RE: Use of www-rdf-dspace for comments re: early draft note, DSpa ce H istory System

comments below.

- Mick

1.2 History System
s/"The History System describes how the metadata associated with a work changes over time"
 /"The History System describes how the content and metadata associated with a work changes over time"/

- It would add clarity to add a section "About this document" between 1.2 and 1.3 which refers to the context (History System Statement of Work), the purpose of this document (to capture for the record what the History System does and does not do in its current form, and to capture issues and recommendations), and what will come next (some work to implement those recommendations).

2.1 Namespaces

"Used for internal purposes, the http://www.dspace.org/ is used to contain properties that are required by the implementation of the History System.  They are not generally useful metadata."
--> it would be helpful to list some examples of (or to enumerate) such properties, so the validity of the assertion is more clear.

2.2.1 Collection
--> has_part properties model the relationship between this collection and the items that it contains.
[true?  not true?]

"Each Item has associated Dublin Core metadata"
--> This statement fits more logically in 2.2.2, which describes items.

2.2.2 Item
"A Manifestation of a conceptual work..."
--> We might have a rich dialogue regarding the mapping of FRBR to objects/classes within Dspace.  It is not clear that the mapping suggested here would yield consensus.  While this may well lead to an open issue, I suspect that it may be out of scope for the present work on the History System.  Time permitting, I will follow up with a separate note on the topic.

2.2.4 Community
"Set of Epersons that are logically or functionally grouped".
--> I do not believe that this is what is modelled within a DSpace Community.  Rather, the community models a set of community-specific metadata (as described in the properties section of the document), as well as the set of collections within the community (via has_part properties).

2.4.2 Item Properties
-->potential issue.  Dspace consistently assigns CNRI Handles as Item URIs (Dublin Core: identifier.uri).  Is there a need to identify / be able to query an Item's assigned handle (as opposed to other URI's which may have been exogenously assigned)?  i.e. should a Dspace-specific sub-property exist for the primary persistent ID's that DSpace assigns?

2.5 Values
--> This section appears to be incomplete.

3.1.2 Duplicate Properties
--> It would be helpful to see an instance of the example used... I'm not convinced that hasPart belongs in DublinCore - if it is turning up there it may be an error that should be corrected, rather than cause for a subPropertyOf relationship.

3.1.3 Lack of Type Information
--> potential issue: ought DSpace objects have subclass relationships from appropriate Harmony base classes (e.g. DSpace item is subclass of Harmony artifact?)

Potential Additional Issue Sections:

== FRBR Mapping & Usage
--> This is of interest to some PIs but is likely out-of-scope for Jason's engagement.

== Completeness of Resource and Event Model
--> One potential issue in the current implementation of the History System may be that it does not currently model sufficient resources and/or events to capture "changes of archival interest".  In particular, most goings-on in the administrative interface (e.g. Group Management and Item Viewing policies) are not currently modelled in the History System.

== Hasn't there been some discussion of what the [recommended] namespace URLs resolve to?  i.e. stated desire for these to resolve to an RDF Schema?



> -----Original Message-----
> From: Bass, Mick [mailto:mick.bass@hp.com] 
> Sent: Wednesday, May 07, 2003 10:48 AM
> To: SIMILE public list
> Subject: Use of www-rdf-dspace for comments re: early draft 
> note, DSpace H istory System
> 
> 
> 
> One of the implementation efforts underway within SIMILE aims 
> to improve the current implementation of the DSpace History 
> System.  The DSpace History System produces RDF that is 
> intended to usefully describe the temporal history of 
> significant events and changes to the state of archived 
> materials (content and metadata).  See the statement of work [1].
> 
> An *initial, early draft* of the first deliverable indicated 
> in the statement of work is available [2]:
> 	"Review the output of the DSpace history system
> 	 and write a research note that enumerates and describes
> 	 the known issues with the RDF that it produces."
> Please bear in mind that this document is an early stage 
> initial draft.
> 
> I have requested comments from several individuals on behalf 
> of the author, and propose to use this list as a vehicle to 
> collect such comments (for referenceability and archival).  
> The purpose of this note is to establish a thread for these 
> comments, and to offer a heads up to the wider community that 
> observes this list.
> 
> - Mick Bass
> 
> [1] 
> http://web.mit.edu/simile/www/resources/history-harmony/histor
y-statement-of-work.htm
[2] http://web.mit.edu/simile/www/documents/historySystem/descriptiveNote/descriptiveNote.pdf


=============================================
Mick Bass


Manager
Research and Business Development
HP Laboratories
Hewlett-Packard Company
1 Cambridge Center
Cambridge, MA 02142


617.551.7634 office    617.551.7650 fax
617.899.3938 mobile    617.627.9694 residence
bass@alum.mit.edu      mick_bass@hp.com
=============================================

Received on Wednesday, 7 May 2003 11:35:29 UTC