[thelist] popgurls preview

[aardvark]

On 16 Jan 2005 at 1:05, Richard Bennett wrote:
[...]
> I'm not saying they'll necessarily fail, or that there's anything wrong with 
> your site in this regard, simply that well-formed code (be it html4 or xhtml) 
> is easier to parse with any parsing tool, and that by validating xhtml you 
> will catch these errors, while html4 validators allows unclosed tags to slip 
> through. 

i can believe that closed elements are easier to parse, but i am 
suspect of your assertion in practice... IE converts HTML/XHTML to 
its own MSHTML internally to parse... which, unfortunately, does 
things like drop some closing tags, remove quotes from attributes, 
and other nastiness...

i suspect browsers don't treat XHTML as anything other than HTML 
internally... i can't back that up, but i think there are years of 
anecdotal evidence to support that...

but, yes, an HTML validator may let an unclosed <li> slip through, 
but it's also technically not an error according to the spec... i, 
however, close them, and require my staff to do so as well...

> > to what tools are you referring?  what sort of real-world impact does
> > a non-self-terminated tag have on them?
> None probably, my point was that if you require a certain standard from people 
> you work with, either you can ask for xhtml/508 which will give you clean 
> code with alt/title attributes and more, or you can ask for html4, and then 
> specify that alt/title attributes are required and <li> and <p> needs to be 
> closed etc etc, which seems more problematic to me.

not really... you can ask for html/508/wai, after all... that kinda 
precludes some of the issues you're mentioning... and if you tell 
your people to close their tags (with obvious exceptions) as a matter 
of course, you don't need to manage it any further than that...

> > i'm on joe clark's accessibility mailing list, too,
> Which is that exactly? I found 6 but not sure which is for accessibility...

erm, perhaps i was remiss in mentioning it... it's a list for toronto 
developers, but since i am close and often head up to the pointy city 
in the great white north, i get to lurk and occasionally get spam-
filtered by members when i post...

> > and i haven't 
> > heard any discussion of that nature... more along the lines of non-
> > semantic/structural problems (which i don't believe i have)...
> Again, nothing wrong with your site, this is just about whether or not xhtml 
> is a good default standard to go for, or not.

i think it is... but not in every case...

> Joe himself writes: "Like all my new uploads, the page validates as proper 
> XHTML 1.0, so it will not be incompatible per se with any browser or screen 
> reader."
> http://www.joeclark.org/access/crtc/digitaltableintro.html

i gotta check on that... i still remember someone posting an example 
page in XHTML, using the proper MIME type (otherwise it really isn't 
valid, now is it?) that caused a browser to break... gonna go google 
now...

> That's what I mean too, not that xhtml is the ONLY way to be accessible.

no, it certainly isn't...

interestingly, i think the last time pages were generally accessible 
was way back before tables were made part of the spec, and font tags, 
etc... everything was marked up in a tag that had a purpose, and 
there weren't tables to obfuscate the content, or plug-ins to hide 
it, or browser issues, etc...

ultimately, we've finally come 'round full circle...

ok, michelle yeoh is still the best bond girl... kickin' butt and 
takin' names...