Re: HTML+RDFa intro: examples before history, please

On 04/22/2010 11:56 AM, Paul Cotton wrote:
> I'm curious to see how the new Facebook use of RDFa
> compares to the HTML+RDFa draft
> http://www.w3.org/TR/2010/WD-rdfa-in-html-20100304/

Facebook's use of RDFa is compatible with both RDFa 1.0 and RDFa 1.1. A
conformant RDFa Processor would extract a valid set of triples from
Facebook's Open Graph Protocol markup.

While some of the data modeling experts are concerned about the way the
data is modeled, the RDFa community views Facebook's announcement as a
very positive event for the semantic web movement.

As stated in the previous e-mail[4], the next HTML+RDFa heartbeat will
be updated to refer to the RDFa Core 1.1 document instead of the RDFa
1.0 document. Ivan has written a very good post on some of the newer
features that will be supported in the next HTML+RDFa heartbeat:

http://ivan-herman.name/2010/04/22/rdfa-1-1-drafts/

> I think I'm not the only reader who skips past all
> the blah blah blah text and looks for the first
> example that I can copy and paste and use.

For casual readers, the RDFa Primer[1] provides a number of simple
examples and is where we would expect those that are unfamiliar with
RDFa to start.

The bulk of the RDFa examples and rules are in the RDFa Core 1.1[2]
document. All language-specific implementations are placed into "thin"
specifications - for example, XHTML+RDFa[3] contains only the rules that
are specific to XHTML. We needed to do this to reduce and eliminate
duplication of text between specifications - that being a priority
because when you repeat the same things in different specs the language
almost inevitably gets out of sync and causes interoperability issues.

With that editing criteria, it is doubtful that we will provide any
examples that exist in the RDFa Primer and the RDFa Core document in the
HTML+RDFa document or the XHTML+RDFa document.

While this does create readability issues, we feel that the trade-off is
worth ensuring that text is not duplicated between specifications.

> The whole 1.1 History section is just more
> blah blah to skip over, from that perspective.
> I strongly suggest moving it to an Acknowledgements
> section at the end.
> By way of example, see
> Acknowledgements and Change History
> in the GRDDL spec
> http://www.w3.org/TR/grddl/#changes

Good suggestion, I will make that change in the next version of the
HTML+RDFa document.

> And please add lots of examples, especially
> a hello-world example in the intro.

Are the examples in the RDFa Primer and the RDFa Core 1.1 document
sufficient? Or would you rather every document (RDFa Core, XHTML+RDFa,
HTML+RDFa, SVG+RDFa, ODF+RDFa, etc.) contains a basic examples section?

> In fact, I hope that over time, the HTML + RDFa spec
> becomes less of a diff from [XHTML+RDFa] and
> more of a self-standing spec.

The current plan is that RDFa Core will contain the bulk of the
language-agnostic examples and processing rules, and that thin
specifications will align RDFa Core to each implementation language
(XHTML, HTML, SVG, ODF, etc.). If one were to visualize this as a
standard "spec stack", it would look like this:

+------------+-----------+----------+----------+
| XHTML+RDFa | HTML+RDFa | SVG+RDFa | ODF+RDFa |
+------------+-----------+----------+----------+
|                 RDFa Core 1.1                |
+----------------------------------------------+

Does this e-mail help explain some of the design decisions that the RDFa
Working Group and HTML WG have assumed while authoring these
specifications? Can you live with these design decisions?

-- manu

[1] http://www.w3.org/TR/xhtml-rdfa-primer/
[2] http://www.w3.org/TR/rdfa-core/
[3] http://www.w3.org/TR/xhtml-rdfa/
[4]
http://lists.w3.org/Archives/Public/public-html-comments/2010Apr/0066.html

-- 
Manu Sporny (skype: msporny, twitter: manusporny)
President/CEO - Digital Bazaar, Inc.
blog: PaySwarming Goes Open Source
http://blog.digitalbazaar.com/2010/02/01/bitmunk-payswarming/

Received on Friday, 23 April 2010 15:52:05 UTC