W3C home > Mailing lists > Public > www-html@w3.org > March 2006

RE: XHTML 2.0 - dfn : Content model and usability (PR#7832)

From: Rob Simpson <Rob.Simpson@BlackBox.com>
Date: Fri, 24 Mar 2006 09:52:34 -0500
Message-ID: <C070BEDCDF191F42A7BB08BF8C70631743EFB7@SATURN.corp.bbns.com>
To: "'www-html@w3.org'" <www-html@w3.org>

Your argument that <dfn> is better than the legacy <dl> makes the optimistic
assumption that <dfn> will be properly used in most cases.  

I strongly agree with your argument that <dfn> is far too primitive.  It
sounds to me like it's best use is for HTML authors trying to spam search
engines by marking keywords, regardless of whether or not there is an
associated definition, basicly saying "make my page more important because I
claim it _defines_ such-and-such".

So <dfn> is more likely to be purposely misused, while <dl> is "misused" for
layout simply for historical reasons.  IMO, I would rather see only the
latter.  Therefore, I think it would be useful to add the tags to the list
module that are needed for the purpose for which <dl> is often "misused",
that is:

     <il> <!-- an indented list -->
          <li>first tab position <!-- like "dt", but not a term being
               <il> <!-- next level of indentation, nested under parent <li>
                  <li>second tab position <!-- like "dd", but not a
                          could be on next line (style="display: block") or
                          same like (style="display: inline) like "dl

There are enough other changes that are needed to convert an XHTML 1 page to
XHTML 2; changing all non-definition <dl>s to <il>s will just be one more
bullet point on the web pages that describe how to do that.  (Well, maybe
two, including removal of the deprecated "compact" layout attributes from
"<dl compact>"s.)

Since XML Schema allows for the same element name under different parents
(with either the same or different content models), you could even do this,
creating a definition of a single term using the existing tags:

     <dfn>The <dt>tiger<dt>, <dd><jk:taxon>Panthera tigris</jk:taxon></dd>,
is <dd>a large, striped Asian felid</dd>.</dfn>


-----Original Message-----
From: Jukka K. Korpela [mailto:jkorpela@cs.tut.fi]
Sent: Friday, March 24, 2006 3:25 AM
To: 'www-html@w3.org'
Subject: Re: XHTML 2.0 - dfn : Content model and usability (PR#7832)

On Thu, 23 Mar 2006, Rob Simpson wrote:

> Regarding the "<dfn/>" element, why is yet another "definition of term"
> element needed in HTML?

It's a holdover from HTML 3.2. It isn't even legacy markup, since it has 
been used very little; it exists in theory only.

> Wouldn't "<dl><dt
> id="def-acronym">acronym</dt></dl> the definition" suffice,

For what? What is the purpose of using definition markup? The <dl> has 
largely become a poor man's layout tool (for certain types of lists) 
rather than structural markup. Continuity with previous versions of HTML 
would be a good reason _not_ to use the same element name if you wish to 
specify an element that _really_ means a list of definitions. You don't 
want to send the message "hey, this is good old <dl> that the W3C has 
always told us to use for lists of stuff with indented description below 
each major item".

Technically, <dfn> is a standalone inline element, so it is syntactically 
rather different from <dt>.

Both <dfn> and <dl> would be worse than useless in XHTML 2.0 especially 
since they are have essentially been copied from HTML 3.2 via HTML. It's 
illogical to have markup (nominally) for a definition list without having 
markup for a definition. A single-definition <dl> would be an artificial 
approach to giving an isolated definition. Moreover, it would apply only 
to a small subset of definitions: those that are expressed as separated
into a definiendum and a definiens. The <dfn> markup is far too primitive, 
since it only says "this is a defining occurrence" (essentially, it's 
just a kind of highlighting) without specifying what constitutes the 

If you want definition markup, you need a good analysis of what 
definitions are - a  definition of "definition", really (see 
http://www.cs.tut.fi/~jkorpela/def.html for an attempt) _and_ some idea of 
how definition markup could be made popular among authors _and_ some 
support to such markup in popular software (especially browsers and 
search engines). This is really a chicken and egg problem.

Jukka "Yucca" Korpela, http://www.cs.tut.fi/~jkorpela/

This email and any files transmitted with it are confidential and are
intended for the sole use of the individual to whom they are addressed.
Black Box Corporation reserves the right to scan all e-mail traffic for
restricted content and to monitor all e-mail in general. If you are not the
intended recipient or you have received this email in error, any use,
dissemination or forwarding of this email is strictly prohibited. If you
have received this email in error, please notify the sender by replying to
this email.
Received on Friday, 24 March 2006 14:52:47 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 27 March 2012 18:16:05 GMT