- From: Anne van Kesteren <annevk@opera.com>
- Date: Thu, 14 Jun 2007 13:54:34 +0200
- To: "Sean Owen" <srowen@google.com>
- Cc: "Ben 'Cerbera' Millard" <cerbera@projectcerbera.com>, "mobileOK WG" <public-bpwg-comments@w3.org>
On Wed, 13 Jun 2007 17:26:29 +0200, Sean Owen <srowen@google.com> wrote: > The tests require the page to validate as XHTML and thus as XML, so > mobileOK pages are being authored against a device that will validate, > in theory. Some browsers may parse this as text/html tag soup, which > should not present a problem (right?) -- the document happens to be > nice and correct XHTML. > > The reverse isn't true, right -- valid HTML is not necessarily valid > XHTML. But the test isn't proposing the reverse. I understand the > danger you cite in serving malformed XHTML, but surely serving > well-formed valid XHTML does not present this problem. One problem is that the validator allows xmlns="http://www.w3.org/1999/xhtml" to be omitted for instance. Another problem is that authors write scripts that don't guarantee well-formed output. Yet another problem is that there are some differences between XHTML processed as HTML and XHTML processed as XML such as scripting, styling and XHTML syntax differences. In XHTML you can write <textarea/> in theory, but not when it's processed as HTML, etc. > Dom says that some mobile devices will reject "text/html" document or > otherwise treat them as not suitable for mobile, since they expect a > type consistent with mobile-friendly formats like XHTML MP and Basic. That's sad. -- Anne van Kesteren <http://annevankesteren.nl/> <http://www.opera.com/>
Received on Thursday, 14 June 2007 11:55:33 UTC