<i> vs. <cite> (was: Markdown-Discuss Digest ...)

[Jelks Cabaniss]

Jon Noring wrote:

> If one wants their documents to be acceptably viewable without

> supplying CSS (that is, use only the browser's default style sheet),

> then one should use <i> and <b> (or <em> and <strong>) instead of

> <span>. If one wants to add semantic meaning as to *why* something is

> highlighted, one can always add a class to the <i> and <b> tags, viz.

> 

>    <p>The title of the article is <i class="articletitle">How To

>    Properly Mark-Up XHTML</i>.</p>


Indeed.  I'd take it even further in your particular example, though (since
the `<i>` is so media-specific):

     <p>The title of the article is <cite class="articletitle">How To
     Properly Mark-Up XHTML</cite>.</p>

Perfect candidate for `<cite>`!


> On a related note, XHTML 2.0 is dumping <i> and <b> in favor of <em>

> and <strong>, respectively, which from a semantic point-of-view makes

> more sense 


Good, though I must say I wonder when -- not to mention if! -- XHTML 2.0
will ever emerge from the womb.  I'm sure you remember the precocious (and
ill-fated!) "HTML 3"... (and no, I *don't* mean "HTML 3.2", which was the
subsequent cop-out "Attempt to more or less express Netscape's behavior in a
DTD!"  :)


> (albeit does not solve all the semantic issues with regards

> to highlighted text.)


I'm not sure it's possible.  Even the [TEI] (I know you know this Jon, but
for the benefit of others who may not -- one of the most semantic tagsets
out there) has a generic `<hi rend="whatever">` (equivalent more or less to
`<span style="whatever">`) or even just `<hi>`, that, because of the TEI's
primary function of marking up *printed* texts and manuscripts, allows the
markup scribe to indicate, "I either don't know or I won't say *why* this
was highlighted; I'm just going to say that it *was* highlighted, and
describe the type of highlighting it has with the `rend` attribute."  It's a
semantic cop-out as to the *meaning* of the highlighting.  OTOH, it *does*
give "semantics" of the print (or handwriting) of the original book (or
manuscript), and that may be all that's needed for the particular local-use
scenario.  And it can further serve as a "first-stage markup" where later,
more sophisticated editors can refactor all the `<hi>`s into more
semantically appropriate elements, while still retaining the `rend=`
attributes from the earlier `<hi>`s.

In XHTML, that last step would be analogous to changing the italicized title
in your example above to a cited title, or changing italicized/bolded
whatevers to `<appropriate-element class="whatever">`...

Oh, I'm rambling; I need to get back to work. :)

[TEI]: http://www.tei-c.org

/Jelks