Semantic Web Interest Group IRC Chat Logs for 2003-10-25

00:30:27 <eaon|bughunt> eaon|bughunt is now known as eaon|zZz

01:21:22 <dmwaters> {global notice} Hi all! I'm going to do some major rehubbing of the network in about 45 mins. Our main european hub will go down for maintenence later this morning,and we need to work around it. Sense this does effect almost the entire network, I wanted to give people heads up. I'll give progress messages in wallops, (/mode your_nick +w) to turn them on.

02:07:34 <ChanServ> [#rdfig] This channel is logged and blogged: http://logicerror.com/rdfIRCWelcome

02:08:05 <ChanServ> [#rdfig] This channel is logged and blogged: http://logicerror.com/rdfIRCWelcome

02:17:13 <ChanServ> [#rdfig] This channel is logged and blogged: http://logicerror.com/rdfIRCWelcome

10:11:58 <arnarl> hi

10:33:16 <Arnia22> Arnia22 is now known as Arnia

10:47:09 <eaon|zZz> eaon|zZz is now known as eaon

12:35:36 <Dorward> Dorward is now known as dorwAFK

12:35:45 <zool> i'm having problems getting to http://www.w3.org/2002/07/owl and hte directory below it asks for authentication

12:35:56 <zool> anyone know what is up with that?

12:42:47 <libby> ait jusrt downloaded ok for me

12:42:52 <libby> ait jusrt downloaded ok for me

12:42:58 <libby> arrgh

12:47:53 * zool frowns okay

12:48:02 <zool> i'm just wondering if i'm mad to say:

12:48:05 <zool> <owl:Property rdf:about="http://www.w3.org/2002/07/owl#Property" rdf:parseType="

12:48:05 <zool> Resource">

12:48:05 <zool> <rdfs:domain rdf:resource="http://www.w3.org/2002/07/owl#Class"/>

12:48:05 <zool> </owl:Property>

12:48:31 <zool> and also say <owl:Class rdf:about="http://www.w3.org/2002/07/owl#Property">

12:48:33 <zool> etc

12:48:41 <libby> I could email it to you?

12:48:52 * libby doesn't know anough about owl

12:48:55 <zool> that would be cool, thankyou!

12:48:59 <libby> np :)

12:49:05 <zool> well s/owl/rdfs/ there also

12:49:14 <zool> owl seems shinier though

13:41:47 * eaon is away: 10 mins

13:44:05 <dorwAFK> dorwAFK is now known as Dorward

13:50:25 <DanC> no, zool, I don't think { owl:Property rdfs:domain owl:Class } makes sense.

13:50:49 <DanC> What I think you mean is: { ?P a owl:Property } => { ?P rdfs:domain owl:Class }

13:51:22 <DanC> aka { owl:Property rdfs:subClassOf [ owl:onProperty rdfs:domain; owl:hasValue owl:Class ] }.

13:53:45 <zool> well, i should show you the context in which i'm doing it

13:55:07 <zool>http://mutemap.openmute.org/onto.rdf

13:55:08 <dc_rdfig> A: http://mutemap.openmute.org/onto.rdf from zool

13:55:14 <zool> oh balls

13:55:19 * eaon is back (gone 00:13:30)

13:55:39 <zool> well from that file model, i am using it to generate an interface to create models

13:55:47 <zool> s/file//

13:56:59 <zool> (e.g. you'd give it a model consituting bits of several ontologies, with annotations, and that determines what classes of thnig and their properties a bot knows about, or a html forms interface knows how to draw

13:57:34 <zool> so i thought i'd use the same interface so people can create an ontology to power another model

13:57:55 <zool> (that file may sitll have rerors in it i havent checked)

13:58:27 <zool> so to rewind, in this scheme i want to be able to create things of class Property, but i also want classes to be able to have the proprty property

13:59:03 * zool stars hard at danc's notation

13:59:07 <zool> e

13:59:35 <zool> and ?P happens to be owl:Property

13:59:45 <zool> gosh

13:59:55 <zool> okay i'm not quite there yet

14:01:19 <zool> hm it doesnt like my partial-understanding use of parseType either

14:02:25 <eaon> eaon is now known as eaon|lunch

14:07:39 <WillWare> Are there any vocabularies for geneology? I've been trying to put one together.

14:08:20 <zool> there's a foad family one somewhere, can't remember author sorry. maybe eikeon?

14:08:49 <libby> yeah eikeon has some data

14:11:29 * libby thinks LTJake is doin g some stuff too, for gedcom. he's in #foaf, though quiet at moment

14:11:50 * eikeon waves, hi libby zool

14:12:45 <libby> heya!

14:22:19 <WillWare> Here's what I've got for my geneology stuff so far.

14:22:27 <WillWare> Classes and properties:

14:22:27 <WillWare> Person (rdfs:subClassOf foaf:Person)

14:22:27 <WillWare> // why not just use foaf:Person??

14:22:27 <WillWare> gender -> literal ("male" or "female")

14:22:27 <WillWare> born -> Event

14:22:28 <WillWare> died -> Event

14:22:30 <WillWare> moved -> Event

14:22:32 <WillWare> Location

14:22:34 <WillWare> Use the chefmoz stuff to fill in details (street,

14:22:36 <WillWare> city, state, lat, long, ...)

14:22:38 <WillWare> Couple

14:22:40 <WillWare> person1 -> Person

14:22:42 <WillWare> person2 -> Person

14:22:44 <WillWare> begin -> Event

14:22:46 <WillWare> end -> Event

14:22:48 <WillWare> Marriage (rdfs:subClass of Couple)

14:22:50 <WillWare> Event (in the relativistic sense, a point in 4-space)

14:22:52 <WillWare> where -> Location

14:22:54 <WillWare> when -> dc:date

14:23:12 <WillWare> Oops, forgot one more property, child: Couple->Person

14:23:14 <zool> hey eikeon

14:23:39 <libby> for when/where, you might l;ike to consider using rdfical and geo

14:24:23 <libby> e.g http://planb.nicecupoftea.org/archives/000074.html, http://swordfish.rdfweb.org/discovery/2003/09/foafcal/foafgeo.rdf

14:24:36 <libby> might not be appropriate though

14:25:24 <libby> geo for lat/long is a good plan, I think chefmoz use that anyway: http://www.w3.org/2003/01/geo/

14:25:40 * libby wonders if jibberjim knows what sort of output LTJake is getting

14:26:17 <JibberJim> see the foaf logs

14:26:26 <libby> yeah, checking

14:26:28 <libby> ta

14:26:52 <JibberJim> it seemed like he was using pixel's relationship schema

14:26:59 <JibberJim> and foaf

14:27:47 <JibberJim> We discussed an IFP for them too the other night, conclusion was giving them a URI wouldn't be disastrous...

14:27:48 <libby> I think this is an example willware: http://red13.alternation.net/gedcom/I0009/foaf

14:28:01 <libby> them? people?

14:28:12 <JibberJim> yes.

14:28:24 <libby> e.g. dead people?

14:28:43 <libby> (I'm thinkng of tthe chet baker discsussion on rdfweb-dev)

14:29:02 <JibberJim> yes, about 4 nights, I'd search, but...

14:29:27 <libby> s'ok, thanks

14:30:08 * libby gathers a few links together

14:30:25 <JibberJim> see http://ilrt.org/discovery/chatlogs/foaf/2003-10-21.html

14:30:45 <libby> BLURB:geneology RDF schemas and discussion

14:30:46 <dc_rdfig> B: geneology RDF schemas and discussion from libby

14:30:48 <libby> ta

14:32:05 <libby> B:[http://ilrt.org/discovery/chatlogs/foaf/2003-10-21.html#T00-18-04|discussion in #foaf about IDs for people and foafnaut]

14:32:06 <dc_rdfig> Added comment B1.

14:33:10 <libby> B:[http://red13.alternation.net/gedcom/I0009/foaf|LTJake's sample output]

14:33:10 <dc_rdfig> Added comment B2.

14:33:37 <darobin_> darobin_ is now known as darobin

14:33:50 <libby> B:[http://ilrt.org/discovery/chatlogs/rdfig/2003-10-25.html#T14-07-39|willware's model]

14:33:50 <dc_rdfig> Added comment B3.

14:34:01 <WillWare> B:[http://willware.net:8080/wares.rdf wware's sample output]

14:34:02 <dc_rdfig> Added comment B4.

14:34:42 <WillWare> B4:[http://willware.net:8080/wares.rdf|wware's sample output]

14:34:42 <dc_rdfig> Replaced comment B4.

14:34:51 <libby> neat :)

14:35:04 * libby looking for pixel's schema, can;t find it...

14:36:10 <JibberJim> Wil

14:36:40 <JibberJim> Will's parent's can't be identified outside that file though...

14:37:12 <WillWare> My parents are so hopelessly computer illiterate that that's not so much of a problem

14:39:37 <JibberJim> It is, because when your cousin wants to talk about their uncle being the same, then there's no way of doing it.

14:39:42 <libby> B:note: [http://rdfschema.info/2003/properties_by_popularity|a very good place to look for schemas]

14:39:42 <dc_rdfig> Added comment B5.

14:40:20 <WillWare> So I need one of those inverseUniqueProperty things?

14:40:58 <eaon|lunch> eaon|lunch is now known as eaon

14:41:01 <libby> ahahh

14:41:10 <JibberJim> IFP's, but there aren't really appropriate ones.

14:41:19 <eikeon> Wonder what we could do to get more people to assert their vocabularies as owl:Ontology

14:41:21 <JibberJim> unless maybe you give them a homepage.

14:41:37 <libby> B:[pixel's relationship schema|http://www.perceive.net/schemas/20020729/relationship/relationship.rdf] covers many relationships; used by LTJake

14:41:38 <dc_rdfig> Added comment B6.

14:41:42 <WillWare> Oh they don't have a URI! OK, now I get it.

14:42:01 <WillWare> Isn't willware.net:8080/wares.rdf#willsDad a URI, though?

14:42:11 <mdupont> mdupont is now known as md-afk

14:42:12 <WillWare> (albeit a lame one)

14:42:40 <eikeon> I guess I could start by expanding the example at: http://rdfschema.info/2003/submit

14:44:53 <WillWare> I made my thing an owl:Ontology because I shamelessly copied most of it off FOAF.

14:45:06 <libby> heh

14:45:13 <libby> its a useful way of picking them out

14:46:45 <eikeon> WillWare: What's the url of your owl:Ontology ?

14:46:59 <libby> is there a fewwtext search eikeon, a la http://swordfish.rdfweb.org/discovery/2001/03/schemas/ - this old example?

14:47:00 <WillWare>http://willware.net:8080/family.rdf

14:47:00 <dc_rdfig> C: http://willware.net:8080/family.rdf from WillWare

14:47:22 <WillWare> C:|Will's geneology vocabulary

14:47:22 <dc_rdfig> Titled item C.

14:48:06 <WillWare> Gee, I forgot to put 'child' there too.

14:48:10 <libby> s/fewwtext/freetext/

14:50:24 <WillWare> ok, updated the vocab to include 'child'

14:50:35 <eikeon> libby: Not yet. Would be a good addition

14:53:04 <WillWare> Duty calls, gotta go. Thanks Libby JibberJim et al

14:53:58 <eikeon> Upps... I didn't type fast enough...

14:54:12 <eikeon> was just pointing out that WillWare's rdf had an error in it.

14:54:53 <eikeon> At 46:2: Can have at more one of rdf:ID, rdf:about, rdf:nodeID

14:55:03 <eikeon> s/more/most/

14:55:05 <libby> maybe he'll read logs

14:55:18 <libby> do you still crawl rdf eikeon?

14:55:27 <eikeon> Yep.

14:55:38 <libby> how often?

14:55:51 <libby> how many triples are you getting (roughly)?

14:56:16 <eikeon> IIRC, once per day.

14:56:23 <libby> right

14:56:45 <eikeon> Not sure how many... I should work on sites some today. I was curious the same thing the other day.

14:57:03 <eikeon> I did notice there are some 7000 contexts now.

14:57:11 <libby> wow

14:57:17 <libby> I'll check mine

14:57:34 <libby> I've got about 350,000 triples, but not crawling regularly

14:57:45 <libby> matrtb found about 750,000 recently

14:57:50 <libby> mattb even

14:57:55 <JimJibber> JimJibber is now known as JibberJim

14:58:06 <eikeon> I'm not sure if the 7000 is counting the 404s too or not.

14:58:32 <eikeon> Also... would be nice to get a handle on how many of them are foaf vs other documents.

14:59:24 <libby> yeah

14:59:35 <libby> mysql> select count(distinct predicate) from triples;

14:59:35 <libby> +---------------------------+

14:59:35 <libby> | count(distinct predicate) |

14:59:35 <libby> +---------------------------+

14:59:35 <libby> | 1078 |

14:59:36 <libby> +---------------------------+

14:59:42 <libby> not so many...

15:03:29 <JibberJim> yeah, which is why I was thinking of not using hash for predicate.

15:04:14 <eikeon> Looks like a fair # of my triples are from: http://www.mindswap.org/2003/nciOncology.owl

15:06:21 <libby> ble, I'm a dork, meant to look for source not predicate

15:07:54 <WillWare> Yup, I do sometimes read logs. Thanks eikeon, I was unsure about that.

15:10:08 <libby> I've got 4143 sources.

15:11:09 <WillWare> C: [http:http://willware.net:8080/cgi-bin/viewcvs.cgi/Jena-2.0/src/wware/Foo.java | Jena test code for geneology stuff]

15:11:09 <dc_rdfig> Added comment C1.

15:11:21 <eikeon> WillWare: You're welcome

15:13:18 <eikeon> libby: I'll need to implement a better len function for the triple store backend that I'm using -- currently it computes the length by iterating over the whole store ;(

15:13:44 <eikeon> Was hoping I'd already done that.

15:14:26 <libby> aw

15:14:27 <dngor> @count () (is a type of) (verb)

15:14:29 <knowp> Your query matched 618 factoids

15:14:52 * libby trying to fix scutter. last time it cashed with an out of memory error.

15:15:59 <WillWare> C1: [http://willware.net:8080/cgi-bin/viewcvs.cgi/Jena-2.0/src/wware/Foo.java | Jena test code for geneology stuff]

15:16:00 <dc_rdfig> Replaced comment C1.

15:16:29 <WillWare> C1: [http://willware.net:8080/cgi-bin/viewcvs.cgi/Jena-2.0/src/wware/Foo.java| Jena test code for geneology stuff]

15:16:29 <dc_rdfig> Replaced comment C1.

15:27:37 <darobin> has anyone done something RDF(ish) around horse races?

15:28:55 <libby> heh

15:29:29 <darobin> I'm considering it :)

15:29:38 <libby> neat!

15:30:20 <darobin> enter horses that got what position at what race being how long with what jokey and what weather

15:30:41 <darobin> then some inference rules, and enter details for a race and get a prediction back

15:30:57 <libby> cute :)

15:31:08 * libby looks forward to it!

15:31:08 <darobin> dunno, maybe I'll get rich ;)

15:31:39 <darobin> I have a friend that's telling me what aspects are important, but I need to find data...

15:36:33 <libby> B:might be worth using [http://www.w3.org/2002/12/cal|RDFical schema for events] and [http://www.w3.org/2003/01/geo/|geo vocab for lat/long]

15:36:33 <dc_rdfig> Added comment B7.

15:59:25 <md-afk> md-afk is now known as mdupont

16:53:54 <DanC> woohoo! debian supports nxml-mode. http://packages.debian.org/unstable/editors/nxml-mode.html

17:14:24 <dajobe> DanC: yeah, it's been there for a couple of weeks.

17:14:51 <dajobe> it also comes with rdf/xml RNC from the syntax spec, but not enabled by default for '.rdf' files

18:00:36 <golbeckSanibel_> golbeckSanibel_ is now known as golbeckSanibel

18:36:44 <libby> hi LotR, urgen :)

18:36:53 <urgen> :-)

18:36:56 <LotR> hi libby

18:36:56 * libby just off out, unfortuntely

18:37:02 <LotR> heh

18:37:02 <urgen> that's ok

18:37:10 <libby> l8r!

23:09:48 <D[a]vey> D[a]vey is now known as Davey

23:22:20 <deusx> deusx is now known as _deusx