This site keep receiving questions asking for help to identify Chinese characters. Example:

I think these questions are too specific to be useful to anyone learning Chinese on this site. And having too many of such questions will reduce answerers to mere CAPTCHA solvers. It would be good if we can have an off-topic close option specifically for character identification as this is not the same as asking for a translation.

A link to Google translate and an instruction to teach users how to use the handwriting input option or to count strokes and lookup a character by its radical can be useful to reduce the quantity of such questions which adds little to no value for the site.

Just to clarify, I am not proposing to close all such questions as the decision still lies within the community. There will always be some exceptions such as in the case of seal script.

1 Answer 1


I don't get it, what about the person's prior effort? Every person who learned Mandarin was taught character recognition by someone. I do understand the need not to trivialize the website but I do not believe that all character recognition questions should be grouped together as "bad". When effort is shown, I believe character recognition becomes a valid question. If I know 4 out of 5 characters but not the 5th, if I come to the site and ask a character recognition question in this case, I think it ought to be registered as a valid question. No amount of technology would replace a human teacher. Character recognition is taught in school everyday and is definitely part of any Q&A on a Chinese language site.

  • The decision to close a question or not still lies in the hands of the moderating community. Whether effort is shown or not is a very subjective issue which I do not wish to discuss here. What I am proposing, is for a better close reason for questions that fit into the category of character identification because it is not the same as translation. My objective is to reduce the amount of such questions because it adds little to no value for the community. Also, I don't see how regular script 楷书 is so difficult to decipher if you can learn to look up by strokes or its radical with a dictionary.
    – 杨以轩
    Aug 29, 2014 at 10:47
  • 1
    @QuestionOverflow - The value added is subjective. For example, I placed a conversational question in context, with 12 missing characters out of 60. The remaining characters needed only identification but the conversation can be priceless to new learners (many of whom only receive the standard conversations in most grammar books). The point is that such questions may be framed in non-trivial/productive ways. Some speakers may consider the conversations unspecialized while beginners may add them to a study book. If the site exists to help true learners then the value put in must be recognized.
    – Tommie C.
    Aug 29, 2014 at 11:02
  • 1
    @QuestionOverflow - ... In some cases strokes may be unclear. Especially with old documents. Removing such questions render the utility of images pointless. Users would not be able to ask a question in context for fear of it being declared character id. Sorry for the extended chat.
    – Tommie C.
    Aug 29, 2014 at 11:06
  • This is a Q&A site, so the question you ask must be specific. If it becomes conversational, then it would be like a forum or chat. I am not disputing that some questions itself might contain useful bits of information. But the question must be able to stand by itself without those bits. Otherwise, it loses focus and runs the risk of being off-topic. And, yes, some of the old documents where some words are smudged may require help in identification. Whether it is trivial or not is up to the community to judge.
    – 杨以轩
    Aug 29, 2014 at 11:20
  • Also, I think it would be helpful for users to know why their question is closed if their question is closed, rather than leaving them clueless with a generic close reason.
    – 杨以轩
    Aug 29, 2014 at 11:21

You must log in to answer this question.

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