- From: Nick Kew <nick@webthing.com>
- Date: Mon, 4 Feb 2002 22:38:23 +0000 (GMT)
- To: Charles McCathieNevile <charles@w3.org>
- cc: Jim Ley <jim@jibbering.com>, <w3c-wai-er-ig@w3.org>, <www-validator@w3.org>
On Mon, 4 Feb 2002, Charles McCathieNevile wrote:
> One possibility we discussed with EARL was to declare a "normatively valid"
> transformation, for example by running something through a given version of
> Tidy.
>
> Would this be a useful idea, or a bad one?
Both?
This is effectively what visval (and, underlying it, OpenSP) is doing,
and it would be very useful to me to declare it normatively valid.
OTOH it might be setting a target that would be out of reach to
someone like Jim, working with the browsers' parsers.
As for Tidy, I can immediately see several objections:
(1) For an online service like Page Valet, it's rather a big overhead
to put on top of what it's already doing.
(2) Preprocessing with Tidy is heuristic, not formally rigorous, and
would kill off validation. It would also screw accessibility
checking if we fixed some things up first!
(3) In some cases, Tidy refuses to produce any output.
--
Nick Kew
Site Valet - the mark of Quality on the Web.
<URL:http://valet.webthing.com/>
Received on Monday, 4 February 2002 17:38:28 UTC