W3C home > Mailing lists > Public > public-html@w3.org > August 2013

Re: updated cite definition - please review

From: Jukka K. Korpela <jukka.k.korpela@kolumbus.fi>
Date: Thu, 29 Aug 2013 11:18:01 +0300
Message-ID: <521F03B9.90405@kolumbus.fi>
To: public-html@w3.org
2013-08-29 10:40, Smylers wrote:
> Suppose I'm writing a new page which features a title of a work, and I 
> wish it to be denoted as such. I could just use <i>, but the site may 
> also (or may later) be using <i> for things that aren't titles of 
> works -- and I wish to be able to style titles independently of other 
> italic phrases.

That's a common issue and applies to many different expressions other 
than titles, like names of ships, foreign words, quotations, and 
mathematical variables. And it is not limited to styling. You might wish 
e.g. to be able to write simple JavaScript code that collects all the 
titles (or variables or whatever) in a list, sorts it, and puts the so 
generated content somewhere on the page. You might even apply the 
principle that different items need different markup so that any future 
styling or scripting is easier. But it sounds more reasonable to spend 
time on such markup finesses when the need comes - the amount of work is 
about the same, and you would then be doing something that you know to 
be useful, instead of just expecting it might.

> Obviously <i class=title> would work fine. But so would <cite>. What's 
> the advantage in picking <i class=title> over <cite>, given <cite> 
> already exists and is supported in user agents? (Note, I'm not saying 
> this would be sufficient reason for minting <cite> if it didn't 
> already exist.)

That's a good question. Similarly we can ask whether we should in any 
way discourage authors from using <cite> instead <i class=...> whenever 
they feel that to be natural - using <cite> for quotations,  for names 
of authors, for foreign words, or for whatever. It's really comparable 
to a custom tag, except that it's recognized by all browsers and 
displayed in italic by default.

My answer is twofold, actually revolving around the same basic point. 
Using <i class=...>, or <span class=...> if you don't want default 
italic, or <b class=...> if you want default bold instead, for titles 
means using the same technique as we use for almost all other types of 
expressions. Using <i class=title> is logical since we would use <i 
class=foreign>, <i class=ship>, etc. Secondly, using <i> with an 
author-selected class avoids the illusionary feeling of using "semantic" 
markup, the feeling that seems to be the basic motivation for endless 
debates about the True Meaning of <cite>.

This would help authors to realize that if they want real semantic 
markup at low level, i.e. to say something real and automatically 
processable about the *meaning* of words or expressions, they need to 
use microdata, microformats, or RDFa.

-- 
Yucca, http://www.cs.tut.fi/~jkorpela/
Received on Thursday, 29 August 2013 08:18:23 UTC

This archive was generated by hypermail 2.3.1 : Thursday, 29 October 2015 10:16:34 UTC