8
$\begingroup$

The consensus seems to be that if people find errors in papers, they should contact the author directly, not post a question here.

Unfortunately, it seems that we haven't enforced this policy in this thread. There is an answer whose sole purpose seems to be exactly this – to point out a possible mistake in a specific paper. This has already led to follow-up discussion that isn't really appropriate on this site.

I would suggest that the moderators delete both of these answers, and encourage the participants to discuss the possible errors offline.

$\endgroup$
  • 2
    $\begingroup$ I have locked the question for the time being till we decide what to do about this. $\endgroup$ – Kaveh Apr 18 '11 at 7:02
  • 1
    $\begingroup$ I closed the question (based on the answers and votes so far, there seems to be an agreement about that point). $\endgroup$ – Kaveh Apr 21 '11 at 22:27
6
$\begingroup$

Another option is to close the question.

$\endgroup$
  • 3
    $\begingroup$ But even if the question was closed, the answers would be publicly visible. Hence regardless of whether the question is closed, I would suggest that we delete (at least) the two answers that I mentioned. $\endgroup$ – Jukka Suomela Apr 18 '11 at 6:53
  • 2
    $\begingroup$ I agree with Jukka, we should remove the new answers and tell the two users to continue their discussion off-line. I think we should have closed this question long ago, and I thought about this again when trg787 posted his first answer. $\endgroup$ – Kaveh Apr 18 '11 at 7:07
  • $\begingroup$ For what it is worth, I agree with both Kaveh and Jukka. $\endgroup$ – Joe Fitzsimons Apr 18 '11 at 10:48
  • 4
    $\begingroup$ In any case, the nonsense conversation part should be deleted. $\endgroup$ – Dave Clarke Apr 18 '11 at 19:23
4
$\begingroup$

I do not think that deleting the answers is the right action to take, because in my opinion, the question asked for them. The question itself was inappropriate in the first place. So if the answers should not be on the website, the question should be deleted.

$\endgroup$
  • 1
    $\begingroup$ I don't see any reason to delete the question; closing the question should be enough, and we can do that through the usual process (vote-to-close). $\endgroup$ – Jukka Suomela Apr 18 '11 at 6:50
  • 1
    $\begingroup$ Let's assume that this was not an old question but a new one. Then we would close it in place of deleting it (and after closure it can be deleted by a mod or by votes of high-rep users). But the question is old and there is an old nice highly-up-voted answer by David, I would personally prefer to keep the answer and don't think we need to delete the question and the answer for resolving the issue. (Of course high rep users can vote to delete the question after we revolve the issue we have at hand.) $\endgroup$ – Kaveh Apr 18 '11 at 7:14
  • $\begingroup$ Jukka and Kaveh, you convinced me that the question should not be deleted but it should be just closed. $\endgroup$ – Tsuyoshi Ito Apr 18 '11 at 11:48
  • 1
    $\begingroup$ However, let me clarify my feeling. David’s post indirectly answers the question, whereas trg787’s post directly answers the question, and I think that it is somewhat unfair if trg787’s answer is deleted and David’s answer is not. $\endgroup$ – Tsuyoshi Ito Apr 18 '11 at 11:53
  • $\begingroup$ @Tsuyoshi Ito, personally I think trg787's answer is not an answer to Sadeq's question for the following reason: the question asks about correctness of the algorithm and the paper, trg787's answer seems to be a bug in an implementation that has been fixed in the later version, it only shows that that the previous implementation fails on an instance, it doesn't say anything about the algorithm or the paper. $\endgroup$ – Kaveh Apr 19 '11 at 5:03
  • 1
    $\begingroup$ @Kaveh: I am confused. Does that justify removing trg787’s answer and keeping David’s answer? I thought that your reasoning was that the answers should be deleted because the question should have been closed in the first place, not because trg787’s post is not a good answer. $\endgroup$ – Tsuyoshi Ito Apr 19 '11 at 8:41
  • $\begingroup$ @Tsuyoshi, no. I was just explaining my own feeling since you mentioned yours in the previous comment. :) $\endgroup$ – Kaveh Apr 19 '11 at 22:58
  • $\begingroup$ @Kaveh: I see. But in another comment, you state that you would prefer to remove the new answers, leaving David's answer. I fail to see why it would be good. It sounds like an arbitrary decision. $\endgroup$ – Tsuyoshi Ito Apr 19 '11 at 23:00
  • 1
    $\begingroup$ @Tsuyoshi, David's answer is posted before the policy, and it hadn't caused a problem. The problem is that we should have closed the question after the adoption of the policy. Removing the new answers seem most reasonable choice to me since leaving them and not allowing further discussion can be misleading for people reading the answers (who probably will not read the comments or this meta discussion). I fear that if we remove the lock without removing the new answers it will cause further discussion similar to the current ones which we want to avoid. $\endgroup$ – Kaveh Apr 20 '11 at 5:21
  • $\begingroup$ Removing the new answers will solve the problem IMHO and any further action does not seem necessary to me at the moment. (and yes, I agree, it seem a little bit arbitrary. If you have a better option I am listening. :) $\endgroup$ – Kaveh Apr 20 '11 at 5:23
  • $\begingroup$ @Kaveh: In any case, moderators should close the question, leave a comment discouraging discussions and unlock the question. I think that better options than deleting the two answers are: (1) deleting all the answers, or (2) deleting nothing. In the case of (2), if the situation becomes bad again, moderators will have to handle it by possibly deleting posts or locking the question. $\endgroup$ – Tsuyoshi Ito Apr 20 '11 at 10:56
  • $\begingroup$ @Tsuyoshi, unfortunately locking the questions doesn't seem to prevent commenting. I don't like removing David's answer since that was before the policy came into effect and it has not caused a problem, and as I said leaving the answers will be misleading for people reading the question. Dave's answer is the same as your later suggestion, and I think you should post your former suggestion as an answer if you really think it is the right thing to do. $\endgroup$ – Kaveh Apr 21 '11 at 22:33
  • $\begingroup$ @Kaveh: If you think that I suggested what I do not believe is the right thing, that’s fine. $\endgroup$ – Tsuyoshi Ito Apr 22 '11 at 1:30
  • $\begingroup$ @Tsuyoshi, sorry, I thought that you are listing other possibilities. I understand your concerns about fairness. In any case I think we should close this discussion and a make a decision. I suggest we add a note to the post linking to the discussion here without deleting any answers, and unlocking the post, and remove the answers if we see that the answers continue to generate off-topic discussion. $\endgroup$ – Kaveh Apr 22 '11 at 23:04
3
$\begingroup$

As the question has been around for a while (since September) without complaint, I'd be more inclined to leave it there as is (statue of limitations).

But the lock Kaveh installed should stay, to avoid any further non-scientific comments.

$\endgroup$
  • 1
    $\begingroup$ The question has been around since September, but the two answers that I mentioned are new (this month). $\endgroup$ – Jukka Suomela Apr 18 '11 at 18:45
  • 1
    $\begingroup$ Ah. I see. The answer brought it back in the spotlight. $\endgroup$ – Dave Clarke Apr 18 '11 at 19:22
  • 2
    $\begingroup$ I do not think that it is good to keep it locked. At least Kaveh clearly explained that he locked the question as a temporary measure, so if it is kept locked, there should be an explanation for that. $\endgroup$ – Tsuyoshi Ito Apr 18 '11 at 23:26
  • $\begingroup$ As I said I would prefer if we remove the new answers, closed the question, and then unlock it. If we see that it continues to cause further discussions then we can consider locking it again. $\endgroup$ – Kaveh Apr 19 '11 at 5:07
3
$\begingroup$

To expand Jukka's suggestion (i.e., deleting the troublesome answers) into a separate answer:

  1. Close the question: it is out of scope, even if it wasn't when it was posted. There is no benefit to the site to the kind of discussion that further answers would constitute; and
  2. Delete the answers: it injures the site to flout the site's policy on what constitutes acceptable discourse in this way. There is not a real censorship issue, since the content of at least one of the two answers is hosted elsewhere.
$\endgroup$
2
$\begingroup$

I think the question is appropriate. It concerns a certain famous problem that is certainly within our scope. The paper in question appears to be neither accepted not refuted (apparently mainly because of lacking form), so I think it is valid to ask for the current state. It is certainly a question one TCS researcher might ask another, but likely not a student.

$\endgroup$
  • 7
    $\begingroup$ It is clearly against site's policy. Discussing the general correctness of papers which have not passed the peer-review process are off-topic for this site. The only reason it didn't got closed was that it was before the adoption of the policy. $\endgroup$ – Kaveh Apr 18 '11 at 7:10
  • 2
    $\begingroup$ ps: the current state is as you said: "neither accepted not refuted". $\endgroup$ – Kaveh Apr 18 '11 at 7:42
  • 2
    $\begingroup$ I agree that the question is against policy. Not all plausible questions can be entertained here, especially open-ended ones that require discussion. $\endgroup$ – Suresh Venkat Apr 18 '11 at 15:59
  • 1
    $\begingroup$ @trg787: Although I understand your frustration, it is not wise to use impolite words. $\endgroup$ – Tsuyoshi Ito Apr 21 '11 at 11:08
  • 1
    $\begingroup$ @Tsuyoshi, I'm not frustrated, at all. My main trouble is that I was born too honest and too frank for the cruel world. $\endgroup$ – trg787 Apr 21 '11 at 11:20
  • 1
    $\begingroup$ @trg787: The “although” part in my previous comment is not my main point. $\endgroup$ – Tsuyoshi Ito Apr 21 '11 at 11:31
  • 1
    $\begingroup$ @trg787: If you want, you probably should express your feeling to those who really helped. I did nothing, and I do not even live in Japan. But let’s keep the meta discussion on topic. $\endgroup$ – Tsuyoshi Ito Apr 21 '11 at 11:34
  • 1
    $\begingroup$ It's Ok, brother :) But you are a particle of your people. It's enough to deserve the gratefullness. $\endgroup$ – trg787 Apr 21 '11 at 11:34
  • 1
    $\begingroup$ trg787, this is a professional site. Please remain nice and professional. (please read the site's FAQ if you haven't.) $\endgroup$ – Kaveh Apr 21 '11 at 22:19
  • 2
    $\begingroup$ Is there a point to your rambling about Russian history, especially in the given context? $\endgroup$ – Raphael Apr 22 '11 at 19:08
  • 3
    $\begingroup$ trg787, several of your comments have been flagged by users. I am removing all of your comments which are unprofessional or not relevant to the topic (and will remove feature ones). This is a TCS Q&A site, it is not a general discussion forum, please keep your comments on topic. $\endgroup$ – Kaveh Apr 22 '11 at 22:56
  • $\begingroup$ It's OK, @Kaveh. You did right thing. My apologies to everyone. $\endgroup$ – trg787 Apr 22 '11 at 23:08
  • $\begingroup$ trg787, thanks for understanding. :) $\endgroup$ – Kaveh Apr 22 '11 at 23:11
1
$\begingroup$

I closed the questions, unlocked it, and removed some comments.

I suggest not removing the answers (unless they continue to cause off-topic discussion), and add a note to the questions explaining the situation ("The paper is neither accepted nor refuted. It is off-topic to discuss general correctness of such papers on cstheory.") linking to this meta post.

$\endgroup$
  • 1
    $\begingroup$ kinder garten :) Guys, seems you live too prosper lives. Have you any other problems in your lives? $\endgroup$ – trg787 Apr 23 '11 at 20:30

You must log in to answer this question.

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