Semantic Web Interest Group IRC Chat Logs for 2005-08-22

00:08:07 <chimezie> dc_swig:views

00:08:08 <dc_swig> A: RDF Home Page - Last Update 2004/10/21 (http://www.w3.org/RDF/)

00:08:09 <dc_swig> B: SW Activity: Advanced Dev - Last Update 2003/07/09 (http://www.w3.org/2000/01/sw/)

00:08:10 <dc_swig> C: Dave Beckett's Page about RDF - Last update 08-June-2005 (http://www.ilrt.bris.ac.uk/discovery/rdf/resources/)

00:08:11 <dc_swig> D: Interpreting XML via an RDF Schema - Michael Klein (PDF) (http://www.cs.vu.nl/~mcaklein/papers/ECAI02-SAAKM.pdf)

00:08:26 <chimezie> D:->swig,rdf,xml,owl

00:08:26 <dc_swig> Set keywords for D.

01:18:05 <chimezie> D:See also: W3C [Note|http://www.w3.org/TR/xml-infoset-rdfs]

01:18:06 <dc_swig> Added comment D1.

01:18:29 <chimezie> D:Corresponding RDF [schema|http://www.w3.org/2001/04/infoset]

01:18:30 <dc_swig> Added comment D2.

01:27:57 <crschmidt> ^q construct {?p2 ?prop ?p. } WHERE { ?prop rdf:type <http://www.w3.org/2002/07/owl#SymmetricProperty>. ?p ?prop ?p2. }

01:27:57 <julie> Total of 542 statements: Here's the first three. {(r0_r1114530965r33008), [http://purl.org/vocab/relationship/colleagueOf], (r0_r1114530965r32995)}, {(r0_r1114530965r32998), [http://purl.org/vocab/relationship/colleagueOf], (r0_r1114530965r32995)}, {(r0_r1114689381r708), [http://purl.org/vocab/relationship/colleagueOf], (r0_r1114530965r32995)},

01:33:18 <crschmidt> ^construct {?p2 ?prop ?p. } WHERE { ?prop rdf:type <http://www.w3.org/2002/07/owl#SymmetricProperty>. ?p ?prop ?p2. }

01:33:26 <julie> Created 542 statements based on CONSTRUCT query: Model size changed by 481.

01:33:41 <crschmidt> ^construct {?p2 ?prop ?p. } WHERE { ?prop rdf:type <http://www.w3.org/2002/07/owl#SymmetricProperty>. ?p ?prop ?p2. }

01:33:49 <julie> Created 1023 statements based on CONSTRUCT query: Model size changed by 454.

01:33:59 <crschmidt> hmph.

01:35:03 <crschmidt> logger: pointer

01:35:03 <crschmidt> See http://www.ilrt.bris.ac.uk/discovery/chatlogs/swig/2005-08-22#T01-35-03

03:18:15 <alk> alk is now known as aminorex

03:56:40 <chimezie> ^q select ?comment {<http://schemas.4suite.org/4ss#rdfdocument> rdfs:comment ?comment}

03:56:40 <julie> A repository document that represents a named subgraph (a scope) and the set of RDF triples contained within. The name of graph the RDF triples belong to is set to the absolute repository path of the RDF document. When the content is requested , the subgraph is serialized into RDF/XML and returned. When new content is provided, it is parsed into the subgraph as RDF/XML. So, a repository RDF document can be thought of as a window

03:56:40 <julie> to the subgraph through which RDF triples can be added and retrieved.

03:57:36 <chimezie> ^q select ?ancestor {<http://schemas.4suite.org/4ss#rdfdocument> rdfs:subClassOf ?ancestor}

03:57:36 <julie> http://www.w3.org/2004/03/trix/rdfg-1/Graph, http://schemas.4suite.org/4ss#xmldocument

03:58:43 <chimezie>http://metacognition.info/ontologies/4SuiteRepository.owl

03:58:44 <dc_swig> A: http://metacognition.info/ontologies/4SuiteRepository.owl from chimezie

03:58:57 <chimezie> A:|4Suite Repository Ontology - Expanded

03:58:58 <dc_swig> Titled item A.

03:59:32 <chimezie> A:->swig,4suite,owl

03:59:34 <dc_swig> Set keywords for A.

04:00:21 <chimezie> A:Expounded and linked with appropriate / related Vocabularies (WordNet/Foaf/Dublin Core)

04:00:21 <dc_swig> Added comment A1.

04:00:52 <chimezie> A:Also added a healthy dose of literature references (via rdfs:seeAlso)

04:00:52 <dc_swig> Added comment A2.

04:02:20 <chimezie> A:The aim is to use the formalism of OWL/RDFS to serve as a framework for documenting the 4Suite repository (to augment the very sparse documentation that does exist) as well as provide a formal model for the underlying RDF Graph that 'drives' the Repository

04:02:21 <dc_swig> Added comment A3.

04:02:42 <chimezie> ^q select ?ancestor {<http://schemas.4suite.org/4ss#repository> rdfs:subClassOf ?ancestor}

04:02:42 <julie> wn16:File-System, wn16:Repository, http://schemas.4suite.org/4ss#container

04:04:06 <chimezie> ^q select ?ancestor,?literature {<http://schemas.4suite.org/4ss#documentdefinition> rdfs:subClassOf ?ancestor;rdfs:seeAlso ?literature}

04:04:06 <julie> wn16:Typology https://www6.software.ibm.com/developerworks/education/x-4suite5/x-4suite5-2-5.html, http://schemas.4suite.org/4ss#xmldocument https://www6.software.ibm.com/developerworks/education/x-4suite5/x-4suite5-2-5.html

07:14:24 <russell_> russell_ is now known as optika

07:33:16 <etnaRosso> hi all

08:16:13 <etnaRosso> someone can help me to understand rdf?

08:17:07 <etnaRosso> leasa

08:17:15 <leasa> hello

08:17:51 <etnaRosso> hello, how do you do?

08:18:08 <leasa> Ok I think

08:18:33 <leasa> what's up

08:18:43 <etnaRosso> can you help me to create my first rdf document?

08:18:56 <leasa> yup

08:19:06 <leasa> or at least I can try

08:22:39 <etnaRosso> ok

08:23:07 <etnaRosso> first thing is declaration of namespaces and xml version, it's right?

08:27:33 <etnaRosso> so using <rdf:RDF xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#" xmlns:doc="http://www.w3.org/2000/10/swap/pim/doc#">

08:27:59 <etnaRosso> i can use properties in 22 rdf sintax and the ones in pim, right?

08:38:38 <leasa> sorry I was looking elsewear for a bit

08:38:40 <leasa> the RDf line looks right

08:38:59 <leasa> but sometimes (i think) you may want to use entities to

08:39:06 <leasa> so you may want to start with

08:39:28 <leasa> <?xml version="1.0" encoding="UTF-8"?>

08:39:35 <leasa> <!DOCTYPE rdf:RDF [

08:39:38 <etnaRosso> yes you are right

08:40:02 <leasa> <!ENTITY doc "http://www.w3.org/2000/10/swap/pim/doc#"

08:40:11 <leasa> ]>

08:40:15 <etnaRosso> ok rdf syntax ns and doc can be defined ontologies?

08:40:59 <leasa> if you may be using it as rdf:resource="&doc;anyURI"

08:41:17 <leasa> otherwise i do not think you need to start with enterties

08:41:31 <leasa> yes they can

08:41:43 <leasa> hold on someone need to tlak to me hear

08:41:44 <etnaRosso> mhm

08:41:56 <etnaRosso> ok

08:54:58 <kjetilkWork> etnaRosso: You could also have FOAF-a-matic generate some RDF/XML from you and start by dissecting it: http://www.ldodds.com/foaf/foaf-a-matic.html

08:55:12 <etnaRosso> yes i know

08:55:38 <etnaRosso> I would like to understand how to use web ontologies to create valid rdf documents

08:56:03 <kjetilkWork> ok

08:56:27 <etnaRosso> can you help me?

08:56:45 <kjetilkWork> I don't have a lot of practical experience, I'm afraid

08:57:03 <kjetilkWork> far from enough to explain it to someone else in a grokkable form... :-)

08:57:04 <etnaRosso> don't be afraid

08:57:18 <etnaRosso> grokkable?

08:57:19 <kjetilkWork> and there is little time today...

08:57:31 <kjetilkWork> "grok" = fully understand

08:57:43 <kjetilkWork> hacker sland

08:57:44 <kjetilkWork> err

08:57:47 <kjetilkWork> slang

08:58:12 <etnaRosso> we can use a simple ontology for example

08:59:19 <etnaRosso> choose one

09:00:26 <leasa> FOAF?

09:01:48 <etnaRosso> ok

09:02:58 <etnaRosso> foaf is defined using shema?

09:03:03 <etnaRosso> schema

09:08:39 <bjoern_> [11:07] <bjoern_> [[[

09:08:39 <bjoern_> [11:07] <bjoern_> latest.asis Q= 0.094

09:08:39 <bjoern_> [11:07] <bjoern_> latest.rss.new Q= 0.094

09:08:39 <bjoern_> [11:07] <bjoern_> ]]] -- http://lists.w3.org/Archives/Public/public-rule-workshop-discuss/latest

09:08:40 <bjoern_> [11:07] <bjoern_> now that's a good interface

09:08:41 <bjoern_> [11:07] <bjoern_> and http://lists.w3.org/Archives/Public/public-rule-workshop-discuss/latest.rss.new is gone...

09:18:29 <etnaRosso> foaf seems too comples

09:18:49 <etnaRosso> complex, can anybody suggest me some simpler specification?

09:23:15 <jeen> etnaRosso, what exactly do you want to know?

09:23:55 <etnaRosso> how i can use web ontologies

09:27:26 <jeen> well, that's a very broad question of course, but the beginning seems to be there in the bits that you posted. to be honest though, you probably shouldn't focus on the (XML) syntax too much right now, and instead try to understand things in terms of the more abstract (graph) layer.

09:27:38 <CaptSolo> chaals :)

09:27:58 <CaptSolo> etnaRosso: in what way is foaf complex?

09:28:30 <etnaRosso> cause it inerith a lot of classes

09:29:28 <CaptSolo> etnaRosso: but Windows or Linux desktop contains shitloads of code and classes, yet you use them and probably say they are simple to use

09:29:56 <CaptSolo> as about FOAF - it's actually very simple, especially if you don't focus on XML representation

09:30:02 <CaptSolo> there is a Person

09:30:06 <etnaRosso> yes but i'm investigating

09:30:15 <etnaRosso> the web ontologies structures

09:30:18 <CaptSolo> and there is relation foaf:knows

09:30:21 <chaals> yeah. The trick for FOAF is not to try and understand everything it can do, but to understand what you want to do and how FOAF does that.

09:30:38 <CaptSolo> that's the basic, plus some properties to describe persons

09:30:48 <chaals> etnaRosso, well then you have a problem. The ontologies structures are really really complicated, since they are the size of the web.

09:30:50 <CaptSolo> allows you to say:

09:31:02 <CaptSolo> person1 foaf:knows person2 .

09:31:08 * jeen digs up a book recommendation

09:31:16 <chaals> (each one starts out really simple, but they have to be the size of teh web so you can actually deal with the information out there)

09:31:23 <CaptSolo> person1 foaf:mbox somemail@gmail.com .

09:31:25 <jeen> This book :http://www.amazon.com/exec/obidos/redirect?tag=openrdforg-20&creative=9325&camp=1789&link_code=as2&path=ASIN/0262012103

09:31:39 <jeen> "A semantic Web primer"

09:31:54 <jeen> gives a very good introduction to the concepts behind web ontologies, both concepts and syntax.

09:32:08 * jeen uses the word 'concept' an awful lot :S

09:33:45 <etnaRosso> jeen it seems not very hard to use ontologies to create rdf documents

09:34:18 <etnaRosso> can i paste the definition of person in the foad spec?

09:34:19 <jeen> it isn't, but the learning curve can be steep :)

09:35:26 <etnaRosso> do you think i should improve the my rdf/xml's knoledge first?

09:39:13 <chaals> well, that is a good idea to understand RDF that you read.

09:39:25 <antoin> hello world

09:39:37 <etnaRosso> hello antoin

09:39:42 <antoin> i use Image magick and i have problem with IPTC

09:39:51 <chaals> Even before that you need to understand that what RDF describes is a set of relations - the syntax is not the most important thing.

09:39:56 <antoin> I want to write an IPTC field in jpg

09:40:02 <antoin> and that don't works

09:40:11 <antoin> somebody can help me??

09:40:46 <antoin> sorry for my english but i'm french

09:42:05 <jeen> antoin, I think you are in the wrong #swig. this is the Semantic Web Interest Group.

09:42:35 <CaptSolo> etnaRosso: what i used before (the two lines ending with '.') was a valid and commonly used way to express RDF

09:42:58 <CaptSolo> as you see, XML is not needed to represent RDF.

09:43:29 <etnaRosso> Solo you are the Millenium Falcon Commander?

09:43:45 <CaptSolo> That's why people here suggest to learn RDF before concentrating just on the XML representation

09:43:59 * CaptSolo looks outside to see where's his Falcon parked.

09:44:05 <CaptSolo> etnaRosso: happens so :)

09:44:10 <etnaRosso> ;)

09:44:26 <etnaRosso> i prefeer use xml so i can use rdf validator

09:45:29 <CaptSolo> install a validator locally that can validate NTriples (the language i used here)

09:46:12 <CaptSolo> CWM (from W3C) or Rapper (from Redland RDF framework) should do

09:46:42 <antoin> ok so sorry jeen....i'm log out

09:46:42 <etnaRosso> capt i like xml

09:46:44 <jeen> also: you should remember that an RDF validator does not actually validate very much. it checks the syntax and the graph structure, but not much else.

09:46:50 <jeen> no problem antoin :)

09:47:14 <CaptSolo> etnaRosso: ok, but don't fall into trap thinking RDF is XML

09:47:39 <jeen> so it won't give you errors if you don't use the "correct" ontology, for example.

09:47:51 <CaptSolo> hi libby :)

09:47:54 <jeen> hi libby

09:48:43 <libby> morning all :)

09:49:02 <chaals> libby!

09:49:28 <dajobe> hello

09:49:40 <dajobe> . o O (maybe we should ask the rdf validator to be renamed to rdf/xml validator...)

09:53:02 <libby> did you find morten chaals?

09:53:18 <chaals> nope...

09:53:25 <libby> boo

09:53:32 <chaals> I was too tired to go out anyway.

09:53:35 <chaals> :-(

09:53:37 <libby> :(

09:53:47 <libby> no! chaals! you ill ;)

09:54:00 * jeen goes for coffee

09:55:09 <Yudai_mm> Yudai_mm is now known as Yudai_mob

09:56:52 <bjoern_> .g foaf rdf pr0n

09:56:54 <phenny> foaf rdf pr0n: http://books.slashdot.org/books/03/09/18/0410238.shtml

09:56:56 <bjoern_> .img foaf rdf pr0n

09:56:59 <phenny> foaf rdf pr0n: http://www.kanzaki.com/docs/sw/foaf-simple.gif

09:57:13 <etnaRosso> lol

09:58:56 <bjoern_> .gc "Reality Distortion Field"

09:58:59 <phenny> "Reality Distortion Field": 31,900

09:59:40 <chaals> well, I stayed at home and had a chat...

10:00:00 <etnaRosso> good

10:01:21 <etnaRosso> Capitain Solo

10:01:30 <etnaRosso> so we have and rdf abstract sintax?

10:03:00 <CaptSolo> yep

10:03:19 <CaptSolo> though i would really start with RDF Primer

10:03:28 <CaptSolo>http://www.w3.org/TR/2004/REC-rdf-primer-20040210/

10:03:28 <dc_swig> B: http://www.w3.org/TR/2004/REC-rdf-primer-20040210/ from CaptSolo

10:04:39 <etnaRosso> but usign rdf description i make an assertion?

10:05:28 <CaptSolo> damn, now i have to name it

10:05:39 <CaptSolo> B:|RDF Primer

10:05:39 <dc_swig> Titled item B.

10:06:46 <CaptSolo> B:repetitio mater studiorum est

10:06:47 <dc_swig> Added comment B1.

10:07:20 <etnaRosso> i don't think so

10:07:44 <CaptSolo> about what?

10:08:01 <etnaRosso> repetitio mater studiorum est

10:08:11 <CaptSolo> why not?

10:08:50 <etnaRosso> cause devices had to contain data

10:08:56 <CaptSolo> you make an assertion by forming a triple consisting of a subject, property and object (propery value)

10:09:12 <etnaRosso> good

10:09:20 <etnaRosso> you aare very nice person

10:09:29 <libby> he is indeed

10:09:29 <kjetilkWork> danja has a recent blog entry with pointers to docs:

10:09:34 <kjetilkWork>http://dannyayers.com/archives/2005/08/01/rdf-tutorial/

10:09:34 <dc_swig> C: http://dannyayers.com/archives/2005/08/01/rdf-tutorial/ from kjetilkWork

10:09:54 <kjetilkWork> C:| Danja blogs about (lack of) RDF tutorials

10:09:54 <dc_swig> Titled item C.

10:10:49 <CaptSolo> :) thanks - just trying to help

10:11:15 <etnaRosso> leyt us examine http://www.w3.org/1999/02/22-rdf-syntax-ns

10:12:43 <etnaRosso> to create an assertion we need a subject, a predicate and an object

10:13:50 <CaptSolo> correct

10:14:35 <etnaRosso> subject is a property, the domain is an rdf statement

10:14:54 <etnaRosso> the range an rdf-schema#Resource

10:15:33 <CaptSolo> i won't be able to go into more details now - got some things to do

10:15:45 <jeen> ah, but that is not what you use for creating RDF statements, normally.

10:15:53 <CaptSolo> you may continue exploring it 'from the roots' - as you are doing now

10:16:07 <jeen> to create an rdf statement (in xml) you do the following:

10:16:13 <CaptSolo> yep - jeen has the point :)

10:16:30 <jeen> <rdf:Description rdf:about="somesubject">

10:16:48 <jeen> <somePredicate rdf:resource="someObject"/>

10:16:54 <jeen> </rdf:Description>

10:17:10 <etnaRosso> great

10:17:16 <CaptSolo> etnaRosso: to use it (as RDF is pretty straighforward) you don't need to explore RDF ontology in depth

10:17:24 <CaptSolo> etnaRosso: just use it :)

10:17:36 <CaptSolo> etnaRosso: but you may explore it when you get bored

10:17:45 <jeen> the properties 'rdf:subject', 'rdf:predicate' etc. are for creating reified statements. I wouldn't bother with that just yet if I were you :)

10:17:47 * CaptSolo wonders if he said something wrong now :)

10:18:02 <etnaRosso> suppose that the subject is a foaf person

10:19:35 <jeen> right, so you want to express that 'person1' is of type foaf:Person. You do that like this:

10:19:45 <jeen> <rdf:Description rdf:about="#person1">

10:19:58 <jeen> <rdf:type rdf:resource="&foaf;Person"/>

10:20:03 <jeen> </rdf:Description>

10:21:01 * jeen uses the &foaf; entity as a shorthand, normally you should write down the complete URI for the FOAF ontology in that place.

10:21:23 <jeen> or, you can write it down in abbreviated form:

10:21:31 <jeen> <foaf:Person rdf:about="#person1"/>

10:21:54 <jeen> both syntaxes encode the same RDF statement.

10:47:07 <etnaRosso> did someone know where i can find the says predicate?

10:48:02 <leasa> whay do you meen by the says predicate?

10:49:28 <etnaRosso> a person that make an assertion

10:51:11 <danbri> i sometimes use foaf:made as a relation between an agent and an rdf document

10:51:19 <danbri> different granularity...

10:51:39 <etnaRosso> made?

10:52:14 <etnaRosso> it sounds as 'creation'

10:52:51 <danbri> yes, its similar to dc:creator

10:53:03 <etnaRosso> dc

10:53:11 <danbri> but there's currently too much variation in ways to use dc:creator (though that's maybe being fixed...)

10:54:21 <etnaRosso> Do i have to search in dublin core for web ontologies?

10:55:00 <danbri> nothing's compulsory

10:55:11 <danbri> you might look at the dc terms to see if they're of use to you

10:55:13 <danbri> sorry, gotta run

11:00:21 <etnaRosso> byes

11:02:14 <mutt> mutt is now known as swh

12:51:53 <etnaRosso> hi all

13:44:11 <Pike_> I'm thinking about a backwards-compat trick to fix the Mozilla bug about using rdf:type instead of rdf:instanceOf. One I just came up with would be to return the rdf:type resource if queried for rdf:instanceOf from the RDF service

13:44:20 <Pike_> anyone feeling like commenting on that?

13:45:39 <crschmidt> What's the difference in meaning... oh, Mozilla looks for instanceOf first?

13:46:53 <Pike_> crschmidt: it looks for instanceOf only right now

13:47:08 * crschmidt nods

13:47:20 <crschmidt> ^q select ?p where (?p rdf:instanceOf ?o)

13:47:21 <julie> Query returned no results

13:48:07 <Pike_> the problem I face is, extensions use the instanceOf resource explicitly to hack stuff

13:48:44 <Pike_> so I can break them, or think about a way to be somewhat backwards compatible. Without being able to actually fix the datasources in general

13:49:16 * crschmidt nods

13:51:31 <Pike_> are there sematic differences between instanceOf and type? i.e., is there something I'd break if both were the same resource?

13:51:58 <Pike_> apart from folks actually using rdf:type or rdf:instanceOf as source for some of their arcs

13:52:27 <crschmidt> I have never seen instanceOf before, but looking at old documentation, it seemse that type and instanceOf have the same semantic meaning as a predicate

13:53:25 <Pike_> crschmidt: yeah, instanceOf is old, that's why it's in the mozilla code :-/

13:54:21 * crschmidt nods. This is what you get for being on the front of a technology wave, sadly

13:54:23 <danbri> yup, guha implimented to the older rdf draft

13:54:35 <danbri> 98ish

13:55:37 <Pike_> would there be any usecase for having arcs off of rdf:type?

13:56:01 <danbri> eg?

13:57:53 <Pike_> rdf:type -> nc:foo -> "ugly"

13:59:15 <sbp> rdf:type rdf:type rdf:Property .

14:00:55 <chimezie> rdf:type owl:equivalentTo dc:type

14:01:06 <chimezie> er.. thats not right

14:01:11 <chimezie> the other way around

14:01:37 <chimezie> althogh is irrelevant w/ owl:equivalentTo

14:03:13 <Pike_> still looks like that would be equivalent to instanceOf being the subject

14:03:22 <Pike_> but this is fishy, at best

14:04:35 <crschmidt> it's useful only in a very small subset of cases, and is probably not a concern at all if you're doing an API change: You don't need to replace subject searches for rdf:instanceOf with rdf:type, becuase people in their extensions have probably never looked for something with a subject of rdf:instanceOf

14:07:15 <Pike_> I don't trust our extension authors

14:07:25 <Pike_> they know jack about RDF in general

21:46:58 <[aniasis]> [aniasis] is now known as aniasis

22:52:38 <temp_> temp_ is now known as wwhateley

23:34:19 <ndw> ndw is now known as ndw|afk