[WEB4LIB] CSS, HTML4, and DHTML4

[Thomas Dowling]

----- Original Message -----
From: "Don Yarman" <yarmando at ghpl.org>
To: "Multiple recipients of list" <web4lib at webjunction.org>
Sent: Wednesday, February 23, 2000 12:13 PM
Subject: [WEB4LIB] CSS, HTML4, and DHTML4

> We recently contracted with a local multimedia design company to
> update the look and function of our Web site.  Based on what
> I've learned by reading the conversations on this list, I asked
> that the new site use cascading style sheets, that W3C standards
> for HTML4 be followed, and that the code be validated as well
> as run through Bobby.  When the new pages were shown to us last
> week, none of these things happened.  I immediately asked them why
> not.
>
> A week later, they've responded by interrogating me, and the
> veneer of my knowledge is about to crack under pressure.  They've
> asked:
>
> > We can make some of the changes you requested, by removing some
> > of the tag information, however, converting the pages completely
> > to CSS will take quite a bit of work. My question is, what is
> > the purpose of this? What is the issue here? Are you looking at
> > creating a set of DHTML 4.0 standards and why?
>
> My real is answer is that I think Thomas Dowling is a god, that I
> hang on his every word and will do whatever he says, but I recognize
> that this answer is unlikely to sway them.

[Thanks, Don, your check is in the mail.]

>
> My best answer is that what I've asked for is the official standard.
> CSS streamlines the code, and given my Administration's love of
> graphics-heavy sites, any streamlining I can do is good.  As long
> as we're going to try to determine the display font of our pages,
> then CSS seems to be the way to go, as it seems to be the way
> the Web is going.

Possible response #1: "I assume you'll provide free 24-hour turnaround
when we alert you to invalid markup that fails cross-browser or
cross-version support."

  Every major release of Netscape has "broken" some feature that
  authors were relying on in the previous one, usually by no longer
  honoring some snippet of invalid markup, and Netscape 5--uh, 6--
  is going to set new records in this regard.  Add to that IE
  compatibility, and whatever you want to say about other browsers
  and non-PC platforms, and you get a pretty clear idea of how far
  outside the published specs you can go.  Yes, I know that none of
  these browsers does everything in the spec, but that's a
  discussion for another day.

  [By the way, the current (March 7) issue of PC Magazine has an
  article on cellular phones with web browsers built into them.  This
  is really happening, and a lot of our sites are going to look
  pretty useless on them.]

#2: "Since only you know exactly what your markup rules are, thanks for
implicitly volunteering to maintain our pages for us indefinitely.  It's
all I can do to teach our staff the rudiments of standard HTML."

  HTML, as a standard, has a formal definition that can be included
  in training for authors and editors.  Presentational hacks are figured
  out ad hoc and can't be authoritatively documented.  And however
  awful the dreck put out by [enter an HTML editor here], it can at
  least open valid HTML.

#3: "Excuse me while our legal counsel discourses upon the likelihood that
our publicly funded library's web site will ever be required to comply
with WCAG level AA or something comparable."

  This could be a real CYA issue in the not very distant
  future, and it strikes me that publicly funded institutions
  charged with serving the public are prime candidates for that
  bellwether lawsuit just waiting to happen.  Double-A
  compliance with the Web Content Accessibility Guidelines
  includes validating HTML.

#4: "Not quite all of our users are using DSL yet.  When they complain
about having to wait for an entire page to load before they see anything,
can we just have them call you for a discussion on the advantages of using
tables for layout?"

  Old codgers like me will remember that Netscape's first major
  selling point over Mosaic was that it could render the top of a
  page while the bottom was still downloading.  This feature is
  effectively thrown away by reckless use of tables for formatting,
  especially in browsers that don't understand <COL> or <COLGROUP>
  for explicit column layout.


Actually, keep all these arrows in your quiver for the time being.  You're
the paying customer; you're asking for work that conforms to relevant
industry specifications; you're being up front about your expectations;
and you're waiting on a project they agreed to.  I'd say the onus is on
them to defend their decision to create a design that they knew was not
what you had asked for.


Thomas Dowling
OhioLINK - Ohio Library and Information Network
tdowling at ohiolink.edu