I agree that we should all encourage use of valid HTML. However,
developers of HTML consuming application and tools, whether they work for
Microsoft or a hole-in-the-wall, do not have the luxury to spit back bad
HTML pages. Only the developers of HTML producing applications and tools
have the ability to improve the situation.
I think that portable HTML generation framework which embodies the latest
HTML spec would go a long way toward improving the level of HTML standard
conformance. If W3C is interested, I can help.
> From: Daniel W. Connolly <firstname.lastname@example.org>
> To: Don Park <email@example.com>
> Cc: Kim Liu <KLIU@us.oracle.com>; firstname.lastname@example.org
> Date: Saturday, September 21, 1996 8:39 PM
> In message <199609210219.TAA21489@gw.quake.net>, "Don Park" writes:
> >A good HTML parser should treat the HTML specs as GUIDELINES only and
> >for maximum deviation from the spec.
> "Good" is a relative term.
> I'm fond of saying "We all create the world around us."
> What that means in this case is: the more the tools encourage
> users to make sloppy HTML, the more people will make sloppy HTML.
> I realize that a small shop or consultant might not have a huge
> impact on the market as a whole and would be, to a certain
> extent, just pissing in the wind if it were too strict.
> But every little bit counts. The HotMetal users seem to enjoy the
> confidence they get from dealing with valid HTML, for example. It
> seems to have earned them a place in PC magazine etc.
> Anyway, www-lib is about building shared understanding in the
> development community. And I'd like the shared understanding to be
> represented in the specs, and not in the black-magic and lore
> surrounding the specs. I know that keeps the consultants employed,
> but it's bad for everbody in the long run.
> Please, stick to the specs (or get them changed) and encourage others
> to do so.