This shows you the differences between two versions of the page.

ww2 [2009/10/25 03:38] UJYQrbUlcNNonz
ww2 [2009/10/26 07:12] (current)
akfarrell old revision restored
Line 1: Line 1:
-Mxg637  <a href="http://axicdwufopnn.com/">axicdwufopnn</a>, [url=http://jccvromyleju.com/]jccvromyleju[/url], [link=http://cnwxrzpfdagv.com/]cnwxrzpfdagv[/link], http://gxmewrilhmtr.com/+==== RDF - WTF? ==== 
 +**Notes by [[http://twitter.com/incanus77|Justin Miller]] -- sorry for typos/omissions** 
 +RDF is machine-readable information about information 
 +triples: subject-predicate-object 
 +anything in a relational DB can be represented in triples 
 +    have a username: “joesmith is a user of X” 
 +SPARQL: query language across resources 
 +predicates themselves can, as human knowledge, be described in RDF as well 
 +alternatives to using XML for RDF (which turned out to be verbose and bad) 
 +    Turtle 
 +practical applications 
 +    biggest: FOAF (friend of a friend) 
 +        describes social network information 
 +        build it yourself, publish it on your site 
 +        becomes your social network identity 
 +        LiveJournal auto-publishes it for everyone 
 +        WordPress, Blogger, Movable Type as well for years 
 +        vast majority of live RDF today is FOAF 
 +post info about RDF in Drupal 
 +    http://drupal.org/project/rdf 
 +    http://drupal.org/project/fileframework 
 +tool called Zotero - like Evernote 
 +    Firefox plugin 
 +        bookmark a site, tag it (like Delicious) 
 +        also: senses bibliographic info on the page & extracts it for your records 
 +        example: your on an Amazon book page, Zotero saves info into your database 
 +question: why use FOAF vs. some other format?  
 +    it’s already a standard 
 +    tools already exist 
 +N3 allows you to define formalae 
 +    example: X is father of Y, A is cousin of Y, implies X is uncle of A 
 +Thetus: “knowledge discovery software” (based in Portland) 
 +interesting application: trust networks as extension of FOAF 
 +downside: hard to store a lot of it 
 +    generally you store it in your own format and export RDF (or make it available on demand) 
 +    Billion Triple Challenge: project trying to tackle this challenge 
 +question: anyone looked into using distributed things like CouchDB to store it?  
 +    problem is they are usually key-value, not triples 
 +    someone needs to build a distributed storage DB for this?  
 +    storage of quads: adding security or lineage info as well 
 +DBpedia: http://en.wikipedia.org/wiki/DBpedia 
 +    trying to take all of Wikipedia’s info and putting it into RDF 
 +question: can we mashup the CIA World Factbook with Wikipedia?  
 +    that’s the dream of RDF, really 
 +    if everything has an absolute URI, then it’s more easily possible 
 +question: what about connecting FOAF info between, say, LJ and Blogger?  
 +    unfortunately a lot of current use is internal to organizations 
 +    but you could write your own FOAF integrating across services and publish that 
 +question: what happens when things change names (say: countries) 
 +    depends on how you define your predicates 
 +        “this country is another name for this country” 
 +question: can you explain the OWL (Web Ontology Language) language?  
 +    http://www.w3.org/TR/owl-features/ 
 +    includes mathematical and logical relationships 
 +    example: saying a pine tree is a type of plant implies that plants include trees 
 +also check out Yahoo’s YQL session today at 6:00pm
ww2.txt · Last modified: 2009/10/26 07:12 by akfarrell