-1

This question already has an answer here:

Original question was resolved; I'd missed the backtick escapes.

However my second point remains: In comments, unescaped URIs have the scheme stripped off them as part of the process of converting them to links. Thus, a careless attempt to discuss http://foo.com in comments displays as foo.com.

I don't think stripping the prefix off actually serves any useful purpose. Browsers generally assume the http: scheme and automagically reintroduce it, but other tools generally don't, and not every discussion of URIs is in the context of a browser.

So, low-priority suggestion: Were it my system, I'd get rid of the stripping. It has the potential to cause confusion, and I don't think it has value to offset that.

Concrete suggestion: Rather than displaying as foo.com, display as http://foo.com

Up to you, of course.

EDIT: Yes, I saw the possible dupe. Like the backticks, the answer suggested there is a workaround for the rendering problem... but I still consider the default rendering a problem. It may make sense to combine this question into that one.

marked as duplicate by Shadow The Curly Braced Wizard, Martijn Pieters, Hugo Dozois, user213963, Lucifer Jan 3 '14 at 0:47

This question has been asked before and already has an answer. If those answers do not fully address your question, please ask a new question.

  • 3
    use back ticks xmlns="http://foo.com", – OGHaza Jan 2 '14 at 14:45
  • 2
    The browser doesn't "automagically" reintroduce the http:// it's still there but only in the html anchor. I think you'd have a better case for arguing that the www. shouldn't be hidden, since http:// is undoubtedly implied when no specifying a protocol before a URL. – OGHaza Jan 2 '14 at 15:11
  • Sorry, I wasn't clear... Browswers, if you type a URI without a scheme into their command line, will try prefixing it with http:// as one of their steps in trying to resolve it. Thus, telling the user to type foo.bash (without escaping) into their browser is OK, because when they do so the right thing will happen despite the scheme having been hidden. Telling them to user that URI in non-browser contexts causes unnecessary confusion; if they copy it visually they get one result, if they copy-URI they get another result, and the former may not work for their application. – keshlam Jan 2 '14 at 15:16
  • I'm suggesting that rather than displaying as foo.com by default, we should display as 'http://foo.com' – keshlam Jan 2 '14 at 15:18
  • @keshlam did you see this? – Shadow The Curly Braced Wizard Jan 2 '14 at 15:23
  • Thanks SW -- good workaround, but I'm suggesting we shouldn't have to work around; the default's a bad choice. – keshlam Jan 2 '14 at 15:28
6

You quote URIs that are not meant to be linked by adding backticks around them:

`xmlns="http://foo.com"`

is rendered as xmlns="http://foo.com".

This works both in comments and in posts; you are discussing code, so mark it up as such.

Note that this not a bug; the MarkDown parser used here explicitly makes all URLs into clickable links, see https://meta.stackoverflow.com/editing-help#bare-urls

  • Good to know, thanks. Not at all obvious to newer users -- you might want to think about hinting more strongly -- and I still don't think that stripping the scheme out of the visual representation actually serves a useful purpose, but I'm willing to drop this from bug to wishlist and accept "working as designed, wontfix" as an answer. – keshlam Jan 2 '14 at 14:49
  • Also note that the issue isn't clickable links, but the fact that the clickable links' visible rendering has the scheme prefix stripped off it, creating unnecessary confusion for not obvious gain. – keshlam Jan 2 '14 at 15:13
  • For most links you want the readable version in the comment. Shadow linked you to a dupe if that is the real problem you are having. – Martijn Pieters Jan 2 '14 at 15:15
  • Question rephrased... and yes, this may be a dupe. Feel free to close it if so. – keshlam Jan 2 '14 at 15:21
  • Duplicate question, but the answer given there is a workaround rather than a resolution. I'd still prefer to see the needlessly misleading behavior fixed. If folks would prefer, we can make this a whole discussion a comment upon that q and kill this one. – keshlam Jan 2 '14 at 15:30

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