Re: [metadata] Who will consume our metadata?

On 04 Feb 2014, at 17:24 , AUDRAIN LUC <LAUDRAIN@hachette-livre.fr> wrote:

> Book discovery is the first and great use of metadata. ONIX for Books is completely appropriate for that and as publishers we contribute to make it more better with our EDItEUR membership.
>  
> But this level doesn’t belong here to the Open Web Platform.
> What belongs to the OWP, is the question of readers consuming content with semantics. And here is the first problem when we have to use a dumb language as HTML5. We have to help HTML5 speak to bring our ebooks intelligence and our readers knowledge.

Luc,

I think we have to separate two notions.

- When I mark an element in HTML5, say, a section element, as being a 'chapter' or an 'introduction', that attaches some semantics to that specific part of the book. For me that is similar to what the Behavioral UI Task Force is looking at; today, this is achieved by @epub:type, in future I would argue that it should be handled by something like @ebook-type; it needs a set of terms which are very specific to the book structures. Is this what you mean when you say 'help HTML5 speak to bring our ebooks intelligence'?

- When I talk about the author, the editor, the price, the description, but even the access rights, etc, of a book, that is metadata that is on a different level than HTML5. The same metadata terms that are used for these can be attached, in theory, to a book fully in SVG, for example. That is the type of metadata that external tools, services, data centers, catalogues, etc, would use.

At least to my mind, the metadata task force is more about the second, what was also called 'structural semantics' is more about the first. And I think they are different in syntax, in goals, in terminology, etc.

Does this make sense, or am I completely wrong?

Ivan


> Because ebooks are not only novels but all sorts of structured content from dictionaries to cooking books, from tabletop books to travel guides. And in that perspective, we are just at dawn from what will be consuming content in the future, even from the titles we today distribute mainly on paper.
>  
> Luc
>  
> De : Cramer, Dave [mailto:Dave.Cramer@hbgusa.com] 
> Envoyé : mardi 4 février 2014 16:29
> À : public-digipub-ig@w3.org
> Objet : [metadata] Who will consume our metadata?
>  
> For metadata to accomplish something, it needs to be both created and consumed. For us book publishers, ONIX is a good example of a very successful metadata standard. The people who make books create ONIX records that describe the books. We then send these records to the retailers, who use this information to populate their web pages. Consumers then read about the books, and (we hope) buy them. BISAC is another similar example—bookstores wanted to know which shelves to put the books on. Publishers, who presumably have read their own books and know what they are about, assigned codes to tell the bookstores what they needed to know. 
>  
> Who are the consumers of all the other metadata we're talking about? Many of us publishers have already implemented some kinds of semantic data, like putting epub:type="chapter" in our ebook content. But is anyone doing anything with that information? We also want metadata to drive the discovery of our books. How would that work when most of our content is not exposed to the web (due to file formats, DRM, or the need for payment)? Who is listening, besides Google's spiders?
>  
> Dave
> 
> 
> :: :: ::
> 
> 
> Dave Cramer | Content Workflow Specialist | Hachette Book Group | 237 Park Avenue NY | NY 10017 | 917 207 7927 | dave.cramer@hbgusa.com
>  
>  
> This may contain confidential material. If you are not an intended recipient, please notify the sender, delete immediately, and understand that no disclosure or reliance on the information herein is permitted. Hachette Book Group may monitor email to and from our network.


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