W3C home > Mailing lists > Public > public-rdf-in-xhtml-tf@w3.org > July 2008

Re: RDFa Primer comment

From: Shane McCarron <shane@aptest.com>
Date: Thu, 17 Jul 2008 09:44:22 -0500
Message-ID: <487F5AC6.40300@aptest.com>
To: Ben Adida <ben@adida.net>
CC: Carlos Tejo Alonso <carlos.tejo@fundacionctic.org>, "public-rdf-in-xhtml-tf.w3.org" <public-rdf-in-xhtml-tf@w3.org>

Let me attempt to clear this up, since I started it...

RDFa in XHTML: Syntax and Processing (http://www.w3.org/TR/rdfa-syntax) 
defines a markup language called XHTML+RDFa.  This is an XML markup 
language, and you can use XML authoring tools to create documents that 
use it and are valid.

What does that mean in the real world?  Most people who are developing 
real content (as opposed to some random blog somewhere) are concerned 
that content be "valid".  Valid content is content that "validates" 
using a validation tool such as the ones built into popular commercial 
editing tools or web-based ones such as http://validator.3.org  So, if 
you care about validation and are creating complete web pages, you need 
to write your content in a markup language that permits the constructs 
you want to use.  In this case, XHTML+RDFa as defined in the XML DTD at 
http://www.w3.org/MarkUp/DTD/xhtml-rdfa-1.dtd

Once you have valid content, you probably want that content to go to 
every user agent out there.  And that's fine.  Even though user agents 
don't explicitly "speak" this new markup language, they are perfectly 
happy to process it.  That's because the XHTML Family of markup 
languages defines processing rules that ensure such compatibility, and 
also provides guidelines for writing content that will even work on 
non-XML user agents such as Internet Explorer (see 
http://www.w3.org/MarkUp/drafts#xhtmlmime for a link to the latest version).

If you are only creating fragments of web pages, and you don't know 
where they will be incorporated (e.g. a blogging environment) or you 
don't care about validation... then RDFa in XHTML: Syntax and Processing 
defines a collection of attributes and what those mean when parsed.  If 
you use them in the defined manner they are likely to work anywhere you 
would care to use them - user agents won't care - we designed it that 
way.  Unless you define your own markup language, the content will not 
"validate".  But if you don't care, apparently some parsers will not 
care either.

Obviously, my bias is that you create valid markup.  It is more likely 
to be portable, and it is more likely to be parsed correctly.  The "tag 
soup" days of the 90s are going the way of the Dodo - and that's a very 
good thing.



Ben Adida wrote:
>
> Carlos Tejo Alonso wrote:
>> So, where is possible to add RDFa flavour (XML, HTML, XHTML...)?
>
> Carlos,
>
> Currently, we've specified RDFa for XHTML1.1, but we've designed it to 
> make it easy to add to HTML, too. As for generic XML, Yahoo is already 
> doing XML+RDFa with their DataRSS project.
>
> In other words, while the specification is currently limited to XHTML 
> 1.1, don't let that stop you from pushing the technology forward, 
> especially given that most of the parsers already work with non-X HTML.
>
> -Ben

-- 
Shane P. McCarron                          Phone: +1 763 786-8160 x120
Managing Director                            Fax: +1 763 786-8180
ApTest Minnesota                            Inet: shane@aptest.com
Received on Thursday, 17 July 2008 14:45:15 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Thursday, 17 July 2008 14:45:16 GMT