W3C home > Mailing lists > Public > public-lod@w3.org > November 2009

Re: Subjects & Tagging - Help?

From: Nathan <nathan@webr3.org>
Date: Tue, 03 Nov 2009 20:53:09 +0000
Message-ID: <4AF09835.4040009@webr3.org>
To: Bill Roberts <bill@swirrl.com>
CC: Alexandre Passant <alexandre.passant@deri.org>, public-lod@w3.org, "kidehen@openlinksw.com" <kidehen@openlinksw.com>
Bill Roberts wrote:
> 
>> Finally, regarding dc:subject, a tag can be used not as a subject 
>> (think of a webpage tagged "cool" or "todo", they are probably not 
>> used as subject) so the semantics of dc:subject is probably not what 
>> you want here.
>>
> I think this comment from Alexandre gets to the core of the matter.  
> Tagging has grown up as a kind of "metadata-lite", ie a quick way of 
> categorising blog posts etc into topics, often fairly broad topics.  Of 
> course it can be a very useful way of finding related material.  However 
> if you are able to say something more precise, then I think you should - 
> eg by using dc:subject or foaf:primaryTopic, to say that a blog post (or 
> other document) is about a particular thing that you have a URI for.
> 
> You might also want to have tags of course, in their role of more 
> general categorisation and the Common Tag ontology looks a nice way to 
> handle that, especially with its capabilities for saying which user 
> added a tag - so letting you handle an applicatoin like delicious.com.
> 
> Cheers
> 
> Bill
> 

Thank you all,

and.. sorry to keep asking! However I still don't feel like I've got 
this down yet; I can follow most ontologies, and most custom / common 
data sets are pretty simple to map; geo data is geo data, a person is a 
person with a name etc - however pretty rapidly a huge chunk of data 
will be going in to the cloud from all the "bloggers" out there, and 
dev's making CMS systems; and if we don't have a standard way of 
associating our content that's well known then.. well most of the data 
(masses of it) will be pretty much junk I guess.

so.. to nail this once and for all, can we take the following article:

http://www.google.com/hostednews/afp/article/ALeqM5hU45P_cGaJc9YjXHHpDrNf9VBuWg

and the following (auto) extracted subjects/terms/tags (each with a uri 
for the dbpedia resource, and in the case of countries a uri for the geo 
resource too)

topics
	Foreign relations of Iran
	Science and technology in Iran
	Iran  United States relations
	Politics of Iran
	Nuclear program of Iran
	Enriched uranium
people
	Barack Obama
	Dmitry Medvedev
	Ali Asghar Soltanieh
	Nicolas Sarkozy
orgs
	United Nations
	International Atomic Energy Agency
	White House
geo
	Iran
	Tehran
	United States
	Moscow
	France
	Qom
	

In your opinion(s) what would the relevant triples be to correctly 
associate the content to these resources (remember I have the relevant 
URIs for each literal above); primary consideration of course goes to 
the ontologies people would use to find this data in the future.

Many thanks in advance, I can assure you any time spent on this bit of 
guidance won't be wasted.

Nathan
Received on Tuesday, 3 November 2009 20:54:06 UTC

This archive was generated by hypermail 2.3.1 : Wednesday, 7 January 2015 15:16:01 UTC