    Nov 25 '15 at 20:05
  • $\begingroup$ Because your answer quotes me, I informed you of my edits (thinned redundancy and improved readability) out of courtesy to your answer. Knowing you were aware of my edit, I deleted my comment- but the question didnt change. $\endgroup$
    – OnStrike
    Nov 25 '15 at 20:35
  • $\begingroup$ @theNamesCross : Please don't delete comments. There's no issue if you say "Edit:" and add more material to your question, but changing the substance of your argument multiple times in order to protect it from the points made in an answer, is disingenuous. Since you've also ducked the issue of examples and references, your whole argument is now looking like some Catch-22 diatribe, wherein an answer should be downvoted more because it's been downvoted, regardless of its merits. $\endgroup$
    – John Duffield
    Nov 25 '15 at 21:02

You must log in to answer this question.

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