- From: Jukka K. Korpela <jkorpela@cs.tut.fi>
- Date: Tue, 18 Oct 2005 14:27:19 +0300 (EEST)
- To: Livio Mondini <livio.mondini@tiuvizeta.it>
- Cc: www-validator@w3.org
On Tue, 18 Oct 2005, Livio Mondini wrote: > Yes David, but validator respond *big and in green*: > > This Page Is Valid -//W3C//DTD XHTML 1.1 //EN! > This is a problem, i think. It is correct up to and including the word "Valid", ignoring the somewhat pompose-looking capitalization for now. The problem is that the validator tries to be helpful by identifying HTML version, which is just nonsense in SGML or XML terms: the document type declaration only identifies the document type definition, which in turns specifies the syntax of markup. None of this involves any "HTML version" in any way. A validator should say "yes" or "no", and in the latter case, it should present the reportable markup errors. Trying to guess that the markup was really meant to be HTML of some kind seems to lead to endless confusion. If the validator said "yes" (or "yes, it is valid") with no reference to any DTD and still less to any specific name, the results might still be confusing when the document uses a DTD with "html" as the root element's name but isn't HTML. Yet, in that case, the validator message would not be outright misleading. Besides, if the simple and correct answer were given instead of complex and potentially misleading (or wrong) answers, there might be room for simple statement like the following: "The validator verified that the document at ... conforms to the Document Type Definition that it declares." This might be inconvenient, since then the user (author of document, as a rule) would need to find out what this means. I still think this would be far better than giving her or him an easily digestible completely wrong idea thereof. -- Jukka "Yucca" Korpela, http://www.cs.tut.fi/~jkorpela/
Received on Tuesday, 18 October 2005 11:27:37 UTC