1
$\begingroup$

Should any tag for a distribution include that -distribution, or only when necessary for disambiguation?

For example, the following distribution tags do not include that suffix:

But the following do:

| |
$\endgroup$
  • 1
    $\begingroup$ -1 to indicate disagreement on the basis of @gung's argument, but I've nothing against the question's clarity, research effort, usefulness, stylishness, etc. On those bases it'd be a +1 any day. $\endgroup$ – Nick Stauner May 3 '14 at 22:31
  • $\begingroup$ Are you saying you down-voted the question based on disagreement? What about the question is disagreeable? $\endgroup$ – feetwet May 4 '14 at 12:56
  • $\begingroup$ Yep. See gung's answer... $\endgroup$ – Nick Stauner May 4 '14 at 20:12
  • $\begingroup$ Obviously I'm relatively new here, but this seems really weird: It's a question for discussion. I can see disagreeing with an answer, but disagreeing with a well-formed question because you agreed with an answer to the question? Your background suggests some grounding in logic and philosophy: perhaps you could describe the premises or reasoning under which such an action makes sense? $\endgroup$ – feetwet May 4 '14 at 23:50
  • 3
    $\begingroup$ I understand your question as containing a proposal as a premise for discussion. I don't see a need for the proposal, and I've upvoted answers to this effect. Downvotes on meta-sites (excluding MSE, the big one) are often used differently. In case it's relevant and you weren't aware, they have no ramifications for reputation here (except badges). Pardon any confusion; it's a bizarre policy that had to be explained to me once personally...I've upvoted some questions on MSE that question the policy too, FWIW. Though I've tried to follow it here, I'm not entirely sure it's a good system... $\endgroup$ – Nick Stauner May 5 '14 at 0:12
  • 2
    $\begingroup$ @feetwet voting works differently in meta to the main site. Here voting indicates agreement/disagreement with a position, not with the quality of the post. It's not personal in any way. $\endgroup$ – Glen_b May 5 '14 at 0:17
  • 1
    $\begingroup$ Personally I'd prefer if there were the usual up and down arrows (to reward post quality, on which badges were based) and a second autotallied +/- which indicated agreement/disagreement (when relevant) and which always gave both + and - numbers (+4 means something different on a 54/48 split than on a 4/0 split) $\endgroup$ – Glen_b May 5 '14 at 0:23
  • $\begingroup$ Thanks, I did not know that. Does Meta have its own intro hidden somewhere? It sounds like there should be a "Proposal" tag in the mandatory Meta tag set. E.g., in this case I really didn't have an idea one way or the other, and the title doesn't even incorporate the key nuance in the discussion. So even re-reading the full question it's not clear what position a voter is taking. If I had a Proposal it would have had a decisive title like, Resolved: Distribution tags should only include "-distribution" when necessary for disambiguation. Now that's a "question" I could see voting on! $\endgroup$ – feetwet May 5 '14 at 2:03
6
$\begingroup$

I have a preference for shorter tag names when possible. That is why I created a tag for , and not [rayleigh-distribution] (but I didn't really think about it).

When there are enough questions categorized under a tag, if there is a demonstrated need, we can create additional tags that will further differentiate a category. As an example, I split up when it seemed to become problematically heterogeneous (in my mind) and created new tags for , , , . I do believe that having the right tags, and having threads tagged appropriately, is very important to the organization and usability of the site. However, I think think that this will always be a work in progress, requiring some continued maintenance as the site evolves; I don't think it is very possible or useful to try to predict what tags / differentiations will be needed in advance. For instance, you make a good point about the ambiguity of , but I haven't noticed an actual problem. If someone else has, or if one emerges, we could split the tag up. Until then, I'm inclined to let it be.

| |
$\endgroup$
  • $\begingroup$ Words of wisdom, mother Mary. +1 for sneaking in Beatles references (and the other merits of your answer, of course)! $\endgroup$ – Nick Stauner May 3 '14 at 22:27
  • $\begingroup$ (+1) It's a pity that we still have [fisher] tag; it currently has 75 questions. All of these 75 threads should really be retagged and the tag eliminated. Do you remember if back then you removed the tag completely (and it re-appeared afterwards) or did you keep it on some threads? $\endgroup$ – amoeba Nov 2 '17 at 22:06
  • $\begingroup$ I didn't remove all of them, @amoeba. Consider Fisher Distance for feature selection, it doesn't fit into any of the categories above, it is from Oct 31, '12, & it has never been edited (to add or remove the tag, or for anything else for that matter). The idiosyncratic ones that didn't fit into the categories I came up with, I just left as is. $\endgroup$ – gung - Reinstate Monica Nov 3 '17 at 0:07
  • $\begingroup$ I see. Well, currently I would say that if there are idiosyncratic uses of [fisher] tag then it's better to remove any Fisher-related tags from them. As long as [fisher] tag exists, people are going to use it for all sorts of things... Do you think this is worth a separate Meta discussion? $\endgroup$ – amoeba Nov 3 '17 at 7:32
  • 1
    $\begingroup$ Nah, we can just start going through the threads & removing the tag, @amoeba. $\endgroup$ – gung - Reinstate Monica Nov 3 '17 at 12:14
  • $\begingroup$ OK. I might post something though, perhaps somebody else would join the effort. $\endgroup$ – amoeba Nov 3 '17 at 20:58
4
$\begingroup$

Given that we also have a tag for , users can always amend a generic tag like in order to avoid ambiguities. For instance, a question concerning the chi-squared test can equally be used together with to show that the question is not about the distribution. For this kind of tag I think it's beneficial to not have an additional one for the distribution itself. Otherwise we might unnecessarily inflate the number of tags even though existing tags can be used in combination for the same purpose. After all this flexibility is implied by allowing the use of multiple tags per question.

| |
$\endgroup$

You must log in to answer this question.

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