3
$\begingroup$

Google has a habit of displaying cards, and I am sure several of you have seen these by now:

enter image description here

(search link)

The way I understand the purpose - of displaying cards for each answer (and not just the accepted answer) - is that it helps the viewer to get alternative ideas about the question, and that is a great thing, when all of the alternative views are correct.

However, it is not so great in such cases:

enter image description here

(search link)

Notice here that answer 3 and 4 are displayed just as prominently as answer 2. But, if you'll visit the actual question (archive link), you'll find that answer 3 and 4 are actually greyed out, with a score of +0/-4 and +0/-5 respectively. Answer 2 instead has a decent score of +2/-1. Realize that this has a potential to sway innocent Google searchers unless they open the actual SE page.

Now, we know that the main reason Google is pulling out these cards from SE sites is to retain the viewer on its search page. Well, assuming they're successful in preventing even 10% of their viewers from clicking the SE link (a reasonable success rate given Google is a smart company), and assuming a few of them choose to prefer answers 3 and 4 over answer 1 (probably because they support some extreme/obsolete view), that's still a lot of misguided folks, especially in case of the very popular questions.

Since we can't force Google to change their algorithm[a], we can at least take steps to minimize the damage by taking our own little steps. I wish to propose that any answer:

  • that is sufficiently low quality and does not attempt to answer the given question, or is completely inaccurate, or plagiarized from an existing answer
  • where the author has not responded to comments seeking clarification, or has replied with inaccurate comments
  • has been over thirty days since posted

should be flagged as VLQ/NAA so it can be quickly deleted.

So, should we flag such answers as NAA/VLQ so they can be quickly deleted? Is it ok, or are there problems with this approach?
I myself don't see any harm with this approach, but I might be missing out on some fine detail.


Meta-background: Nothing much, but that I myself used to downvote a lot of answers in the past, even those which were already greyed out. But I never flagged them.[b] And, I did not realize that they would be shown in Google searches nonetheless. I agree their being shown in the answer cards is a trivial issue, there may not be more than fifty such examples at the present time, but I still thought it might be worth bringing this idea to the attention to the rest of the community. It just gives an incentive for flagging and getting rid of such answers. I found one example above, but there might be many more such questions.

Welcome all constructive feedback! Thanks! :)


[a]: Shadow Wizard, who's literally all over MSE, had this to say in regards to this issue: "SE staff mentioned in the past they don't have any part in those experiments Google are doing, and Google did not ask for any cooperation from SE side"

[b]: I do not definitely exclude the possibility that I am the only with this habit. But considering I have had the chance to view several, year-old, greyed out answers, which should've been deleted by then, I think I'm definitely not the only one.


Actually, this is no longer an issue. Currently, Google has reverted to the previous search display as it used to be. For example, the first search link mentioned in the post above gives:

enter image description here

...as it used to displayed earlier.

If a similar issue ever comes up again, due to Google or otherwise, we can always have a look back at this discussion.

$\endgroup$
  • 1
    $\begingroup$ In case anyone is unaware how far Google has gone in bringing these cards (any many, many other related cards) into our lives, here's a nice in-depth article about the same. $\endgroup$ – Gaurang Tandon Mar 20 '18 at 6:48
7
$\begingroup$
  • Sufficiently low-quality answers, non-answers, and plagiarised answers are already flagged, and usually consequently deleted (probably actually hidden from public view). This should continue, in my humble opinion.
  • There is some merit in preserving a little wrong, downvoted answers precisely because it shows that these alternative approaches have been deemed incorrect by the community. That is, it stands as warning for future readers. Even more so if there is an extensive comment section. An exception would be a case in which the main answer(s) have discussed these misconceptions at length, I suppose, but this should be decided on a case-by-case basis.
  • In the ideal case, wrong answers are corrected by OP. If this edit is done a large amount of time after initally posting, or if I feel it is a large part of the answer, I like to add a disclaimer after the edit detailing why and how the original was flawed. (I know you specifically mentioned that OP's non-responsiveness is one of the criteria but the emphasis should stand.)
  • A final scenario is when the OP is responsive but disagrees that or does not understand why their answer is wrong after it has been pointed out. A forceful edit is against SE policy (as it should be), so in this case the stock action is also further persuasion, comments, and downvoting. We could also encourage the answerer to ask a new question.

It goes without saying but if the post is corrected, and you happen upon it, removing the downvote is good policy (I think) to catalyse editing wrong posts by the OP.

Sidenote: In which browser are these boxes showing up?

$\endgroup$
  • $\begingroup$ "There is some merit in preserving a little wrong, downvoted answers precisely" But the problem with these is that Google search doesn't know if these answers are wrong. A better course of action in such cases would be, in my humble opinion, (1) for the answerer to put a line in the start of their answer "This answer has been determined wrong by the community but stands to clear misconception" (2) delete that answer and let the accepted answer clear the misconception. (I think many great answers here not only state the answer but also clear related misconceptions.) (cont) $\endgroup$ – Gaurang Tandon Mar 15 '18 at 1:39
  • $\begingroup$ "extensive comment section" sometimes what happens is that the comment section isn't 2-3 comments of mistakes but a dozen comments of back and forth dialogue, and the latter is difficult for any reader who wasn't a part of the original convo to understand. "I like to add a disclaimer after the edit detailing why and how the original was flawed" Definitely, 100% agreed. "This should continue, in my humble opinion." Yup, sure. " A forceful edit is against SE policy " Yup, we should never forcefully edit in such a case. This I think is one exception where downvoting is our only option. (cont) $\endgroup$ – Gaurang Tandon Mar 15 '18 at 1:43
  • $\begingroup$ And thanks for your answer btw! :D "In which browser are these boxes showing up?" I didn't check other browsers, sorry for that, but I saw cards/boxes these in Chrome. $\endgroup$ – Gaurang Tandon Mar 15 '18 at 1:44
6
$\begingroup$

I would like to add one point which I find very important concerning:

sufficiently low quality and does not attempt to answer the given question, or is completely inaccurate, or plagiarized from an existing answer

If an answer is plagiarised, it should not be flagged as VLQ or NAA, but instead with a custom flag telling the moderators exactly that, preferably with a source. Our team usually then knows if this is a one time case, or has occurred multiple times, and then we would us the appropriate action.
(This of course also applies to plagiarism from outside of our site.)

$\endgroup$

You must log in to answer this question.

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