this touches on a number of related issues: identifiers, naming,
cross-walking between user-supplied vocabulary and extant metadata
for resource, as well as granularity.
it has been useful and is desirable that serials have an identifier,
the ISSN (as it happens, a dumb number assigned to each
manifestation, now linked by the iSSN-L), with full use being the
number and the key title (the latter being a modification to
disambiguate the shared use of the same string).
with geo-graphic terms, like placenames, the geo-referencing plays a
comparable role, although not as a dumb number and allowing geo-
spatial operation across the variant vocabulary of use and resource.
as James indicates, a lot of work exists, and it would be good if DC-
like schemes could benefit, both de jure and in deployment.
hth
Peter
On 3 Feb 2009, at 13:38, Leslie Carr wrote:
>
> On 3 Feb 2009, at 12:03, Andy Powell wrote:
>
>>> I think its important that we move to an acceptance that
>>> *explicit* i.e.
>>> coordinate reference based georeferencing is preferable to
>>> *implicit* i.e. purely textual
>>
>> Depends! What's the user requirement?
>
> Most of the time when I want to indicate "London, UK" then I it is
> providing a context about a specific event or a specific entity.
> In the repositories I have responsibility for, this may be the
> location of the conference where the item was presented, or the
> location of the institution to which the author is affiliated.
>
> So "London, UK" might mean "the British Library", or "Novotel
> Hotel, Euston Road" or "Birkbeck" or the "University of Middlesex".
> It is traditional in bibliographic citations to include the city of
> the journal publisher, or the city of the conference, so perhaps
> that explains my split attempts to have both a general and specific
> location.
> --
> Les
--
The University of Edinburgh is a charitable body, registered in
Scotland, with registration number SC005336.
|