33

I formulate this feature request particularly with language sites in mind, but it may also be useful to other Stack Exchanges:

On some, if not all language sites, blockquotes (triggered with a > at the beginning of the line) are the preferred way to format larger language examples. Now, sometimes, it’s useful to give examples of incorrect language use to illustrate something. In this case it’s desirable that the reader can directly see that the example text is incorrect because:

  • Writing that examples are incorrect in the accompanying text can be tedious and break the text flow.
  • If the information that an example is incorrect can only be deduced from the accompanying text, it is confusing for readers who only skim a post, e.g., to recapture some information.
  • People who are not fluent in the language of the post (which can be English or the language the site is about) may have difficulties understanding the accompanying text and thus might miss that the text is wrong. Since language learners are among these people, they might mislearn aspects of the language due to this.

Right now, there are mainly three options to do this:

  • Use HTML strike-through. This has the advantage that it is generally understood, but it makes the text very difficult to read. Also, this is not markdown.
  • Use a symbol at the beginning of the example, usually an asterisk (*). This is good to read, but usually only understood by people who know this notation.
  • Write “[wrong:]” or similar at the beginning of every example. This is generally understood, but still breaks the reading flow and is somewhat unaesthetic.

Therefore, it would be nice to have a formatting option for such incorrect blockquotes, similar to spoiler markdown. This should be rendered similar to blockquotes but in a way that is generally understood to be negative. To give an example for the blockquote style used by most sites, the background could be tinted red and the left bar could be zigzagged:

enter image description here

While this feature may not be that important, its cost–benefit ratio should be comparable to the spoiler markdown, in particular as its implementation can mostly happen in parallel.

  • If somebody has a good English example post that could use this (preferrably from ELU or ELL), please tell me. I could only find posts in German with a quick search. – Wrzlprmft Nov 30 '14 at 23:18
  • 8
    This could be useful for programming sites as well. It would have to work for code blocks, though, not only for quote blocks. It's common to quote wrong code from the question to explain what exactly the problem is, and then include a corrected version of the same code. Would be kind of nice if they looked different. Example here: stackoverflow.com/a/27216207/3530129. The first block is "bad", the second "good". – Reto Koradi Dec 1 '14 at 6:42
  • 2
    We need to make sure the difference is clear to color-blind users as well. I wonder how obvious it would be based on the zig-zag at the left. – snailboat Dec 23 '15 at 12:21
  • @snailboat: I am no expert on colour-blindness, but I would expect that a sufficent difference in brightness of the background should be recognisable by everybody. Of course, this does not communicate what the difference is. (Also, isn’t such colour-blindness extremely rare?) – Wrzlprmft Dec 23 '15 at 12:26
14

We recently looked into this feature request. I really liked this idea; it rates very high on that cool scale. But it also raised a lot of issues about adding a change like this to Markdown, and we concluded that we are not able to implement this.

Using open standards in Stack Exchange like Markdown, Open ID, CC-WIKI, LaTeX, Gravatar, etc were selected for their high accessibility and complete re-usability of our content. But this feature-request takes a big step away from that philosophy on two counts:

It is a very proprietary markup. One of the hallmarks of Markdown is that it's supposed to be clear and readable even when displayed in *plain* text. Even when we've gone outside the Markdown standard, we were careful to make sure the text is still readable to anyone who doesn't implement it. But this would be inventing a markup that literally changes the meaning of what is written underneath. While we may be able to create our own visual cues that this text is incorrect (including on mobile and all the apps that use our content), the text will become misleading-to-incorrect to everyone else.

It's fabricating a proprietary way to communicate. I was looking for some real-world use cases for an "incorrect example" format. There aren't any. This does not exist in any other formatting standard that I can find. I think we may be taking a leap of faith that squiggly-indent-red-background will always be intrinsically understood to mean "this is wrong" wherever it is seen.

When correctness and universality are so fundamental to what we do, I don't think it is wise to get into a communication situation like "voting means something different in meta" when it comes to conveying the meaning of what people write.

I love the idea. I just don't think we can do it. :(

  • 2
    Thank you for considering my request. Two remarks though: 1) Prepending an asterisk to wrong examples is commonly used on language Stack Exchange sites right now and also in the linguistic literature. So if you decide to use >* to markup the wrong example, this would indeed work quite well in unrendered markup for audiences familiar with it. On the other hand, I agree that this is a rather small audience. 2) As for the specific way of rendering this markup, I am not fixed on the squiggly-indent-red-background – but then, I have no better suggestion to make right now either. – Wrzlprmft Jan 29 '16 at 16:47

You must log in to answer this question.

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