[thelist] Standards advice

Ben Henick persist1 at io.com
Sat Jan 4 13:05:01 CST 2003


On Fri, 3 Jan 2003 howcheng at ix.netcom.com wrote:

> You have it exactly. Large commercial sites need their content to appear
> essentially the same on as many browsers as possible, thus facilitating the
> "lowest common denominator" approach.

[I answer the questions posed, at the end of my rant, BTW.]

Oh, pfui.

Not disagreeing with your statement, rather fighting the urge to rant
against the dynamics that have resulted in the fact that that's how Things
Are Done.

Anecdotally -

If it's good enough for Wired News, it should be good enough for anyone.

I will be the first person to stand up and defend the need for consistent
presentation in as many browsers as are practicable...

But at this stage, given that 5.x+ browsers have a market share generally
understood to be in excess of 85%...

It becomes painfully obvious that the continued reliance on table-based
design (and all of the still-worse evils concomitant to that) is more the
consequence of ossified policies and complacency on the part of line
folks, more than anything else.

My hope is that eventually Some Big Huge Corporation will experience a
Section 508 bitchslap at the hand of a judge, which seems to be the only
event likely at this point to get things moving with any rapidity down a
different course.


Meanwhile... one of the biggest incentives for not migrating to better
standards support, is that these same companies have huge investments in
systems and infrastructure meant to support standards of the pre-bomb era.

Anyone who's worked with earlier versions of Microsoft Site Server (to
name one example) are probably at this point nodding enthusastically.

Between politics and cost, the consequences should be most obvious!

At a smaller-company level, one also notes that WYSIWYG support for
standards-based production, while improving, still has plenty of room for
growth.

As for the practical considerations of What To Do From Here -

Bear in mind that valid XHTML validates as XML as well.  At least the
general idea.  If that's an intermediary step likely to reduce your future
workload, go for it.  Otherwise, don't worry if you think the efforts
gonna put you into knots (which I doubt it would).

CSS is a different matter.  It's what allows Web markup to be semantic -
that is to say, to fulfill the role for which it was designed.  Period,
end of sentence.

(I just woke up and haven't had my coffee, so I'll leave substantiation of
that assertion aside for the moment in the hope that someone else will
take it up.)

As for (bleah) Netscape 4...

You may find JSSS worthy of investigation, as it's the API through which
Netscape 4 was DESIGNED to support styles.  The best documentation of that
is at developer.netscape.com.

When everything is said and done, attempts to support Netscape 4 with
fully standards-compliant code and markup will fail beyond a certain
point.  You need to decide what happens when that line is crossed.

...But in any event, if you rely on markup to express structure and
semantics - and turn to CSS for nitty-gritty presentation - you can't go
wrong.  ;-)


--
Ben Henick                     "In the long run, men hit only what they aim
Web Author At-Large             at.  Therefore, though they should fail
http://www.io.com/persist1/     immediately, they had better aim high."
persist1 at io.com                 -- Henry David Thoreau






More information about the thelist mailing list