ah-ha!
so slowly 'I' am getting to understand better and wonder if there are
others like me who might find it useful to think about this issue? I
have battled with the descriptions of people, not just because there
are zillions of things that one might want to say about people and,
of course, they would be in application profiles, but because I have
had trouble understanding all the terms that were chosen way back in
theoretical ways. The abstract model (DCAM) helps a lot but it is
very difficult in retrospect to create a model when some of the terms
were just 'chosen'. (I have re-read some of the early docs and know
that at times it was almost exhaustion with the problems that led to
the choices that were necessary to satisfy the very wide range of
domains in which people wanted DC terms to be able to function.)
BUT
if I think of people as being related to the resource, and
specifically the creator, contributor, publisher terms as important
qualifications of the 'relation' term, I am much more comfortable
with those terms. So now I can think about what I might want to say
about 'people'.
First, I extend 'people' to 'agents' because it might not be a person
who creates, contributes, etc... Then I can start to think about how
people might be described. If there are terms that I like in other
metadata sets, I need to find ways of expressing the same info in a
way that conforms to the DCAM if those terms do not already
conform. And when I do that, I can use the DCAM to think about how
I make descriptions of the people (agents) I am interested in. Once I
have my terms for expressing my descriptions, I can use them in a
range of ways. This is possible because within RDF there are good
ways of making and linking statements.
* I can use DC conformant terms to describe the relations the
resource has with an agent - illustrator, photographer, needle-
pointer, ...
* I can use DC conformant terms to describe a person when there is
no particular resource in my mind
* I can use DC conformant terms to relate resources to the person.....
Maybe a lot of us can agree about some sets of terms and we can
develop application profiles for specific communities..
So that is how I now understand the situation, and my encyclopaedia
and dictionaries of biographies colleagues might want to contribute
to the list of reqs for agents, alongside others who work in
government, education, health, sport, ..... (please contact the
Agents Community leaders)
Correct????
Please feel free to correct my story .....
Liddy
On 11/12/2006, at 9:25 PM, Liddy Nevile wrote:
>
> OK - good news - I stand corrected - I didn't realise that the
> agents group was doing this in this way - it is how I too would
> like it done and I think it'd be good for my encyclopaedia
> colleagues -
>
> Andrew, do you think others are making the mistake I have been making?
>
> I should see now if I can get them to help...
>
>
> Liddy
>
>
> On 11/12/2006, at 9:02 PM, Andrew Wilson wrote:
>
>>
>> Exactly. What Pete says - the Agents WG is not limited in the way
>> that Liddy suggests. John Robert and I, the co-chairs of DC-
>> Agents, have been trying to get interest and involvement for the
>> last couple of years with absolutely no result. The Agents WG is
>> working on developing a way of describing agents - at the moment
>> we are changing the WG into a task Group consisting intially of
>> myself, Dan Brickley, and Tom Baker to look at whether and how
>> FOAF meets the needs of the DC community for describing agents.
>> Perhaps this discussion could be copied to the DC-Agents email list?
>>
>> cheers
>> Andrew Wilson
>>
>> Quoting Pete Johnston <[log in to unmask]>:
>>
>>> Liddy,
>>>
>>>> I think that DC agents is a group who are working on what should be
>>>> the DC way to approach vales for such elements as creator,
>>>> contributor, publisher, etc while I think the encyclopaedia people
>>>> want to describe people, not resources....
>>>
>>> Sorry, I don't understand the distinction you are making here.
>>>
>>> People are resources. In the terms of the DCMI Abstract Model,
>>> the "values" for the creator, contributor, publisher etc
>>> properties _are_ people (well, they are "entities" or "agents",
>>> which might also be resources other than people, like
>>> organisations or services).
>>>
>>> So, as Irvin says, yes, this - the description of agents - is
>>> _exactly_ the remit of the DC Agents WG. Scanning the mailing
>>> list archive
>>>
>>> http://www.jiscmail.ac.uk/lists/DC-AGENTS.html
>>>
>>> suggests that the Agents WG has received almost no interest for
>>> at least the last couple of years.
>>>
>>> Pete
>>>
>>
>>
>>
>> --
>> Andrew Wilson
>> [log in to unmask]
|