2012-07-30 19:49, Michael[tm] Smith wrote: >> Because the “validator” is, in HTML5 mode, a heuristic checker, reflecting >> the opinions of its authors at the time of writing its code, not objective >> criteria. > > What authors do you mean? The authors of the tool or the authors of the > HTML5 spec? The authors of the tool. I do not know their exact relation to the authors of HTML5 drafts. And I do not know whether the authors of the tool are trying to check against the most recent W3C HTML5 draft or the WHATWG HTML Living Standard, or who the authors of the tool exactly are. Much of the “HTML5 spec” (i.e., one of the above-mentioned documents, that may change at any moment without prior notice) is still too vague for objectively decidability of conformance. So quite understandably, the authors of the tool have had to make subjective decisions (that is, what *they* regard as proper, or as correct intepretation). They also decide which version of the “HTML5 spec” is used, but this is *not* disclosed in any way. An SGML or XML validator has a rigorously defined task and logic, which have been publicly documented. Not so with the HTML5 mode. YuccaReceived on Monday, 30 July 2012 18:02:07 GMT
This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 30 July 2012 18:02:13 GMT