Web Technologies in a Tailspin (was Re: Uncle Dave ponders the meaning of namespaces)

Date view Thread view Subject view Author view

From: B.K. DeLong (bkdelong@pobox.com)
Date: Fri Aug 11 2000 - 08:55:33 PDT

At 07:36 AM 08/11/2000 -0700, Dave Winer wrote:
>I fear it's hopeless. XML, as promoted by the W3C, is not for people who
>love the Web. We continue to have this argument. It's a total waste of time.
>Try a mental exercise. RDFize RSS. Add namespaces. Let the schema people
>have their way. What once was something that any HTML coder and Perl
>scripter could understand becomes a bookshelf of arcane specs and minutiae.

I think you're totally on the mark Dave. From a Web developer's standpoint
there is going to be an incredible shift in how they need to see the Web.
The only W3C standards I see developers of public Web sites using is XML
1.0, XHTML, XSL/XSLT, and XForms. (and possibly the various forms of XHTML
1.1, Modularization etc). I'm talking about people who create the Web pages
for Disney, Universal Studios, ZOT Group etc.

The problem is that HTML got bastardized from being a core markup language
to something being used strictly to layout text and graphics for a page.
CSS wasn't pushed hard enough when it first came out and you even had the
W3C slipping things like <B> and <BIG> into the mix. Frankly, I'm really
surprised that ALL the style elements haven't been deprecated in HTML 4.01
because the only way to strongly bring all the Web developers from the HTML
mindset to XML is by forcing them to use markup for markup and CSS or XSL
for styling.

I was ELATED yesterday when I was giving someone examples of how CSS and
XML were supposed to work and came across this example (
http://metalab.unc.edu/xml/examples/books1.xml) which worked as it should
in IE5.5 and that's sad. All Web browsers should have been doing the same
within 6 months of CSS 1.0 coming out.

This corruption of "markup" in the HTML sense continues with book authors
writing about content positioning workarounds using tables since no one has
really done a good job of implement positioning from CSS2. People are
groaning about being required to make their Web sites accessible when this
would have been a non-issue if stylesheets were usable cross-browser. Web
developers are having to do entire sites in HTML and then again in WML to
make their sites accessible on WAP-enabled wireless devices when authoring
them once in XHTML with a standard CSS/XSL style for wireless phones would
have sufficed.

Don't get me wrong....RDF with Dublin Core metadata will continue to be
extremely valuable to archivists, librarians and large corporations who
have vast libraries of data that need referencing. Standards like Schemas,
namespaces, and XML-RPC/SOAP will be great for complex B2B applications,
use in binary programs (not your everyday Web site) and projects like the
one my wife is working on creating a system for marking up data for an
occasionally-connected battlefield network using SINCGAR radios.

In my opinion, there are more people working on standard Web sites
(including personal sites) then the complex B2B and special applications of
Web technology that I mentioned and I'd like to see not only the core Web
technologies concentrated on and completed, but more encouragement for user
agents to fully implemented standards before sending product-to-market.
Groups like the Web Standards Project and people like you, Dave, are doing
all they can to make this happen but there needs to be a stronger industry
entity with more clout that will come out and say that MUST happen and
within the next year.

My fear is that if we continue with the standards adoption rate from the
past, technologies like XForms for instance will take 10 years before Web
developers can actually use them if Web browsers continue to put
product-to-market in front of implementing standards.

I am fully aware of how product development cycles work and that what I am
suggesting would take a lot of work and sacrifices however if we go back to
the main point of this rant the only way to get Web developers interested
in all these new standards is to put them into a new realm of thinking when
it comes to Web design - markup and style as two separate entities. That
can't happen when the software Web users utilize to view Web sites doesn't
employ the standards we want Web developers to conform to. And since the
only organize actively pushing for this to happen don't seem to have enough
clout for browser companies to listen, I don't see how this can happen.

B.K. DeLong
Research Lead
ZOT Group

617.542.5335 ext. 204 bkdelong@zotgroup.com http://www.zotgroup.com

Date view Thread view Subject view Author view

This archive was generated by hypermail 2b29 : Fri Aug 11 2000 - 09:00:11 PDT