Coincidentally, I'm 2/3 of the way through providing the "easily" mappable MARC codes lists to RDF.
Here are Geographic Area Codes: http://purl.org/NET/marccodes/gacs/n-usa#location (Use the values here: http://www.loc.gov/marc/geoareas/gacshome.html) And Languages: http://purl.org/NET/marccodes/languages/eng#lang (see: http://www.loc.gov/marc/languages/langhome.html) Tomorrow morning come the countries list: http://www.loc.gov/marc/countries/cou_home.html >From there, well, we'll see how well I can match Geonames. -Ross. On Wed, Nov 11, 2009 at 5:26 PM, Hugh Glaser <[email protected]> wrote: > Absolutement. > More authoritative sources would be great – in fact dbpedia may be as good as > it gets, as it often is (at least for the moment). > > As far as sameAs .org is concerned, I only (aim to, but sometimes I fail) > include URIs that resolve to RDF. > And of course, often need others to have done the hard work of establishing > the link. > So even for the excellent http://www.fao.org/countryprofiles/geoinfo.asp > there are (as far as I can see) no resolvable URIs, nor any sameAs links. > > Best > Hugh > > > On 10/11/2009 09:20, "Bernard Vatant" <[email protected]> wrote: > > Hugh > > The actual problem, as is well shown by your sameas.org <http://sameas.org> > example, is not the lack of URIs for countries, but to figure out which are > "cool" (stable, authoritative, published following best practices). > sameas.org <http://sameas.org> yields 23 URIs for Austria, 29 for France etc. > Supposing they are all really "equivalent" in the strong owl:sameAs sense, > any of those should do, but ... > On the other hand, maybe more authoritative sources are absent of the > sameas.org <http://sameas.org> list, such as the excellent FAO ontology > pointed by Dan. And above all, which is definitely missing are sets of URIs > published by ISO itself. > There is an ongoing work aiming at authoritative URIs for ISO 639-2 languages > by its registration authority at Library of Congress. > http://www.loc.gov/standards/iso639-2/. As I understand, those URI will be > published under http://id.loc.gov/authorities/, so watch this space. I cc > Rebecca Guenther who is in charge of this effort at LoC, she'll certainly be > able to provide update about this, and maybe she's aware of some equivalent > effort for ISO 3166-1. But according to an exchange I had with her a while > ago, ISO itself might be "years away" from publication under its own > namespace, unfortunately. > > Bernard > > > 2009/11/9 Hugh Glaser <[email protected]> > There are quite a few, but I don't know which other ones follow ISO 3166-1. > http://sameas.org/?uri=http://dbpedia.org/resource/Austria > Gives a selection. > Or also > http://unlocode.rkbexplorer.com/id/AT > http://ontologi.es/place/AT > > Our site, http://unlocode.rkbexplorer.com/id/AT > is our capture of UN/LOCODE 2009-1, the United Nations Code for Trade and > Transport Locations, which uses the 2-letter country codes from ISO 3166-1, > as well as the 1-3 letter subdivision codes of ISO 3166-2 > See http://www.unece.org/cefact/locode/ > It also gives inclusion and coords, etc. > We need to do more coref to other than onologi.es <http://onologi.es> . > > Best > Hugh > > On 09/11/2009 21:47, "Aldo Bucchi" <[email protected]> wrote: > >> Hi, >> >> I found a dataset that represents countries as two letter country >> codes: DK, FI, NO, SE, UK. >> I would like to turn these into URIs of the actual countries they represent. >> >> ( I have no idea on whether this follows an ISO standard or is just >> some private key in this system ). >> >> Any ideas on a set of candidata URIs? I would like to run a complete >> coverage test and take care I don't introduce distortion ( that is >> pretty easy by doing some heuristic tests against labels, etc ). >> >> There are some border cases that suggest this isn't ISO3166-1, but I >> am not sure yet. ( and if it were, which widely used URIs are based on >> this standard? ). >> >> Thanks! >> A > > > > > >
