True, I didn't really mean 'a set' - obviously it would be 'sets' of
URLs. But I think Richard's point about material within the cultural
heritage domain sets important boundaries - most people listed as
makers, designers, patrons, etc in collections databases are notable
enough for wikipedia/dbpedia, but you would start to find multiple
references to them across Europeana/Digital Public Library of
America/Trove etc - doing something with the existing data in those
big repositories would be a good start. Barry/Dominic - will the
British Museum's ResearchSpace stuff tie into other systems or would
people bring their data to ResearchSpace?
More contextual information is definitely a help but it's still
difficult to automate matches. I've seen various academic projects
that use manual verification of potential concordances (i.e. is the
Dorothea Erxleben in this site the same person as Dorothea Leporin on
that site) and as a microtask it would be perfect for a focused
crowdsourcing project.
There are various projects to build historical gazetteers to provide a
Geonames/Foursquare for the past (and as usual the classicists/ancient
world people are quietly getting on with it in e.g.
http://pelagios-project.blogspot.co.uk/p/about-pelagios.html), etc,
etc.
On Sarah's point - I don't know if it helps but in the past I've
separated the 'display date' from underlying date fields (period,
earliest date, latest date, designed/manufactured/distributed date,
etc) to provide a neat, computer-friendly version while keeping the
original data as messy as it needs to be. It's an attempt to have all
records show on things like timelines even when the exact date isn't
known or isn't a simple question.
Cheers, Mia
On 13 December 2013 11:10, Frankie Roberto <[log in to unmask]> wrote:
> Mia wrote:
>
>> Thank you, Richard, for sharing your work* and your conclusions. I
>> agree - any uses of open cultural data would benefit from an
>> authoritative set of URLs about relevant people, places, materials,
>> events and concepts to use as a shared reference. In practical terms,
>> does anything currently or almost in existence come close to that now?
>
> This is definitely an issue.
>
> My bet is that we’re unlikely to get a single set of authoritative URLs / ids for people/places/etc.
****************************************************************
website: http://museumscomputergroup.org.uk/
Twitter: http://www.twitter.com/ukmcg
Facebook: http://www.facebook.com/museumscomputergroup
[un]subscribe: http://museumscomputergroup.org.uk/email-list/
****************************************************************
|