XHTML 2.0 suggestion: <caption> for <object>, <attribution> for <blockquote>

Two things on my XHTML 2.0 wishlist are extending <caption> to be a
parameter element of <object> as well as <table>, and adding a structural
tag for the attribution of a <blockquote>. Figure captions and blockquote
attributions are common parts of documents that, in my opinion, should be
marked up structurally.


In my formal XHTML 1.1 documents, I code figures with markup like:

<div class="figure">
   <object data="filename.png" type="image/png"></object>
   <p class="caption"><em>Figure 1:</em> Description.</p>
</div>

However, in browsers that cannot render the image (like the text-only lynx),
the browser presents a caption without a figure. If <object> treated
<caption> elements within it as another sort of parameter and not as
replacement text:

<object class="figure" data="filename.png" type="image/png">
   <caption><em>Figure 1:</em> Description.</caption>
</object>

The caption would be rendered beneath the object if and only if the object
itself was rendered. Figure and caption could be seamlessly eliminated when
the figure is not rendered for whatever reason (text-only user agent,
"display: none", unsupported MIME type, &c.).

If I wanted the caption rendered regardless, then I would set the
replacement markup to the same caption:

<object class="figure" data="filename.png" type="image/png">
   <caption><em>Figure 1:</em> Description.</caption>

   <em>Figure 1:</em> Description.
</object>


My blockquotes often contain attributions. Today I use CSS style sheets
format "p.attribution" by removing paragraph indent, increasing the top
margin to separate it from the text, and mark up the blockquotes like:

<blockquote>
      <p>Some historians insist that Duncanthrax was general of the Royal
Militia. One legend even suggests that Duncanthrax was a demon who assumed
human form. Another legend describes him as a former rope
salesman.</p>
      <p class="attribution"><cite class="book">The Great Underground
Empire: A History</cite>, Froboz Mumbar</p>
</blockquote>

Though <caption> could be extended to blockquotes as well, I can't think of
a reason to caption a blockquote other than as for attribution. The
identification of source sematics of attributions also has value beyond the
descriptiveness of a caption.

Agents that read the structure of documents could benefit by recognizing the
difference between quoted material and its source, or perhaps use
attributions in other ways (such as an aid to bibliography preparation).
They'd also have presentational uses; user agents that do not support style
sheets could format them with leading space, and those that do could use the
tag as a style sheet descriptor.

Received on Saturday, 10 May 2003 02:08:51 UTC