- From: Toby Inkster <tai@g5n.co.uk>
- Date: Mon, 15 Nov 2010 10:56:20 +0000
- To: Ivan Herman <ivan@w3.org>
- Cc: RDFa Working Group <public-rdfa-wg@w3.org>
On Mon, 2010-11-15 at 11:18 +0100, Ivan Herman wrote:
> From a technical point of view, I must admit I feel a bit uneasy by
> the fact that the your document requires changes on the processing
> model.
I assume you're talking about inferring typeof="" on <entry> elements.
(Actually the Wiki also says to do this for <feed> - it shouldn't - I'll
fix that.)
It's not really my document that requires it - it's the existing Atom
1.0 syntax and semantics. Given:
<feed xmlns="http://www.w3.org/2005/Atom">
<entry>
<link rel="self" href="page1" />
<link rel="next" href="page2" />
</entry>
<entry>
<link rel="self" href="page2" />
<link rel="previous" href="page1" />
</entry>
</feed>
It is the entries that are the subjects of those links, not the feed
itself. If we don't set a new subject for each entry, then we'll just
get junk data back parsing existing Atom feeds.
Yes, this does require coding rather than configuration, but if you
wanted to abstract this kind of processing away from your parsing code
you could add typeof="" attributes via an XSLT preprocess. (And you
could do the same for about="" on <head> and <body> in HTML.) You'd have
a single XSLT preprocessing file for each host language much like you'd
have a single default profile.
PS: if you dislike Atom+RDFa, wait until you start trying to parse
OpenDocument's RDFa. :-)
--
Toby A Inkster
<mailto:mail@tobyinkster.co.uk>
<http://tobyinkster.co.uk>
Received on Monday, 15 November 2010 11:00:09 UTC