[crm-sig] 4 new issues

[Tony Gill]

OK, here are a few discussion issues I would like to propose:

1. Scope Notes for Properties
I think we need to provide scope notes for properties, to make their 
meaning more explicit. The need for property scope notes has also been 
voiced by the ABC/Harmony folks.

2. Collection class
I've always been unhappy about the absence of a Collection class in the 
CRM, but couldn't think of a good response to Martin's question about what 
it's unique properties should be. Well, Martin himself came up with a 
candidate property for the Collection class, "Curated by", so I'm very 
happy to propose that we now add this class. My initial inclination would 
be to make it a subclass of E24 Physical Man Made Stuff, since I think we 
should limit it's scope to the kinds of physical collections that museums, 
libraries and archives deal with (i.e. explicitly and deliberately exclude 
collections of conceptual objects such as ideas, dreams etc.), but it 
should probably include electronic objects.

3. Modeling States
At the DELOS harmonisation meeting with ABC/Harmony in Darmstadt recently, 
the notion of "states" came up, and it became very clear that this central 
aspect of the ABC/Harmony model was not really addressed by the CRM at 
all. For example, how would we model an assertion that an object O was at 
a location X at time T? The CRM can model a change of location event, but 
this is not exactly the same. Do we need to be able to model states in the 
CRM? I don't know, but it would make harmonisation with the ABC model a 
lot easier, so it's certainly worth adding to the list of issues.

4. Causality
Another issue that has come up before at least once in the CRM SIG (I 
believe Steve brought it up) is the modeling of causality -- to explicitly 
say that event E1 had some kind of causal role in the occurrence of event 
E2. I think that the CRM deliberately avoids this at present, favouring a 
more neutral and objective modeling paradigm. However, there are some 
causal relationships that are sufficiently unarguable that we may want to 
explicitly identify them: I propose that we could do this using the 
Property Scope Notes. For example, the scope notes for the property 
"destroyed" could identify it as being a causal relationship.

See you in Paris next month,

Cheers,

T.
----------------------------------------------------------------------
Tony Gill <> tony.gill at notes.rlg.org
Research Libraries Group <> http://www.rlg.org/
1200 Villa Street, Mountain View, CA 94041 USA
Voice: +1 (650) 691-2304 <> Fax: +1 (650) 964-1461






martin <martin at ics.forth.gr>
Sent by: owner-crm-sig at ics.forth.gr
17/09/2001 07:12 AM

 
        To:     crm-sig at ics.forth.gr
        cc: 
        Subject:        [crm-sig] 2 new issues


I just placed two new issues on the discussion list:

location of physical features
motivation of man-made objects

as issues nr 40 and 41. Please do not hesitate to formulate
issues for the next meeting, even if you are not sure if the subject
is already covered.