OASIS Mailing List ArchivesView the OASIS mailing list archive below
or browse/search using MarkMail.

 


Help: OASIS Mailing Lists Help | MarkMail Help

office-metadata message

[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]


Subject: Re: [office] Re: [office-metadata] Re: [office] Suggested ODF1.2 items



On Dec 12, 2006, at 8:32 AM, Svante Schubert wrote:

> I don't like approach of start & end flags. I will expand the example 
> of the solution I prefer to the following:
>
> <p><span/><span meta:class="foo"/><span 
> meta:class="foo">blahblah</span></p>
> <p><span meta:class="foo">something</span><span 
> meta:class="foo"><span/></p>

I think this is what Patrick and I had in mind, in the sense that this 
particular case could be achieved in other ways.

However, from the model perspective, you'd have to distinguish the 
subject and the predicate. The meta:class attribute would just be 
assigning a type to a subject, but the subject here would be unclear. 
Maybe meta:about would be more appropriate?

> All elements with @meta:class="foo" are of the same semantic and 
> define an area, which is not related to the XML hierarchy.
> In this special case - representing an alternative to a start & end 
> tag - this area is a sequence of content, but it might be as well 
> clustered all over the content.xml.

The one problem is that I'm not really sure I'd call this an "area" 
formally.

For example, let's say we have a list:

<text:list meta:about="event-1">
   <text:li meta:property="http://ex.net/event#participant";>Jane 
Doe</text:li>
   <text:li meta:property="http://ex.net/event#participant";>John 
Smith</text:li>
</text:list>

The items in the list are not one object (area) from a model 
perspective.

Bruce



[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]