Should we transcribe text from image?

Question

I submitted an edit for the following question:
How is 니까 used in this sentence?

In that edit, I transcribed the text from the image:

Q. 집에 (       ) 비가 내리기 시작했다

A. (1) 도착하거나     (2) 도착하기까     (3) 도착하거든     (4) 도착하려고

The edit has been reviewed, but rejected based on the following reason:

This edit does not make the post even a little bit easier to read, easier to find, more accurate or more accessible. Changes are either completely superfluous or actively harm readability.

While I disagree with the reason given, 2 people decided otherwise. I think this is something that should be discussed.

Should we transcribe text from image?

Answer 1

Yes, certainly. There is no reason to have a photo of text that contains the crux of the question itself. It makes the question virtually impossible for others to find through search.

The primary goal of Stack Exchange is to compile a collaborative work of sharing knowledge for those who come after. If the author is having trouble transcribing the text (or transcribed it incorrectly), then the community can help improve the post by using the wiki editing features to make the question more comprehensive and searchable.

But be mindful — screen captures of questions are often copied from a text book or other sources the author almost certainly did not have permission to copy. I don't suspect that was the case here, but please try not to use images where the text is needed to understand what you are asking.

Answer 2

While I agree that this might not make the question easier to find, it definitely makes it easier to read and thus, more accessible. Here are a few reasons:

• Visually impaired users can struggle to read text on image, especially when the contrast is low. In this case, the contrast is not that great - I, for example, had some issue to read it.

• Screen readers can't process images. People using them won't be able to understand the context, and will need to struggle to guess based on the answers. Even though, in that particular case, the image has been transcribed in the answers, I highly doubt that visually impaired users will go as far as to read them if they can't understand the question.

• There's no harm done in making it easier to be found. The text seems to come from a resource helping people to learn Korean. Even if the content might not be of paramount importance, other people using the same resource to learn Korean might struggle on that question. (As the author did.) Allowing search engines to index it should be welcomed.

People might post more and more screenshots of text written in Korean. Typing Korean on a occidental keyboard might not be easy, and many people might find it more convenient to directly post a screenshot instead of writing the content. Thus, we should encourage this kind of edit.

Answer 3

Usually, I would say yes - but I'm aware that there are a couple of questions where I have put in photographs of text:

What's the difference between 상태동사 and 형용사? - where the photo serves as confirmation of a reference - and

How does 15th Century Hangul writing/type differ from modern Hangul?, where the writing would be very hard to reproduce and would lose context.

So I think this has to be considered on a case-by-case basis - a blanket rule might not be appropriate.