W3C home > Mailing lists > Public > public-html@w3.org > August 2009

Re: The cite and pubdate attributes

From: Ian Hickson <ian@hixie.ch>
Date: Sat, 15 Aug 2009 11:16:17 +0000 (UTC)
To: Lachlan Hunt <lachlan.hunt@lachy.id.au>, Toby A Inkster <tai@g5n.co.uk>, Joe D Williams <joedwil@earthlink.net>, Bruce Lawson <brucel@opera.com>
Cc: public-html <public-html@w3.org>
Message-ID: <Pine.LNX.4.62.0908150025430.6420@hixie.dreamhostps.com>
On Mon, 10 Aug 2009, Lachlan Hunt wrote:
>
> The cite and pubdate attributes now defined for the article and section 
> elements don't seem to be very well designed.  It's not entirely clear 
> what problem they are meant to solve, or use cases they are addressing.  

cite="" on <section> is solving the problem that Chaals doesn't know where 
the parts of his document come from.

cite="" on <article> is solving the same problem as <a rel=bookmark>.

pubdate="" on <article> is solving the problem that there's no other way 
to associate a publication date with a blog entry in HTML, in particular 
to allow for conversion to Atom.


> Specifically, the time element can be used for marking up a date and 
> time. The only benefit that pubdate currently has over reusing the time 
> element is that it is directly associated with the specific article. But 
> there are possibly other ways in which this could be achieved by finding 
> a way to associate the time element with the sectioning element.
>
> See the possible solutions we discussed briefly in IRC.
>
> http://krijnhoetmer.nl/irc-logs/whatwg/20090810#l-290

I don't see any feasible solutions proposed there. The best solution seems 
to be a class name, but people disagree that we should allow predefined 
class names.


On Mon, 10 Aug 2009, Toby A Inkster wrote:
>
> Certainly both can be replaced with a little RDFa fairly easily.
> 
> <section id="foo" xmlns:dc="http://purl.org/dc/terms/" about="#foo">
>   <h1 property="dc:title">Foo</h1>
>   <p>
>     This section was first published at
>     <time property="dc:issued">2009-01-01 12:00:00</time>
>     and is based on
>     <a rel="dc:source" href="http://example.com/data">this data</a>.
>   </p>
> </section>

That doesn't work for syndication (since the URL would no longer point to 
the <section>), so it's not a workable solution in general.


On Mon, 10 Aug 2009, Joe D Williams wrote:
>
> If I understand Toby's comment, certainly whatever ud or rd is propsed, 
> this transient conditional commentary markup and data should be handled 
> by the selected candidate technology. I think it is most important to 
> keep the actual 'spec' keystrokes at a minimum. Please don't clobber the 
> basic 'normative' text and art with a lot of 'informative' commentary, 
> or in these cases, more like 'critical' items or a running 
> judgement/target/replacements/whatever that these additions represent.
>
> I think I would be just as happy looking at a second window that just 
> includes this 'critical' (in the context of providing a sanctioned 
> informed reasoned opinion from some otherwise impartial concensus group 
> within the topic WG). Anyway, please give me the option of reading a 
> naked one Also please recognize that the spec does not always run in all 
> contributing member browsers now.

I've no idea what the above means.


On Mon, 10 Aug 2009, Bruce Lawson wrote:
> 
> For what it's worth, when redesigning my site to use HTML 5, I discarded 
> the idea of using pubdate as it seemed to me hidden metadata that also 
> duplicated the data in most blogpost's content.
> 
> The time element in the header of this article already adequately 
> expresses the information
> 
>     <article>
>         <header>
>             <h2 id="post-1915"><a href="blah">Accessibility of HTML 5
> video</a></h2>
>             <time datetime="2009-07-30">Thursday 30 July 2009</time>
>         </header>
> 	<p>Brilliantly witty, incisive prose, in a gloriously elegiac style
> reminiscent of <cite>Cider With Rosie</cite>.</p>
>     </article>

That <time> element isn't semantically linked to the article. It could 
just as easily be the author's birthdate.


I've removed cite="" from <section> and <article>, since their use cases 
are either weak or solved by other problems. I've not removed pubdate="" 
since I don't see how else to get an unambiguous date out for conversion 
to Atom without annointing a class name. I've also not done anything with 
<blockquote cite=""> at this time, since the arguments against them (not 
quoted above, but mentioned in passing in the original e-mails from which 
the above is quoted) were weak to non-existent. While <blockquote cite=""> 
is hardly a commonly used attribute, its presence does not seem to have 
done as much harm as one usually finds is caused by such features, and a 
surprising number of people who contribute to the standards process seem 
to like having the ability to cite their sources using that attribute.

I would like to drop pubdate="" also, if we can do so in some manner that 
still solves the aforementioned problem.

-- 
Ian Hickson               U+1047E                )\._.,--....,'``.    fL
http://ln.hixie.ch/       U+263A                /,   _.. \   _\  ;`._ ,.
Things that are impossible just take longer.   `._.-(,_..'--(,_..'`-.;.'
Received on Saturday, 15 August 2009 11:16:58 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Wednesday, 9 May 2012 00:16:43 GMT