Re: [VE][108] Error Message Feedback

Hi David

Thanks for taking the time to reply!

I have to admit, I was unaware of a CSS alternative to the <hr> 'color"
attribute: I shall have to do more study.

I've also found it very hard to get definitive info on DTD, and must admit
to always pasting the same line in as I don't know what the alternatives
are.  There also seems to be endless warnings about not using code that is
too up to date, as only computers with the very latest browsers installed
will render it correctly. Is this a valid point?  As a Mac user, still stuck
on OS9, even with the latest (or last) versions of the various browsers
installed I find many problems viewing pages: even on Microsoft's IE site.

Yours
Gavin Scott
www.jazz-trio.com



From: David Dorward <david@dorward.me.uk>
Reply-To: www-validator@w3.org
Date: Tue, 23 Nov 2004 13:12:13 +0000
To: Gavin Scott <gavin@bassplayer.co.uk>
Cc: www-validator@w3.org
Subject: Re: [VE][108] Error Message Feedback


On Tue, Nov 23, 2004 at 12:55:38PM +0000, Gavin Scott wrote:
> I just wanted to make a comment, however, about the attribute "color" for
> the element "hr". Whilst I realise this is not truly valid HTML 4.01, it is
> used by by far and away the biggest browser (IE), and has no ill effect on
> any other browser.  I think it would make sense to acknowledge this, with a
> remark, rather than calling it an error.

The validator compares the given document against the given DTD. If
the attribute doesn't appear in the DTD then it is an error. If you
wish to use non-standard code (which I do not recommend, especially
when the same effect can be achieved using CSS - which is the correct
tool for such presentational matters) then you should use a DTD which
includes those attributes. (Of course, if you do this then what you
will be writing still won't be HTML, but it will be valid markup).

> I also can't see any harm in incorporating it as an option into
> HTML.

That would be an issue to raise with the HTML Working Group. However,
given that (a) development on HTML has ceased in favour of XHTML and
that (b) since HTML 4.0 (which was published back in about '96 and
went under development before then) has focused on pulling *away* from
the presentational junk added during the first browser wars it seems
very unlikely that more presentational junk will be added to the
specification.

> I don't know if this is the right place to air this, but it also seems to
> make sense to add alternative English (as in non-US) spellings to common
> HTML code, such as colour, and grey.  Having had years of these spellings
> drummed into us it is very difficult for us 'Limeys' to spell these words
> wrongly all the time!

As a fellow Brit I can only say that syntax highlighting editors,
tools like the validator and practise make tend to make this problem
go away. If it doesn't, then it shouldn't be a non-trivial operation
to write a simple preprocessor to change attributes with Oxford
spelling to Webster spelling.

It would, however, be another matter for the HTML working group, but
one I doubt you'll see much enthusiasm for. It would lead to all sorts
of interesting situations where both spellings of the attribute could
appear attached to the same element.

-- 
David Dorward                                      http://dorward.me.uk