Hi Karen,

please also see my latest post on the subject. I just want to comment on your 
differentiation between "distribution" and "publication".

> Would it cause problems if  two types of elements are identified within a 
> media package:  a "distribution" element to represent the actual file at its 
> distribution location, and "publication" elements that represent publishings 
> and logical links for a user to access the media. The  "publications" don't 
> have to know about each other, but the "publication"s do know about the 
> related "distribution" object.

This would only work if there is a 1:1 relationship between the original media 
package element and its role within a publication, i.e. if the files are taken 
as is and copied to a location where they are accessible via a URL. With 
metadata this will must likely be not the case. Many channels will read only a 
subset or create a digest, transform them to a different format and then store 
the result maybe in a database. In these cases you cannot create such a 
distribution element but only an element that notes that e.g. the episode 
dublin core has been used. See my other post for further details.

Christoph


_______________________________________________
Matterhorn mailing list
[email protected]
http://lists.opencastproject.org/mailman/listinfo/matterhorn


To unsubscribe please email
[email protected]
_______________________________________________

Reply via email to