W3C home > Mailing lists > Public > public-html@w3.org > May 2007

RE: Cleaning House

From: Lee Roberts <lee_roberts@roserockdesign.com>
Date: Wed, 2 May 2007 23:48:33 -0500
To: <www-html@w3.org>, <public-html@w3.org>
Message-ID: <008c01c78d3e$4e05cc00$ea116400$@com>

The specification _does not_ need to continue to use deprecated tags.
Standards change and old standards are grandfathered, but new standards
update and replace older standards.  HTML5 is a new standard; it should not
be a rehash of the same stuff that's been a thorn in our sides since the
beginning.   

HTML4 is a poorly written standard, but was a best attempt.  I've taught
standards-based Web design at a local technical college.  I've faced more
questions than I care to admit from many confused students.  Many of the
questions posed dealt with deprecated elements and heading tags.  Heading
tags lead many novice developers to think that they can have as high as h12
on a page.  The semantics of heading tags are confusing to the novice.  Even
_accessibility experts_ have made the error.

I've worked with fire safety, building, electrical, plumbing, and
asset/property protection codes.  Each time a new set of standards come out,
they are based upon new technologies and discovered needs.  Existing
buildings are never required to comply with the new standard until new
construction or remodeling is performed.  Depending upon how much of the
building is remodeled determines if the entire building is required to
comply with the new standard or just the newly remodeled section.

What I get out of all the chatter about deprecated tags being supported
FOREVER is somewhere along the line of people don't know how to make
standards.  This is evident in HTML4.x and XHTML1.x with deprecated tags.
If we're smart and we will set up a DOCTYPE HTML5 which people can use, then
Browser Vendors can easily support the author's desired standard set.  If
the author fails to use DOCTYPE HTML5 then Browser Vendors can take the
assumption that they need to fall back to previous standard support mode.

This falls in line with IEEE and other standards bodies.  Backward
compatibility still exists, but we are no longer required to deal with
deprecated codes.  Eventually, instead of NEVER, we will have a standard
that works.

Browser Vendors would need to build support for the new standard on top of
their support for older standards.  Yet, this does not mean we need to
continue support for deprecated tags or tags we NOW realize we don't need.
I foresee HTML6 getting rid of some of the new stuff being submitted for
consideration in the HTML5 standard.

<b> and <i> were replaced by <strong> and <em> for their semantic values
versus the presentational values of <b> and <i>.  <u> was deprecated or
shall we say replaced by CSS text-decoration.  <font> tag certainly has no
use because CSS replaced that tag.

As I stated previously, we need only one standard.  Previous HTML standard
sets had multiple standards.  For example, HTML4 Strict, HTML4 Loose, and so
forth.  There's no need for this mishmash.  Doing this again simply means
the editors must create multiple sets of standards documents.  This then
causes problems for authors and Browser Vendors alike; they're required to
determine what was really meant.

One standard with a firm, easy to understand, and yes "for dummies"
instruction set will make it easy for the new designer as well as for the
old designer.  We need to remember that we're developing the standard as a
guide for the new designer as much as for the old designer.  When we realize
this, I'm sure we can come to standard that is easy for everyone to
understand.

I don't know who WHATWG is.  As far as I know it is a group of people who
didn't like W3's processes.  As far as I understand, the people of this
group decided to go about their own path and create a new standard.  This
only causes more problems because Browser Vendors, Editor Vendors, and Web
designers can now follow two camps.  The first camp follows the WHATWG.  The
second camp follows W3.  Seems like a lot of confusion to me.

Regards,
Lee Roberts
Received on Thursday, 3 May 2007 04:49:03 UTC

This archive was generated by hypermail 2.3.1 : Monday, 29 September 2014 09:38:44 UTC