W3C home > Mailing lists > Public > public-digipub-ig@w3.org > September 2014

[METADATA] Metadata+Content (ISSUE-3)

From: Ivan Herman <ivan@w3.org>
Date: Mon, 15 Sep 2014 10:59:26 +0200
Cc: Bill Kasdorf <bkasdorf@apexcovantage.com>, W3C Digital Publishing IG <public-digipub-ig@w3.org>
Message-Id: <567397EE-E93B-49EB-BC24-D03B24F58E28@w3.org>
To: Tzviya Siegman <tsiegman@wiley.com>
Hi Tzviya,

ISSUE-3[1] says:

[[[
How does metadata travel with content? 
How does metadata live with content?
]]]

Let me try to understand...

- Metadata can be added to the content using microdata or RDFa (although microdata is for HTML only and not, e.g., to SVG). Microdata is, usually, used with schema.org vocabulary only, RDFa can be used with anything that has a proper URI for the various terms. For more complex metadata, JSON-LD may also be added in an HTML file via a special <script> element. All these are documented, specified approaches.

- It is possible to add any metadata files as part of the EPUB file as a separate file (probably a better choice for larger pieces of metadata. There are syntax choices galore. If RDF is not a prerequisite, then an XML vocabulary can also be used, no issue there.

Of course, these refer to EPUB here, I am not sure what trade publishers do in this respect for printed books; I presume they send the metadata file along to some services. The choices are wide open.

I guess, *from an OWP point of view*, I see only one problem: both RDFa and even microdata (even if in a more covert way) rely on an RDF version of the metadata vocabularies. Whilst this is not a problem for schema.org or for Dublin Core, I am not sure whether that is done for ONIX, for example, afaik only partially (Graham and I have briefly touched upon this issue at some point when we met in London). *If* there is a necessity for considering RDF vocabularies only, then of course continuing and expanding that work may become an issue. I have not checked the similar issues around other vocabularies like PRISM. Clearly, *if* such conversions are even considered, that may be considered as an OWP issue/task. But I do not know the answer to the '*if*'-s. 

I also see a relationship to ISSUE-1 (ie, the identification issue), see my separate mail on that issue's thread.

(I have not seen any reference to these issues in the interviews, but I may have missed something.)

Do I correctly understand the question you are raising? 

Ivan

P.S. That being said, I would think that this whole problem area SHOULD be listed in the metadata document we produce, spelling it out clearly.


[1] http://www.w3.org/dpub/IG/track/issues/3

----
Ivan Herman, W3C 
Digital Publishing Activity Lead
Home: http://www.w3.org/People/Ivan/
mobile: +31-641044153
GPG: 0x343F1A3D
WebID: http://www.ivan-herman.net/foaf#me






Received on Monday, 15 September 2014 09:00:16 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 25 April 2017 10:44:20 UTC