4

This question has been recently asked.

Should questions about identifying languages be on-topic here?

Pro's

  • Such Q's are often interesting by themselves;
  • They may highlight some unusual usage of common languages or provide an interesting info about extinct or rare ones;
  • They may let the users to demonstrate their erudition (may be an important concern for someone);
  • At the moment, Linguistics.SE needs more users and more questions, so even if such questions are not {yet} on-topic, they may be included;

Con's

  • There may be disputes on which language it is, provided that several relative languages or dialects may apply;
  • They are not about linguistics! :)
  • 1
    Thanks for the question. I'll wait to see more community participation and then when there's enough consensus, I'll take the majority vote. :) I added the featured tag. It should bring more attention to it. – Alenanno Jan 11 '13 at 10:17
  • I'm voting to close this question as off-topic because it is now out of date. – curiousdannii May 15 '18 at 16:10
  • @curiousdannii, how does that help? Would it not be more helpful to answer it instead, referencing current policy? – Peter Taylor May 15 '18 at 17:58
4

Note: in the years since this was written, the community has since decided that these questions are no longer allowed.


As you say, there are several good reasons to allow them, and that is enough for me. I vote yes.

1

Few considerations:

  • Methods of automated language identification given a text belong to linguistics. So language identification per se is not off-topic, but just this particular type of language identification as it includes listening to a recorded speech sample.

  • Sometimes identified language family or even a geographic area could be enough for the asker

Also it is not clear to me how language identification should be performed in principle. A web site where speakers of all possible languages listen to all samples provided by askers doesn't seem feasible.

  • RE: automated identification, I would argue that Markov's chains and Vector-space modeling rather belong to Theoretical Comp Science or Computational Science; there are many nice NLP questions there. – bytebuster Jan 10 '13 at 9:38
  • RE: how identification should be done. Again, it's a statistical analysis combined with a probabilistic Vector-space modeling, like on this site. – bytebuster Jan 10 '13 at 9:41
0

More questions is ok with me!

And if there are some disputes about exactly which dialect/language it is then I look forward to some well argued linguistic analysis.

You must log in to answer this question.

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