Ian Hickson ha scritto:
On Sun, 30 Nov 2008, Calogero Alex Baldacchino wrote:
I'm not sure I'm understanding the whole function of the <cite> element,
and perhaps I'm bothering again with ids and references, but the
relationship between a <cite> and a quotation could be disambiguated by
coupling an id and a reference to that id.
Why is the ambiguity a problem?
Well, it depends on the uses the <cite> element is targeted to. If the
'only' purpose (and such can be enough) is to provide the semantics of a
citation in a media-independent manner and as well a stronger fashion
than a 'general purpose italic' can do, but regardless of the actual
subject taken from the cited source (which finds in the <blocquote> and
<q> elements a proper, independent semantics), the ambiguity shouldn't
be a problem: the end (human) user consuming the document should be able
to correctly relate the cited source to the quoted subject just by
extrapolating it from the surrounding prose, unless such text were
really unintelligible (but even in this case, disambiguation would be
out of the <cite> scope, with the above semantics). Otherwise, if there
were any good reason to explicitly relate the source to the subject, or
viceversa, i.e. to make it intelligible to a user agent (perhaps a bot
grouping and joining in one document all contents taken from the same
source, by parsing a series of articles? - surely there must be some
better ways to accomplish that, but perhaps such could make sense for a
somewhat purpose), then the ambiguity concern might be addressed by the
mean of a well defined relationship in terms of html semantics. I just
tried to suggest a solution to a concern I thought you and Sam Kuper
were discussing for some reason, since there is no way to correctly
define such a relationship in terms of relative positions, as you
pointed out.
Q: What problem does it solve?
Uhm... perhaps a first cite could be a complete reference, i.e. a book name
along its author, publisher and IBAN code, or a reference to another
site/blog, its author and a link to the page with the quoted text. Ok, and
now? Let's say any other reference to the same sorce could be shorter and
without the need of any markup, but for styling, while the quotation block
could remind the whole sorce to the reader, for instance, when the user moved
his pointer over it, or focuses the <blockquote>, a tooltip could present the
citation content - as if it where the content of the title attribute - and a
screen reader could speak it aloud after the quotation, and if the referred
<cite> contained a link, a click on the <blocquote> content (or any other kind
of activation) could open the linked page in another tab/window.
That describes how it could be used, but _why_? Is there an actual problem
that isn't solved today that needs solving?
Well, in fact, the above could be done as well by 'playing' with anchors
(but is it still possible to set an anchor somewhere in the document,
such as <a id="foo" />? I haven't found examples for that, perhaps I'm
missing something...). Anyway, maybe, using a <cite> element as anchor,
and a related <blockquote> or <q> as the mean to reach it, might render
a stronger semanitcs than "generic" anchors and hyperlinks (maybe
coherent for an article holding "extended" - complete - citations, as
references, in a proper section, separated from the rest of its
content), perhaps in a similar fashion as a <dfn> element is
referenceable through out its id by an <a> element.
Best regards,
Alex.
--
Caselle da 1GB, trasmetti allegati fino a 3GB e in piu' IMAP, POP3 e SMTP
autenticato? GRATIS solo con Email.it http://www.email.it/f
Sponsor:
"I sensi di un'intesa perfetta" grande concorso Sheba!
* Vinci speciali premi e tanti prodotti Sheba per il tuo gatto! Partecipa ora!
*
Clicca qui: http://adv.email.it/cgi-bin/foclick.cgi?mid=8432&d=1-12
