Re: Last call comments on WOFF (4)

On Feb 9, 2011, at 15:45, Chris Lilley wrote:

> On Wednesday, January 12, 2011, 4:16:26 PM, Bert wrote:
> 
> BB> 4) Section 6 Extended metadata: If it is in XML and is metadata, it  
> BB> would seem logical to have chosen XMP. Existing XMP and RDF tools  
> BB> would be able to read it, no need for new parsers; it could be linked
> BB> to other RDF ontologies, to enable Semantic Web tools to make  
> BB> inferences; and it would be extensible without the need to have  
> BB> different syntaxes for predefined and extended elements.
> 
> We aren't sure exactly what you are asking for.

I'm just trying to reduce the number of different formats on the Web that do more or less the same thing. More formats means more effort, more lines of code, and more bugs.

> Are you asking that the entire metadata section be replaced by a block written in the Adobe XMP language? If so, are there examples where XMP has been used to express licensing information 9for fonts or for other media, such as images)?

XMP is relatively easy to parse and convert into other things, it is widely used (I thought TeX would be too old for it, but just yesterday I discovered that modern versions include XMP in the generated PDF) and it is compatible with software that reads RDF. Creative Commons recommends it for many formats that do not natively have metadata.

There are various vocabularies available that could be useful for WOFF. Specifically for license info, there is the simple but ubiquitous "rights" property from Dublin Core, but also Adobe's rights management vocabulary.

But my comment wasn't specifically about license data. What caught my eye was the "extension" element.

Allowing user-defined extensions is a problem that many working groups have tried to solve without success, but RDF is probably the only somewhat workable solution (because it (1) restricts the syntax to only allow things that express triples and (2) more or less forces extensions to include a link to their definitions). The extension element in WOFF is one of the bad examples. You can only guess at what its content means or where you can find the documentation.

> Are you asking that the entire metadata section be replaced by an entirely open-ended 'anything you want as long as its RDF' block, and if so how could that be interpreted in there is no predefined vocabulary at all?

RDF (I recommend XMP, specifically) allows everything the current extension element allows and more, but additionally links each extension to its definition. It also allows private vocabularies to be redefined in terms of standard ones at a later date, if needed, and without changing any existing files that use the vocabularies.

So, concretely, I'm suggesting that WOFF either allows no user-defined extensions, or, if there is a good reason to allow them, to use XMP for that.

If you do provide space in WOFF for XMP, then it might make sense to put the predefined metadata (licensee, description, vendor, etc.) in XMP as well. But I haven't sufficiently thought about that.

> Tracker, this relates to ACTION-71: Clarify Bert Bos' Point 4



Bert
-- 
  Bert Bos                                ( W 3 C ) http://www.w3.org/
  http://www.w3.org/people/bos                               W3C/ERCIM
  bert@w3.org                             2004 Rt des Lucioles / BP 93
  +33 (0)4 92 38 76 92            06902 Sophia Antipolis Cedex, France

Received on Friday, 25 March 2011 23:08:21 UTC