[thelist] Re: WebReview responds to WaSP browser death march

Paola Kathuria paola at limitless.co.uk
Sat Mar 17 14:51:27 CST 2001


[Sorry for this late response - I'm, um, rather behind on list mail.]

Sabrina Dent wrote:
> well, while the statement may not be qualified correctly, it is still
> accurate... those who use WYSIWYGs *exclusively* (which is the
> missing qualifier) are still going to churn out non-compliant code...

>From what I've seen, WYSIWYG editors made by the companies that also
make browsers (e.g., Microsoft) tend not to produce compliant HTML.
I've recommended HotMetal in the past (when asked by clients which
tool to use to update their own sites); it has no browser bias,
doesn't rewrite HTML and includes a validator against different DTDs
(this was a few years ago).  So, I agree that not *all* WYSIWYG
editors necessarily produce non-compliant HTML.

However, FrontPage (I've used the version before FP 2000) produces
vastly invalid HTML.  The most common problem I've seen is not
opening and closing tags in the right order.  A common sight was
"<h2><b>About the company</h2></b>" (who knows why FP puts <b> tags
around headings!).  MSIE is forgiving of this invalid HTML but
Netscape will show the remaining page in h2 (that is, large and
bold).  There are lots of other problems (perhaps they're fixed in
FP 2000) which make the page take longer to load, display and debug.
Anyway, I digress.

> If I care, or if the clients cares, about vailidation, then I can run
> it through a validator and change the HTML as required.

Well, in theory you could.  But in practice you wouldn't since
FP rewrites the HTML every time you save a page and so you'd
have to revalidate and re-edit previously found invalid HTML.
 
> To be frank, most of
> my clients don't care, and you know what? Neither do I.

HTML validity is something to care about if only because it goes
a long way towards creating accessible sites.  And I believe that
if you produce valid HTML, you won't have to test against every
browser version.

I lurked on the wai-w3c list for a short while and got into a
debate with one of the members off-list.  The impression I got was
w3c-wai are mostly interested in accessibility (the ability to
view a page with any browser, including text-only browsers and
screen-readers), mostly in the context of people with disabilities
and complying with the W3C's Web Accessibility Initiative

My correspondent had the idea that if a web page was marked up
in valid HTML, it is more likely to be accessible to all browsers.

> I care how it
> *looks* -- in all the browsers I can get my hands on -- and not how it
> "complies" with some "standard."

Your quoting of a standard for HTML implies it's a figment of
someone's imagination.  There is an official HTML standard.  That
most browser manufacturers ignore most of the standard is sad but
if there weren't an HTML standard no browsers would exist at all.

> Until the people writing the standards are the people writing the browsers,
> then this is all a semantic/tree-hugging excuse to rant, as far as I'm
> concerned

Hmm, but that already happened and is why browsers now don't comply
with standards.  In order to compete for market share, both Netscape
nd Microsoft dangled a carrot in the form of extended HTML for web
developers to create browser-specific sites (which they did, in
droves).  (And now Macromedia is the company dangling the carrot.)

> The end user doesn't give a hoot as to what the code behind it looks like.

I agree, what the HTML *looks* like is only relevant to the person
who has to maintain it.  There are utilities which remove line breaks
and spaces from finished web pages to cut down the download time.
The HTML hasn't changed, just the formatting.  It's no longer
human-readable but no more or less valid because of it.

> If it works for the users, then the code is good.

I disagree.  Usability, visual design and accessibility (by way of
HTML validity) are completely orthogonal dimensions; high quality
in any one of these doesn't imply a similar quality in any of the
others.

For instance, the evolt site is a good example of putting equal
effort into all three.  The revamped alistapart site is an
example of increased accessibility but decreased usability and
visual design (although I also prefer the default font size),
whilst exclusively-Flash sites may be very usable and can be
visually stunning but they are not accessible.


Paola






More information about the thelist mailing list