1

A long introduction

There's quite a new bug report about using URLs with Unicode characters, which was closed as exact duplicate of an old one. But the old one was marked as bydesign (sort of closed) based on this false assumption:

  • closing as bydesign since there is such an easy workaround (use the normal character form, which Wikipedia always supports) – Jeff Atwood

But there's not only Wikipedia in the world, is it? Actually, this is a real problem, at least for newbies. In fact, there are two problems

  • Unicode in the hostname
  • Unicode in the path

which need to be solved differently. You may be lucky and find a page doing a conversion for the part you need, or you may learn Unicode details, punycode and/or hexadecimal URL encoding (not so much fun, IMHO). There may be other simple solutions, but a newbie hardly knows them.

So, it's a real and unsolved problem, isn't it? It doesn't happen often for most of us. I personally hate Unicode in URLs, but it happens.

The questions

  • How should a newbie find out what to do when the allegedly original question does not contain the answer for their case? In fact, the new bug report concerns the hostname and the old one contains solutions for the path only.
  • Don't you think that questions get closed too quickly sometimes?
  • Shouldn't there be a mechanism preventing situations where long existing bugs (unsolved problems, incomplete answers, etc.) get ignored forever as in this case?
Improve this question
7
  • 3
    Looks like a whine (very) thinly disguised as a set of questions.
    – Michael Petrotta
    Commented Jan 27, 2011 at 5:26
  • 1
    But it is not. I was having such a problem, but it's solved for me now. And it wasn't my question which was closed. I just wanted to point to something I consider to be a problem.
    – maaartinus
    Commented Jan 27, 2011 at 5:30
  • @Oded, as an aside, just curious: I see you're deleting many old bug reports (probably as they're no longer relevant). Any reason why those are not tagged cannot-reproduce or something instead? (Deleting will make it harder for me to find references that I still remember in my head.)
    – Arjan
    Commented Jul 12, 2013 at 10:09
  • 2
    @Arjan - Don't want to clutter the front page and many of these are completely irrelevant (old browsers, pages that are no longer there, code that has been completely re-written etc...). A no-repro is not really correct either.
    – Oded StaffMod
    Commented Jul 12, 2013 at 10:14
  • @Oded Can't you close them as "off-topic because [they are irrelevant]" then?
    – Tobias Kienzler
    Commented Jul 12, 2013 at 10:51
  • @TobiasKienzler - They are not off-topic - they clearly belong on meta and I'd rather not abuse the close reasons. Closing them doesn't help with setting the status either. They would also end up cluttering the recently closed list, where other things are more important.
    – Oded StaffMod
    Commented Jul 12, 2013 at 10:57
  • @Oded If you say they are irrelevant, e.g. due to old browsers that are not supported, I'd claim a bug concerning an unsupported browser is off-topic. But then again, how many people are still using these ancient beasts and will yet manage to navigate meta.SO well enough to find the report stating their browser stinks...
    – Tobias Kienzler
    Commented Jul 12, 2013 at 10:58

1 Answer 1

Reset to default
2
  • How should a newbie find out what to do when the allegedly original question does not contain the answer for their case? In fact, the [new bug report] concerns the hostname and the [old one] contains solutions for the path only.

That's not true. The original does contain the answer for their case: Just use the editor's hyperlink button, which will encode the URL correctly.

  • Don't you think, that questions get closed too quickly sometimes?

Yes, sometimes questions get closed to quickly. And at other times, questions don't get close fast enough, or not at all.

  • Shouldn't be there a mechanisms preventing situations where long existing bugs (unsolved problems, incomplete answers, etc.) get ignored forever as in this case?

The assumption "as in this case" is wrong, but in general: These mechanisms already exist. Suggest edits (or edit directly if you have the rep), post new answers, reopen incorrectly closed questions, etc.

Improve this answer
7
  • 1
    Nope. Using the hyperlink button gives this link, which is invalid. I wonder if I should open a new bug (this time the bug is that the hyperlink button does not encode correctly).
    – Johannes Kuhn
    Commented Jul 12, 2013 at 10:45
  • How is it invalid? It works fine for me.
    – balpha StaffMod
    Commented Jul 12, 2013 at 10:47
  • 1
    Well, maybe it depends on the browser. I use FF 22.0, which displays the url correctly, but tries to access www.erfolg-f%C3%BCr-alle.de (yes, with the %C3%BC) which fails with "host not found". Anyway, I don't think that such an url encoding is standard.
    – Johannes Kuhn
    Commented Jul 12, 2013 at 10:49
  • 1
    bugzilla.mozilla.org/show_bug.cgi?id=309671 tools.ietf.org/html/rfc3986#section-3.2.2
    – balpha StaffMod
    Commented Jul 12, 2013 at 10:54
  • Well, I see. A bug in firefox. Do we really go back to the dark age where we had to do the same thing three times for every browser or do we just say "well, that is a bug in your browser, nag the developers".
    – Johannes Kuhn
    Commented Jul 12, 2013 at 11:24
  • I have no idea what you're trying to say.
    – balpha StaffMod
    Commented Jul 12, 2013 at 11:27
  • In the dark age there was the browser war, document.all, lot of work for web devs to make the site work with every browser. Now most of the stuff should be working on all browsers, but this is not always the case.
    – Johannes Kuhn
    Commented Jul 12, 2013 at 11:40

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