- From: Justin James <j_james@mindspring.com>
- Date: Fri, 9 May 2008 09:48:57 -0400
- To: "'Ian Hickson'" <ian@hixie.ch>
- Cc: "'Steven Faulkner'" <faulkner.steve@gmail.com>, <public-html@w3.org>, "'W3C WAI-XTECH'" <wai-xtech@w3.org>, <wai-liaison@w3.org>, "'Dan Connolly'" <connolly@w3.org>, "'Chris Wilson'" <Chris.Wilson@microsoft.com>, "'Michael\(tm\) Smith'" <mike@w3.org>
> That's not entirely true, we can mandate whatever we want. We just can't > machine-check much beyond presence and non-null-ness. For example, we > require that authors use <h1> for headers, but we can't check it without > a human (at the moment -- AI might get there eventually!). Anything that cannot be machine-checked should not be in the spec. There are too many automated user agents (search engines, screen scrapers, etc.) and too many automated authoring tools (anything generating HTML without user intervention, 99% of Web apps) out there. If it is not verifiable by the logic that general purpose programming languages allow, how is the programmer of such tools going to ensure that the software generates (or properly consumes) HTML that meets the spec? In a nutshell, a non-verifiable HTML spec relies upon a "I know it when I see it" definition of things, which will get us into trouble. If the user of the spec does not share the same internal "I know it when I see it" definitions, then they will be meeting the tech specs but not the assumed specs. J.Ja
Received on Friday, 9 May 2008 13:50:13 UTC