2
$\begingroup$

I think trouble-with-paper is not a useful tag. I suggest that we remove it (by merging with tag-removed).

update:

Tags for non soft questions are mainly used for subject classification and for keywords (there are some exceptions like reference-request). I think that we should keep the number of these (reference-request like) tags as small as possible. And the burden is on the creator of the tag to justify the usefulness of these tags.

The main use of tags is categorizing question to make it easier to search/ignore/... questions. This tag doesn't express much about the content of the question. It seems very unlikely to me that anyone would use the trouble-with-paper tag. It almost feels like tagging a question with a question tag.

$\endgroup$
  • $\begingroup$ I can think of one reason to search on that tag. When writing a blog post about year two of CSTheory, it might be helpful to say "X, Y and Z person got help understanding papers A B and C." I seem to recall that Peter Shor answered a question about the quantum factoring algorithm, though I don't know whether that question is tagged trouble-with-paper. Maybe not a good enough reason to keep the tag, but I don't think its existence is on a par with a question tag. $\endgroup$ – Aaron Sterling Jan 17 '12 at 3:03
  • $\begingroup$ @Aaron, I think that is a very artificial reason for the usefulness of this tag. I might want to do lockup some questions next year satisfying property X, should I create a new tag for that purpose? Should we create a tag for questions about paper Y? ... Another point here is it is not just one person that uses a tag, it is a system that should be helpful generally. It is a pragmatic point, not a theoretical one. We should avoid creating new tags as much as possible and try to use the existing ones to categorize questions, otherwise the system will become less and less useful. $\endgroup$ – Kaveh Jan 17 '12 at 3:23
  • 1
    $\begingroup$ We should retain the tag and be more open towards questions that warrant it. $\endgroup$ – Raphael Jan 17 '12 at 6:46
  • 1
    $\begingroup$ Since some seem to misunderstand this post, this is not about the scope, it is about having or not having the tag. These are separate issues. Questions about published papers are already in the scope and no one is suggesting a change about that. $\endgroup$ – Kaveh Jan 22 '12 at 23:39
4
$\begingroup$

I do not find anything wrong with the tag itself. Why is it bad? I think that it describes the content of the question.

Also I have trouble understanding what “meta tag” really means. It seems to me that people use this term whenever they do not like some tag.

$\endgroup$
  • $\begingroup$ I updated the post and removed the word meta-tag (I agree that it is a quite ambiguous concept meaning different things to different people but if you are interested see these: 0, 1, 2, 3). $\endgroup$ – Kaveh Jan 17 '12 at 0:45
  • $\begingroup$ @Kaveh: As I said, I think that this tag describes the content of the question and therefore it is different from tagging a question as “question.” I do not see why all tags have to be used for subject classification. $\endgroup$ – Tsuyoshi Ito Jan 17 '12 at 1:11
  • 1
    $\begingroup$ @Kaveh: By the way, I had read your link 0 before I posted this answer, and I found it ironical to see “If the tag commonly means different things to different people, it’s probably a meta-tag” because, like you, I think that the same applies to the term “meta-tag.” $\endgroup$ – Tsuyoshi Ito Jan 17 '12 at 1:17
  • $\begingroup$ I don't think it describes the content of the question. Being about a paper is not content (same with homework, big-list, ...). Unlike subject classification (ds.algorithms) and keywords (dynamic-programming). Since subject classification and keywords are the main ways of categorizing articles I think using them in the search is quite natural and therefore they are useful. On the other hand I found it very unlikely that someone will search for trouble-with-paper. $\endgroup$ – Kaveh Jan 17 '12 at 1:29
  • 1
    $\begingroup$ @Kaveh: I do not know why you think it is unlikely that someone will search for trouble-with-paper. $\endgroup$ – Tsuyoshi Ito Jan 17 '12 at 1:53
  • $\begingroup$ why would someone search for it? (AFAIU, it means the OP has trouble in understanding some paper) $\endgroup$ – Kaveh Jan 17 '12 at 2:00
  • 2
    $\begingroup$ @Kaveh: (1) If I want to ask a question about understanding a specific paper, I will search similar questions to decide what is a good way to write the question like that. (2) More importantly, I think that you are assuming that you know how other people use this website, and that attitude worries me. $\endgroup$ – Tsuyoshi Ito Jan 17 '12 at 5:54
  • $\begingroup$ Tsuyoshi, honestly I think this is not a good reason. Let me explain: it is for reasons not related to reading the question or its answers. See my reply to Aaron. I think the difference we have here is that I am looking at the tags as a system and want to see what is a reasonable tag and what is not, e.g. is it OK if someone uses a tag to categorize questions referring to paper X? Is it OK if someone creates a tag for reasons not related to the content of the question? (e.g. the reason you stated) On the other hand it seems to me that you are looking at it as a single tag issue. $\endgroup$ – Kaveh Jan 17 '12 at 23:11
  • $\begingroup$ About 2, I have ideas about how people are using the system but obviously it is not complete, it is an approximation. I didn't see any use for the tag, and now that you and Aaron have come up with reasons I see there can be useful, but the point is general usefulness of tags. Maybe we should start a more general discussion about what is the goal of having tags and what kind of tags are acceptable. SE's general policy (as I understand it) is that these kind of tags are not acceptable (with small number of exceptions like homework tag on Mathematics). $\endgroup$ – Kaveh Jan 17 '12 at 23:19
  • $\begingroup$ But they haven't enforced it on cstheory, so we have more freedom. I don't know what is the policy on MO (will try to check it) but I also haven't seen tags like this there. It would be nice if we had statistics about how often these tags are being used for searching/following/filtering but we don't. As I expressed in the post my view is that for non-soft questions the tags act mainly as subject classification and keywords in articles on AMS MR, ACM DL, arXiv. $\endgroup$ – Kaveh Jan 17 '12 at 23:24
  • $\begingroup$ If you have an alternative general perspective about tagging system and what is an acceptable tag and what is not an acceptable tag I would be happy to listen and learn. We would like to have some general guidelines about tags to keep them useful. If someone creates a tag named X and uses it to tag question satisfying property X and argues that it is useful for reasons similar to your and Aaron's examples, would that be acceptable? (sorry for long reply). $\endgroup$ – Kaveh Jan 17 '12 at 23:27
  • 1
    $\begingroup$ @Kaveh: You can keep saying whatever use cases I come up with is not up to your standard. You have your own conclusion and you do not want to change it because of input from others. You do not have to learn. I lost interest in convincing you. $\endgroup$ – Tsuyoshi Ito Jan 17 '12 at 23:49
  • 3
    $\begingroup$ Kaveh, Tsuyoshi, maybe you should continue on chat, or maybe even take a break for a bit ? $\endgroup$ – Suresh Venkat Jan 18 '12 at 0:17
  • 1
    $\begingroup$ @Kaveh: First of all, “give me a use case or I will delete it” is nothing but unfair. Then Aaron and I explained some use cases when no one really have to. Then you just claimed they are not good reasons because they are not what you think are good. Come on. There seems to be nothing I can do about it. Go ahead and delete the tag if that is all you want to do. $\endgroup$ – Tsuyoshi Ito Jan 18 '12 at 0:19
  • 1
    $\begingroup$ see specific questions about thm x in paper y are ok $\endgroup$ – vzn Jan 22 '12 at 17:01

You must log in to answer this question.

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