Well, John, it looks like we agree on this point:
[John Kunze wrote:]
> I'm very glad you brought this up. If you think, for example, that a goal
> of the DC is to provide only a holder for application-specific metadata,
> with no particular shared semantics, then we have a foundation-level goal
> conflict that needs to be resolved immediately.
I.e., we need to resolve this point.
I would like to know if there are other people on the list who see the
difference of opinion and would also like to see it resolved -- in the
absence of any comments to that effect, I will assume this is a personal
difference of opinion between me and John, that I am missing the point of
the shared consensus on the purpose of this work, and take the thread off
the list (there is certainly enough other traffic at this time!).
[John wrote:]
> Good, bad, correct -- somewhat loaded terms. Let's just say "consistent",
> which doesn't weaken your point. From the User Guide:
>
> "One goal of this document is to encourage metadata consistency as a way
> to promote complete retrieval and intelligible display across widely
> disparate sources of descriptive records. With inconsistent metadata,
> search results tend either to omit desired records or to drown them out
> in a flood of irrelevant records. Effectively, inconsistent metadata
> hides desired records." [ http://www.ckm.ucsf.edu/meta/mguide3.html ]
>
I agree that it would be lovely to have consistent metadata the world over.
I also think it is a holy grail, not because people are stupid or obstinate,
but because there are too many competing _needs_ and _uses_ for metadata --
even when you identify a certain template of metadata attributes (i.e.,
the Dublin Core). _Global_ systems inherently imply heterogeneity (just
look at what happens when you try to specify "address" information in the
global white pages information -- that varies drastically from one country
to the next).
Rather than simply say that I think DC should be a holder for any
application-specific metadata, I would prefer to suggest the refined
statement that this group should focus on (i.e., I thought it _was_ focusing
on) building a good holder, and perhaps _separately_ considering how that
holder can be used for a particular application (e.g., if there are specific
views on how the citation information should be specified so that it can be
collected and served in a particular service). By separating out the two
concepts, a holder can be built that does serve many applications (flexibility
through simplicity), while a solution to one particular application's problems
can be shaped by describing a particular use of the holder.
This can be done, by the way, if the holder model includes the ability
to indicate which semantics are being used by the entity that fills
it. So, I am not saying that no semantics should be shared when attempting
to use these templates, I am saying that they should be shared at an
application-level, not a global-level.
Leslie.
--
----------------------------------------------------------------------------
"_Be_ Leslie Daigle
where you Vice President, Research
_are_." Bunyip Information Systems
-- ThinkingCat (514) 875-8611
[log in to unmask]
----------------------------------------------------------------------------
|