![]() |
|
![]() |
||
![]() |
![]() |
|
[WWW-HTML Mailing List Archive Home] [Messages By Thread] [Messages By Date] Defining processing of invalid documents (Was: Charters for review (TAG issue tagSoupIntegration-54))
From: Ian Hickson <ian@hixie.ch>
Date: Tue, 21 Nov 2006 14:22:24 -0800 Message-ID: <403d38e20611211422i54f4c03fv737628bc1310acbc@mail.gmail.com> To: ed.rice@hp.com, www-html@w3.org (Thread moved to www-html@w3.org at danc's request.) Ed Rice <ed.rice@hp.com> wrote, on w3c-html-cg@w3.org: > > Doesn't including the non-well-formed/invalid documents in the > architecture/design actually validate the design and say that there is > no such thing as a non-well-formed document? No; you can simultaneously say that something is invalid and say exactly how it should be processed. CSS does this; for instance: p { color: "red"; color: green; } ...is completely invalid CSS (you can't pass a string to the property 'color'), but every conforming UA will handle it the same. In the case of HTML5, the HTML5 parser spec: http://whatwg.org/specs/web-apps/current-work/#parser ...defines what is an error (non-conforming syntax) and what isn't. HTH, -- Ian HicksonReceived on Tuesday, 21 November 2006 22:22:39 GMT |
|
||||||||||||||||