Re: [MACCAWS] Webmonkey article: Web Standards for Hard Times

> > On 2002-08-07 11:58 AM, <[email protected]> wrote:
> >
> > > He has a very good point here. What I'm asking is: Should we be
> > > pushing workarounds/hacks to work WITH the older browsers, or settle
> > > with (possibly) less visually appealing & precise layouts which ARE
> > > cleanly laid out but still remain W3C compliant.
>
> We need to clarify what we're talking about when we refer to 'hacks'. I
> can't think of a single commonly-discussed hack that isn't valid CSS,
> and if it's valid, it meets the standard.
>

This is from glish (below).  Please note that is says "a CSS parsing *bug*"
[my emphasis].  This is not a standard, and is not taking advantage of
"standards", and as time go on there will be more and more to relieve the
strange bugs as browser versions and subversions are released.  It's the
anologous situation to what got us here in the first place.

The real issue here is not necessarily standards, but the advantages that
these new technologies bring, specifically, that they 1) seperate content
from code cleanly 2) create markup that renders flexibly and 3) create xml
compliant documents that are easily translated to other forms (in theory,
have *you* tried to create an xhtml parser recently? ;-), and 4) keep web
developers employed [again, my emphasis ;-)].

In so far as we see the adoption of standards as bringing these advantages,
then they're significant advatanges.  But in some ways, we are trading one
for another.  I'm interested to see the working docs for xhtml 2.0, which
are due any day...

My best to you all,

Jonathan Dillon

 #centercontent {
  background:#fff;
     margin-right:199px;
  /*
  IE5x PC mis-implements the box model. Because of that we sometimes have
  to perform a little CSS trickery to get pixel-perfect display across
browsers.
  The following bit of code was proposed by Tantek Celik, and it preys upon
a CSS
  parsing bug in IE5x PC that will prematurly close a style rule when it
runs
  into the string "\"}\"". After that string appears in a rule, then, we can
override
  previously set attribute values and only browsers without the parse bug
will
  recognize the new values. So any of the name-value pairs above this
comment
  that we need to override for browsers with correct box-model
implementations
  will be listed below.

  We use the voice-family property because it is likely to be used very
infrequently,
  and where it is used it will be set on the body tag. So the second
voice-family value
  of "inherit" will override our bogus "\"}\"" value and allow the proper
value to
  cascade down from the body tag.

  The style rule immediately following this rule offers another chance for
CSS2
  aware browsers to pick up the values meant for correct box-model
implementations.
  It uses a CSS2 selector that will be ignored by IE5x PC.

  Read more at http://www.glish.com/css/hacks.asp
  */

  voice-family: "\"}\"";
  voice-family: inherit;
     margin-right:201px;
  }