Comments on: News from the frontier: Geonames on the Semantic Web http://richard.cyganiak.de/blog/2006/10/news-from-the-frontier-geonames-on-the-semantic-web/ A weblog by Richard Cyganiak Wed, 03 Aug 2011 12:16:08 +0000 hourly 1 http://wordpress.org/?v=3.2.1 By: to.rtic.us » Blog Archive » dowhatimean.net » News from the frontier: Geonames on the Semantic Web http://richard.cyganiak.de/blog/2006/10/news-from-the-frontier-geonames-on-the-semantic-web/#comment-64811 to.rtic.us » Blog Archive » dowhatimean.net » News from the frontier: Geonames on the Semantic Web Mon, 30 Apr 2007 06:31:08 +0000 http://dowhatimean.net/2006/10/news-from-the-frontier-geonames-on-the-semantic-web#comment-64811 <p>[...] dowhatimean.net » News from the frontier: Geonames on the Semantic Web [...]</p> [...] dowhatimean.net » News from the frontier: Geonames on the Semantic Web [...]

]]> By: Semantic Web : Concept vs Document « Geonames http://richard.cyganiak.de/blog/2006/10/news-from-the-frontier-geonames-on-the-semantic-web/#comment-12219 Semantic Web : Concept vs Document « Geonames Sat, 21 Oct 2006 07:28:33 +0000 http://dowhatimean.net/2006/10/news-from-the-frontier-geonames-on-the-semantic-web#comment-12219 [...] With feedback and help from the blogosphere we have updated the geonames ontology and fixed some flaws in the corresponding web service. The ontology now includes linked data such as contains (administrative divisions), neighbours (for features with a boundary) or nearby features. The web service is using two URIs to clearly distinguish between Concept (the thing as is) and Document (the document about it). [...] [...] With feedback and help from the blogosphere we have updated the geonames ontology and fixed some flaws in the corresponding web service. The ontology now includes linked data such as contains (administrative divisions), neighbours (for features with a boundary) or nearby features. The web service is using two URIs to clearly distinguish between Concept (the thing as is) and Document (the document about it). [...]

]]> By: Richard Cyganiak http://richard.cyganiak.de/blog/2006/10/news-from-the-frontier-geonames-on-the-semantic-web/#comment-11992 Richard Cyganiak Mon, 16 Oct 2006 22:28:11 +0000 http://dowhatimean.net/2006/10/news-from-the-frontier-geonames-on-the-semantic-web#comment-11992 Bernard, conceptually there's no difference between a Web page and a Web API query result. In both cases, the server accepts a URI and answers with a document (HTML or XML or whatever), which may be generated dynamically. This, I think, is the key idea behind the REST style of web services. I've written <a href="http://dowhatimean.net/2006/10/fixing-ambiguous-concept-uris" rel="nofollow">some more about the 303 thing</a>. Bernard, conceptually there’s no difference between a Web page and a Web API query result. In both cases, the server accepts a URI and answers with a document (HTML or XML or whatever), which may be generated dynamically. This, I think, is the key idea behind the REST style of web services.

I’ve written some more about the 303 thing.

]]> By: Bernard Vatant http://richard.cyganiak.de/blog/2006/10/news-from-the-frontier-geonames-on-the-semantic-web/#comment-11970 Bernard Vatant Mon, 16 Oct 2006 08:06:40 +0000 http://dowhatimean.net/2006/10/news-from-the-frontier-geonames-on-the-semantic-web#comment-11970 Richard, Tim Thanks again for your quick reaction to geonames-in-RDF. Please consider again this is a first release, set up in two weeks through a few exchanges with Marc, and subject to improvements. Actually, I'm amazed by the feedback so far. About the TAG resolution, which I have read again and again trying to make sense of it - and must admit I have not yet ... (and yes, it's a consensus, and yes we have to apply it, even if some very smart people have explained why it's broken ...) So let me make sure what it is the TAG want here on an example : http://ws.geonames.org/rdf?geonameId=3014258 (1) should be only the URI for the "document" as it stands now (although I have hard time considering the result of a query as a "document", let's assume it), and we should have a different URI, like e.g. http://ws.geonames.org/rdf#geonameId=3014258 (2) or whatever, identifying the "Feature" itself, and have for (2) a 303 redirecting to (1). If it's only that, ask Marc to go for it, and he surely will : let me insist again he does all the real stuff, he's the data base and server administrator, and I just wrote a few lines of OWL for him. Richard, Tim

Thanks again for your quick reaction to geonames-in-RDF. Please consider again this is a first release, set up in two weeks through a few exchanges with Marc, and subject to improvements. Actually, I’m amazed by the feedback so far.
About the TAG resolution, which I have read again and again trying to make sense of it – and must admit I have not yet … (and yes, it’s a consensus, and yes we have to apply it, even if some very smart people have explained why it’s broken …)
So let me make sure what it is the TAG want here on an example :
http://ws.geonames.org/rdf?geonameId=3014258 (1) should be only the URI for the “document” as it stands now (although I have hard time considering the result of a query as a “document”, let’s assume it), and we should have a different URI, like e.g. http://ws.geonames.org/rdf#geonameId=3014258 (2) or whatever, identifying the “Feature” itself, and have for (2) a 303 redirecting to (1).
If it’s only that, ask Marc to go for it, and he surely will : let me insist again he does all the real stuff, he’s the data base and server administrator, and I just wrote a few lines of OWL for him.

]]> By: Tim Berners-Lee http://richard.cyganiak.de/blog/2006/10/news-from-the-frontier-geonames-on-the-semantic-web/#comment-11951 Tim Berners-Lee Sun, 15 Oct 2006 18:48:26 +0000 http://dowhatimean.net/2006/10/news-from-the-frontier-geonames-on-the-semantic-web#comment-11951 Bernard, PLEASE use HTTP 303s to from the concept to the document. This was the conclusion that the W3C TAG came to about what should be done. (Wheras ambiguity can be handled by people, it is a mess for machines. A place name and a document about it quite different. ) For example, the tabulator view http://dig.csail.mit.edu/2005/ajar/ajaw/tab.html?uri=http://ws.geonames.org/rdf%3FgeonameId%3D2950159 shows something which "mentions Feature, and is inCountry DE". That's silly. The file mentions the class "Feature". The cty is in Germany. Confusing them really does not help. This is a tremendous set of data you have there. Richard's review is spot on. The data is currently spolied by this architectural flow which could be fairly easily fixed, as is done in semantic wiki sites, among many. (The backlinks is the one thing which is less clear. If there are many, then it may be useful to give instead of all the "contains" give a pointer "containsList" to a virtual documnent which has a list of tall the things Berlin contains. This is like having a pointer to a list of publications on your FOAF page instead of all the publications themselves.) Bernard, PLEASE use HTTP 303s to from the concept to the document. This was the conclusion that the W3C TAG came to about what should be done. (Wheras ambiguity can be handled by people, it is a mess for machines. A place name and a document about it quite different. )

For example, the tabulator view http://dig.csail.mit.edu/2005/ajar/ajaw/tab.html?uri=http://ws.geonames.org/rdf%3FgeonameId%3D2950159
shows something which “mentions Feature, and is inCountry DE”. That’s silly. The file mentions the class “Feature”. The cty is in Germany. Confusing them really does not help.

This is a tremendous set of data you have there. Richard’s review is spot on. The data is currently spolied by this architectural flow which could be fairly easily fixed, as is done in semantic wiki sites, among many.

(The backlinks is the one thing which is less clear. If there are many, then it may be useful to give instead of all the “contains” give a pointer “containsList” to a virtual documnent which has a list of tall the things Berlin contains. This is like having a pointer to a list of publications on your FOAF page instead of all the publications themselves.)

]]>