- From: Henrik Dvergsdal <henrik.dvergsdal@hibo.no>
- Date: Mon, 9 Apr 2007 21:50:02 +0200
- To: public-html@w3.org
On 9. apr. 2007, at 17.57, Anne van Kesteren wrote: > > On Mon, 09 Apr 2007 17:53:32 +0200, Henrik Dvergsdal > <henrik.dvergsdal@hibo.no> wrote: >> Lets say someone wants to create a 100% compliant browser >> completely from scratch, guided *only* by HTML5 and other >> standards external to HTML and with no prior knowledge of the >> functionality of current browsers. >> >> Should this browser then be compatible with "todays content"? > > Yes. > > >> In other words: Should the standard be comprehensive and detailed >> enough to ensure that this browser will be able to handle every >> single page on "the web as it is today" in a way that doesn't >> surprise people to much? > > Yes. Does this mean we have to include detailed, exact definitions and descriptions of every HTML feature that has a certain likelyhood of being used out there? In addition to all the new stuff, I guess this includes elements/ attributes/semantics/prescribed behaviours: * within HTML 4.01, * that have been deprecated (in HTML 4.01 and previous versions), * that became de facto standards without being described formally * that were implemented in some browser(s) but never became part of any HTML standard. How do we determine which elements/attributes/semantics/behaviours that can be safely left out from the standard? How shall we deal with with inconsistent behaviour between current browsers? Shall we choose one of them for the new standard or describe several options? -- Henrik
Received on Monday, 9 April 2007 19:51:20 UTC