Re: [office-metadata] Rough notes (I won't call them minutes just yet)

[Michael Brauer <Michael.Brauer@Sun.COM>]

Hi

Patrick, thank you very much for the notes.

Some notes are inline:

Patrick Durusau wrote:
> Greetings!
> 
> The rough notes I took this morning are attached.
> 
> Comments or suggestions are very welcome!
> 
> Unlike the US Congress, we really can't add stuff that was not said, 
> ;-), but if you think of something we need to add as being in agreement 
> or a remaining issue, post a note to the list and we will get it 
> mentioned next week.
> 
> Hope everyone is having a great week!
> 
> Patrick
> 
> 
> ------------------------------------------------------------------------
> 
> Metadata SC meeting - 31 January 2007
> 
> Roll Call:
> 
> *Bernd
> 
> Bruce
> 
> *Florian
> 
> Gary
> 
> *Patrick
> 
> *Rob
> 
> *Svante
> 
> 
> 
> Svante has volunteered to do Brend's examples in N3 syntax.
> Passed unanimously selected.
> 
> 
> Understanding to Date:
> 
> Using the RDF Model
> 
> Storing RDF/XML in a file package
> 
> May have multiple metadata files
> 
> Registering such files with the 'application/rdf+xml' 
> mediatype in the manifest
> 
> In the metadata files:
> 
> 1. properties should be encoded as elements
> 
> 2. properties should generally be wrapped in a single resource node
> 
> Use of metadata attributes in content 
> 
> Elements proposed to have metadata attributes:
> 
> tables
>     
> images
>   
> equations

Does that means formulas? If yes: Since formulas are attributes, we 
would need the table cell instead.

>     
> fields

I'm not sure about this. There are many different fields with very 
different purposes. Do we really need meta data for all fields. Or do we 
need a meta data field?
>   
> paragraphs
>     
> spans
> 
> Question: Is this a complete list? Do we need to include table:table-cell for example?

You may want to add graphical objects. I would suggest to support meta 
data at least for those elements that have alternative text or 
description attributes.


> 
> Open Questions:
> 
> 1. object literals in content (may be split)

I'm not feeling comfortable with the term "object literal". What we in 
my point of view need is the possibility to assign meta data to some, 
possibly splitted text, in the sense that the meta data says something 
about this text (for instance, who added it, whether it is important, 
and so on).

I will clarify that in a different mail.

> 
> 2. object URIs in content, where subject-object relations may be:
>     i. local-global (document to something else; citations fit this)
>     ii. global-global (talking about other stuff; John's use case)

Regarding ii: My understanding of this is that we have an RDF triple 
where the subject URI does not reference a text literal or other object 
in the content.xml, and where the object of the triple also does not 
appear in the content.xml. In that case, the triple seems not to be 
related to the content.xml at all, and I think we should better move it 
to a separate RDF-XML stream in the package.

> 
> 3. Use of both path and global IRI in a manifest file
> 
> 4. xml:id required?
> 
> 5. Preservation of all metadata? Means content not understood must be
>    preserved. 

We have to careful with this. What works is that we say that RDF-XML 
streams in the package should be preserved, and that we identify a 
couple of XML elements where we also say that meta data related 
attributes have to be preserved. What will not work is to preserve meta 
data at arbitrary elements.

Michael