- From: Boris Zbarsky <bzbarsky@MIT.EDU>
- Date: Tue, 19 Aug 2008 10:19:43 -0400
- To: David Poehlman <david.poehlman@handsontechnologeyes.com>
- CC: "Patrick H. Lauke" <splintered@gmail.com>, "Philip TAYLOR (Ret'd)" <P.Taylor@rhul.ac.uk>, Maciej Stachowiak <mjs@apple.com>, Anne van Kesteren <annevk@opera.com>, James Graham <jg307@cam.ac.uk>, Steven Faulkner <faulkner.steve@gmail.com>, Ian Hickson <ian@hixie.ch>, W3C WAI-XTECH <wai-xtech@w3.org>, public-html@w3.org
David Poehlman wrote:
> Right, either you are compliant with a spec or you are not. It is like
> programming. you can write a lousy app and it may work but... interestingly
> in some programming environments, they won't run at all if not done right.
Not that great an example, since you can write an app "per spec" and it
won't run in some programming environments. In fact this is a quite
common problem with a number of existing languages: following the
language spec is no guarantee of things working, because of bugs in the
compilers, interpreters, virtual machines, standard libraries, etc.
> Ths is not to say you can't choose to be non compliant, but we cannot
> provide for exceptions in the spec because those who don't want to be
> compliant but want to claim compliance will work hard to fit themselves into
> the exceptions.
This seems to be using a circular definition of compliance....
Ideally the spec would have the following properties:
1) Being compliant with the spec is the easy thing to do (leads to
the "right" behavior from the point of view of authors).
2) Being compliant with the spec leads to the "right" behavior
from the point of view of those viewing (hearing, smelling,
whatever) the content.
Sadly, this is pretty difficult because at heart authors are lazy (just
like all of us, and this is not a bad thing!) and the "right" behavior
from the viewer's point of view varies so widely by viewer....
-Boris
Received on Tuesday, 19 August 2008 14:21:24 UTC