When reviewing, some diffs are practically invisible. In particular changing between blockquote (>) and spoiler (>!). Yes, one could inspect things in detail looking at the markdowndiff. But who does this for a minor formatting edit? I would not have noticed this, would I have not fallen prey myself for such an edit.

As a suggestion, such differences should be made more visible, somehow.

This is somewhat similar to this issue.

  • I don't understand this, can you clarify what you mean? The diff of changing a blockquote to a spoiler looks like this: i.stack.imgur.com/6TeVt.png That is hardly "practically invisible". – balpha Feb 21 '13 at 11:44
  • @balpha: Not sure if that changed in the meantime. But even if it was the difference you show now, the actual difference is not visible. That is, it looks like a simple formatting issue. Some indentation or the like and not something related to spoilers. – false Feb 25 '13 at 16:34
  • Nothing changed in that regard. And that diff makes it very obvious that something changed. So "practically invisible" is a big stretch. That's why the Markdown diff exists -- some things just can't be shown visually. – balpha Feb 25 '13 at 16:58

But who does this for a minor formatting edit?

I do.

That being said, I agree things should stand out more simply because not enough people care about the system to use it properly and it's kind of annoying that edits get blindly approved because people don't see little changes that make a big difference. My idea is to have the system compare the Markdown-processed diffs to figure out what exactly got changed, and put a nice little box off to the side pointing to those changes. Also include a box at the top pointing to the other options so users know they can switch between views to see what's going on, to make sure they use the reviewing features we provide to their maximum potential.

See this screenshot I spent a lot of unicorn points to make:

View 1

Or take this more realistic example of how it appears I just randomly added text to the answer, where I actually extracted it from a <!-- HTML comment --> already provided:

View 2

Without looking at the Markdown view, it looks like I'm just adding stuff. But with that nice yellow box saying I also removed some HTML that isn't allowed via the Markdown parser, people now know "hey, there was something else there." This is basically using the "Show rendered output" option, which ordinarily doesn't get used much, and isn't even available when reviewing suggested edits. But it has that nice 200px margin over to the right (for whatever reason) which would be perfect for highlighting key spots where changes were made. Users would just have to scroll down the right side looking for the yellow boxes to see if they missed anything.

I can dream can't I?

  • Good to know there are some people still here who care about this. But do you think that those who skip through quickly do harm on purpose? I think not. I believe it is rather this interface which is so suggestive. What do I have to do? Oh, there are couple of edits to review, that can be done so quickly... At least we do not have the traditional comment spam on this site at all. Like visit n_s_f_w.com (that site actually does exist without _ ...). Maybe another idea: Requeue already (uncontroversially) rejected edits to see how others react. – false Nov 15 '12 at 20:31

You must log in to answer this question.

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