Previous Message
Next Message

Re: [css-d] Divergence between compliant browsers

Sent by Peter-Paul Koch on 29 January 2002 12:12


>Will designers and developers who haven't read Eric's article and the 
>related quarreling in www-style
>(http://lists.w3.org/Archives/Public/www-style/2001Mar/0169.html)
>think web standards are broken/not ready for Prime Time? Will those
>who've read Eric's article and the related quarreling in www-style
>come to the same conclusion?

They might come to a worse conclusion. In my opinion, the key lies in the 
www-style mail you linked to:

"
I'll say it again, breaking compatibility with HTML presentation when it
could have just has easily been made compatible, will do nothing more than
alienate web authors from CSS.
"

Web developers might get tired of the standards, because they behave 
irrationally (seen from a practical web development point of view) or are 
incomprehensible. Far worse than deciding standards aren't ready yet, they 
might decide the CSS standards are too complex and not necessary anyway 
(FONT tags still work, after all).

This would be a terrible setback.

>I've read the relevant www-style threads, respect the intellect and
>integrity of all concerned, and fully appreciate that they know more
>about CSS than I ever will.

What is sometimes forgotten, however, is that we know far more about 
practical web design and development than they do. We routinely use 
combinations of HTML and CSS they never have considered, and it is there 
that most bugs and problems surface.

Therefore, we, as web designers and developers, should be critical of the 
specs. If something doesn't turn out to work in practice, don't say "It's in 
the standards", instead say "Well, we'll work around it somehow".

The result is non-standard sites. Deplorable but inevitable. Standards are a 
tool, not the goal.

An example I posted about earlier: We need a reliable vertical-align, so we 
will use tables instead of CSS. This is a direct consequence of the decision 
not to incorporate a normal vertical-align in the specs.

Who's guilty? The question as such doesn't very much interest me, but let no 
one say the web developers are guilty! We are given some tools and use them 
to the best of our abilities. If the tools lack certain possibilities we 
need, it's not our fault if we turn to other tools.

The lack of practical feeling that sometimes surfaces in the specs is, in my 
opinion, the worst enemy of the standards. Sure, web developers would love 
to use standards, but they have to be comprehensible and have to allow 
roughly the same sort of styling as the traditional methods.

Long-term solution: let the style standardizers pay slightly more attention 
to the practical everyday considerations of web designers and developers.

>I also appreciate that Mozilla/Netscape
>is doing the right thing in fully complying with CSS according to the best 
>interpretation of its experts.

The browser developers might have more concern for the practical 
consequences of their decisions. When the standards are impractical, a 
browser vendor might favour a loose interpretation that will help 
developers. (In fact, such browsers are often most popular with web 
developers)

Is this good? Dunno. Is this practical? Yes.

>But the practical implications of this divergence between IE and
>Mozilla concern me per all the unanswered questions asked above.

I don't see this as something new. Netscape and Explorer have always 
diverged, and in my opinion they're closer now than they ever have been. So 
it's the same as always: check your site in all browsers.

Yes, I'd love to get rid of these compatibility problems but in the 
forseeable future they won't disappear.

>a preference for practical discussion over theory -- a preference I
>share. As a working stiff trying to move my clients closer to W3C
>compliance, I raise these issues in hopes they may generate practical 
>discussion and experience-based tips.

I have one very practical tip: don't use DOCTYPEs. As long as you don't use 
them, the browsers stay in quirk mode and are raliable, though not 
standard-compliant.

I understand that this will not be the best solution in the long term, but 
for the short term it will serve nicely.

Thinking hard about the problems you brought up and not pretending to have 
all the answers,

ppk

_________________________________________________________________
MSN Photos is the easiest way to share and print your photos: 
http://photos.msn.com/support/worldwide.aspx
Previous Message
Next Message

Possibly related: