- From: Aryeh Gregor <Simetrical+w3c@gmail.com>
- Date: Fri, 27 Aug 2010 15:00:28 -0400
On Fri, Aug 27, 2010 at 6:32 AM, Hugh Guiney <hugh.guiney at gmail.com> wrote: > But, I thought XHTML5 was just an XML serialization of HTML5, so why > is this the case? I just read the rationale behind it, but despite not > being best practice shouldn't it be at the very least allowed? This is guesswork on my part, but maybe it's correct. <noscript> in text/html has magical parsing effects, which cannot be replicated in XML. Specifically, if scripting is enabled, its contents will be parsed as plaintext, which is then hidden. In XML, you'd have to specify for every single element that its effects are ignored if it's a descendant of <noscript> and scripting is enabled. For instance, the text/html markup <noscript><meta http-equiv="foo" content="bar"><meta http-equiv="baz" content="quuz"></noscript> will result in a DOM with a noscript element and a single text node child. The equivalent markup in XML must instead result in two elements as children, which the rest of the UA then has to specially ignore at some higher level than the parser. So I'm guessing that browsers didn't implement <noscript> in XML since it would have been a pain and lots of people don't like it anyway; and nobody wants to change it now because it would still be a pain and few sites use XML. But that's just a guess, someone else can give a more informed answer. On Fri, Aug 27, 2010 at 9:59 AM, Julian Reschke <julian.reschke at gmx.de> wrote: > The HTML WG is currently discussing whether it should be deprecated (in > HTML), see <http://www.w3.org/Bugs/Public/show_bug.cgi?id=10068>. > > If the outcome of this is that there are good use cases for <noscript>, I'd > expect that it will also be allowed in XHTML. I'm quite sure that <noscript> was not banned from XHTML just because it lacked use-cases, and the editor currently thinks it has use-cases regardless, so I don't expect it to be allowed in XHTML regardless. Note that it's not just invalid in XHTML -- it actually has no effect, as an *implementation* requirement.
Received on Friday, 27 August 2010 12:00:28 UTC