Offtopic: Web Standards Project

[Eric Eldred]

Tim Bray wrote:
> 
> At 12:02 PM 8/8/98 -0500, W. Eliot Kimber wrote:
> >So I'm wondering what the true motivation of the WSP is: ....
> One interesting
> discussion is, which standards to focus on... my personal bet would
> be XML/CSS/DOM, because the implementations are just happening.  Is
> it worthwhile, at this point in history, trying to retroactively
> save HTML?  Real question. -Tim

I'm not sure what this has to do with xml-dev, other
than the people, but I'll continue the thread here.

I have no opinion about motives.  I'm just a little guy,
who would love to find ways to publish faster both to
the web and to paper.  XML certainly appeals to me, so
any encouragement that way would help.  I see several
directions that would be important to pursue:

1.  Talk to the *HTML editor* people.  They are responsible
for foisting a lot of errors on us--I don't care whether
they are validated or not--they often produce such
garbage (Front Page, Cold Fusion) that I can't read them
with Lynx at all.  The lack of a good, free XML editor is
shameful (don't ask me to use emacs).

2.  I use vi and some Linux tools, but one reason I do 
is that the *filters* from Blueberry and Filtrix included 
in other commercial products are useless.  I'm not
talking about complicated things, just elementary problems
such as curly quotes and em dashes--at some point between
OCR and a web page one needs to edit them by hand.  There
needs to be some simple rule-based filter such as
Omnimark, but cheap or free.  I can write some by hand
in perl, but most people can't.  But dealing with the
white space problems is not trivial.

3.  Let's face it:  Netscape and Microsoft might be
enemies, and they might have almost all of the browser
market between them, but it will be neither necessary
nor sufficient to have them agree on web standards.  Because:

3a. For example, CSS1 provides for text (left and right)
justification.  Netscape 4's justification is force
justification for the last paragraph line (thus often
creating huge wordspacing on it) and Internet Explorer
4's is not force.  I don't believe the standard specifies
one or the other behavior.  In each case, the author has
no control.  Much of the browser behavior that is
consistent between vendors is de facto standardization,
not written down anywhere.

3b. As the very example of the web pages at the
webstandards.org site shows, there will always be a
need for backward compatibility.  Unfortunately, it is
the history of web browsers that has caused all these
problems, and even when version 5 browsers are 
released, we still have to design our pages to account
for earlier ones, and text-only browsers such as Lynx.  
I can't see that webstandards.org can do much about that.
(Well, we can stop using those features, but we don't
need an organization to tell us to stop.)

4. Given that there will for a long time be used web
browsers that can't support XML directly, why doesn't
it make the most sense to do this at the server end?
Keep pages in validated XML and filter them on the
fly to HTML (whatever level your clients support) or
even PDF?  I realize there are problems with that--
who knows what combination of source will confuse or
crash this or that browser--but at least for some time
users can continue to employ the web browsers they
prefer.

5.  Just what level of CSS is webstandards.org going
to mandate that Netscape and Microsoft support fully
in their version 5 browsers?  Just what level of XSL?
I guess you can see there might be development
schedule considerations.

6.  CSS and XML are only a few of the W3C standards.
I'd like to see some more encouragement for full
support of the Accessibility standards (and some
feedback as to how to improve them), the PICS
effort, and especially the RDF effort--there needs
to be an easier way to use metadata and more tools
that use them, such as Dublin Core and others.

7.  Seems to me that Microsoft and others are mainly
interested in XML for practical, proprietary reasons:
to support their CDF "push" technology, "Chrome"
to justify buying a 350MHz Pentium II and get Microsoft
into television markets, EDI replacements as well as
on-line banking, etc.  Although these efforts are
based on a standard XML, the actual implementations
are likely to be locked up and the otherwise "openness"
of XML is lost.  Standards agreements go only so far.

To sum up:  webstandards.org is a good idea and
deserves support.  Netscape and Microsoft need to
be kept to their promises.  We need to address
more closely the significance of de facto standarization.
However, we can disclose and work on some problems
without having to put pressure on those big companies
to do it for us.  We can support the mozilla effort
and work on XML/XSL etc tools that can help everybody.

-- 
"Eric"    Eric Eldred      Eldritch Press
mailto:EricEldred at usa.net  http://eldred.ne.mediaone.net/

xml-dev: A list for W3C XML Developers. To post, mailto:xml-dev at ic.ac.uk
Archived as: http://www.lists.ic.ac.uk/hypermail/xml-dev/
To (un)subscribe, mailto:majordomo at ic.ac.uk the following message;
(un)subscribe xml-dev
To subscribe to the digests, mailto:majordomo at ic.ac.uk the following message;
subscribe xml-dev-digest
List coordinator, Henry Rzepa (mailto:rzepa at ic.ac.uk)