- From: Karl Dubost <karl@w3.org>
- Date: Wed, 27 Oct 2004 12:55:55 +0100
- To: www-qa@w3.org
- Message-Id: <28E27913-280F-11D9-8C06-000A95718F82@w3.org>
Hi QA WG, *MEMBER ONLY Discussion* Hi, I sent an initial message about Conformance and XHTML 2.0 on the HTML Member WG mailing-list. Be careful the discussion is member confidential. http://lists.w3.org/Archives/Member/w3c-html-wg/2004JulSep/0165 http://lists.w3.org/Archives/Member/w3c-html-wg/2004OctDec/ thread.html#36 Basically to broaden my question is: 1) How do we prove semantics 2) How do we test meaning in Markup languages? 3) What does that mean to specify the meaning of an element if it's not enforced by the conformance section? Example: If I create PoemML An element "verse" is created and defined by verse: A single metrical line of poetry, or poetry in general (as opposed to prose). We may define an attribute poemtype for verse, with different values: alexandrine: A line of poetry that has 12 syllables. heptameter: A line of poetry that has seven metrical feet. For example: <verse poemtype="alexandrine">But satire needs not those, and wit will shine</verse> How do we prove that this element has been implemented for the specific meaning it has been defined for? Not proving that the content has the right semantics, but that the element is useful and why it's useful has to be demonstrated somehow. People will use an element because it's useful and it does something specific in an application (browser, authoring tool, search engine, annotation tools, etc...) -- Karl Dubost - http://www.w3.org/People/karl/ W3C Conformance Manager *** Be Strict To Be Cool ***
Received on Wednesday, 27 October 2004 11:55:58 UTC