W3C home > Mailing lists > Public > www-tag@w3.org > May 2002

Re: New issue: error recovery practices (Re: Proposed TAG Finding: Internet Media Type registration, consistency of use)

From: Simon St.Laurent <simonstl@simonstl.com>
Date: 29 May 2002 23:14:15 -0400
To: "www-tag@w3.org" <www-tag@w3.org>
Message-Id: <1022728455.814.2433.camel@localhost.localdomain>

On Wed, 2002-05-29 at 22:20, Keith Moore wrote:
> I'd like to suggest a different architectural tack -
> 
> Errors should be detectable by, or communicated to, the party
> that is in the best position to fix them.

That's an excellent point.  

However, in my experience, the primary and likely only testing that is
done by the people creating the errors is through the end-user
experience.  

Traditionally this has meant viewing in browsers, particularly the
browser that is used by the person (representing the organization)
paying for the work being done.

I would be stunned if a client whose site looked horrible in their
preferred browser would be content with "but the W3C validator says it's
perfect".  To take just one example, a certain browser blows up on a
fair number of legal XML documents, and people shy away from those legal
but supposedly wrong practices.  (Now I have to go track down those
documents, but they're mostly victims of a double parse.)  On the
opposite side, I see a lot of odd DTD syntax, mostly around mixed
content, that "works in XYZ tool."  (Putting #PCDATA in parameter
entities seems to fix all woes.)

Even among the developers I know who take standards seriously,
validation is only a part of the project.  More cynical (or merely
uninformed) developers skip that step entirely and test only in one or
(maybe) two likely environments.

The most cynical among us are trying to figure why exactly slapping a
DOCTYPE declaration on something makes it right, especially the poor
souls who were fond of uppercase MARKUP or who use frames.

If something is going to be standards-compliant, I'd really prefer that
it comply with rules that require reporting of compliance violations.  

Without those rules, _especially_ at the level of end-user experience,
this exercise is quite severely farcical.

-- 
Simon St.Laurent
Ring around the content, a pocket full of brackets
Errors, errors, all fall down!
http://simonstl.com
Received on Wednesday, 29 May 2002 23:08:52 UTC

This archive was generated by hypermail 2.3.1 : Wednesday, 7 January 2015 15:32:32 UTC