Similar questions the same site. Sometimes it happens that I've a problem (when I'm programming in Fortran, or when I'm writing a LaTeX document, ...) that I cannot solve on my own, so I open Google and search for an answer; I enter complete sentences if it's evident to me that there surely exist an already asked-and-answered question, otherwise I just write what I think are keywords to my problem. In the latter case the search process is harder and sometimes it happens that I don't find anything useful, so I decide to post a question. Such a question can result to be strictly related to (even if not really a duplicate of) another question. This question could be merged to the existing one in order to add some detail and make it more visible on the web.

Same questions on different sites. The questions about cross-posting are so many, and they make clear that it is not evil tout court. In general, an assiduous TeX.SE user can be so able to use Vim that he could easly answer a Vi.SE question even if it's barely related to LaTeX. But maybe this user considers vi just a tool, so he can have no vi account and, in turn, cannot see the question. But this question has tags, maybe a LaTeX tag. This tag could be used to make the vi.SE question visible on TeX.SE. To make a real example, this question of mine has been answered on TeX.SE, but maybe it would have fit better to Vi.SE, wouldn't it?

  • Wouldn't it be useful to make SE sites more related to each other than they are now? I mean, a tag can be used to link different sites.
  • Wouldn't it be useful to develop a graphical platform to navigate from one site to the other in order to find an answer? I'm thinking about something like a graph with branches connecting node. I think SO would be a big node connecting many other (smaller) nodes.
  • About how to do something like this, it comes to my mind a program which can be used to automatically build a dependency-map of a program (made up of sub-program and modules) exploiting the source codes.
  • 3
    Because tags on different sites may not be used in exactly the same way? – Catija Jul 15 '16 at 22:42
  • Well, I bet that on Sexy.SE Latex could mean something other than LaTeX. – Enrico Maria De Angelis Jul 15 '16 at 23:05
  • 5
    Don't even need to be hypothetical, there is a latex tag on Arts & Crafts. – Catija Jul 15 '16 at 23:07
  • In this case it's not that they aren't used in exactly the same way. They are exactly two unrelated things. I think these situations can be easily managed. – Enrico Maria De Angelis Jul 15 '16 at 23:18
  • 1
    It makes it significantly more work. Even within single sites, it's possible to have issues like this. M&TV has to be careful about TV show titles that happen to be common genre words... Cooking has to deal with words that have different meanings between American English and British English (and even Australian and Indian English)... like "pudding" or "chips". And what about when there's a tangential relationship but not an actual one... perhaps with the "black holes" tag on Worldbuilding and Physics. Do we really want those two connected? – Catija Jul 15 '16 at 23:35
  • No, we don't. I'm not talking about merging sites without criterion, nor about collecting ALL tags together. But something like.. making bridges which connect specific tags. So vi.SE's and SO's latex tag would be connected to TeX.SE, but not to A&C.SE. – Enrico Maria De Angelis Jul 15 '16 at 23:49
  • Related: meta.stackexchange.com/questions/268885/… – rene Jul 16 '16 at 7:57

Tags are not a static taxonomy that has been organized between sites. Tags are largely ad hoc keywords selected by the person asking the question and loosely organized by the community charged with their management. They are constantly being created and destroyed and edited through the choices made by the individuals that use them.

So rhetorically, how would such a collaboration even be established? Communities don't really have a well-established means of communication between them. Even that rare, unshakable tag match between sites would would have at least as many that are marginal at best. The subject being asked about under these tags differ between sites, so the use of [tag] would likely vary substantially between them. And where these marginal associations are established, the community is no longer free to re-organize their content. A large percentage of these tenuous relationships would likely go out of date quickly, or be broken completely as the sites evolve.

The story here is that there are a few tags that are essentially the same between sites. Maybe on the rarest of occasion, someone might find a better answer in a tag on another site. But for the vast majority of these use cases, this meta data would be incorrect or unreliable at best. I don't think tags are that categorize-able, especially between sites. We would needs a system to determine which tag-sets are eligible and who can create and destroy the pairs. It would be an administrative nightmare.

It would not be a well-trusted system.

You must log in to answer this question.

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