Friend of a Friend (FOAF) IRC Chat Logs for 2003-09-11

00:36:45 <JibJimMob> Hi

00:36:54 <JibberJim> Hi

00:37:07 * JibberJim talks to himself

00:51:51 <JibJimMob> :-)

01:01:34 * JibJimMob waves

09:04:08 <Jibbler> Over the last hundred frames or so, this drawing hasn't changed a noticable amount, so i guess it has settled into a vaguely true respresentation of this channel - http://www.cs.kent.ac.uk/people/rpg/pjm2/temp/frame0526.png

09:31:04 <darobin> nice!

09:31:26 <darobin> Jibbler: considered svg for this?

09:31:54 <Jibbler> darobin: i have, yes

09:31:58 <Jibbler> moreso than before

09:32:25 <darobin> it should be fairly straightforward to do the same, and then add a little interactivity

09:32:32 <Jibbler> I suppose SVGGraphics2D from Batik is the easiest way

09:33:08 <Jibbler> i am an SVG newbie.. i can make some hand-written stuff, but the thing that put me off it is lack of decent viewers

09:33:22 <darobin> asv is decent

09:33:34 * mattb obmutters about linux support

09:33:44 <darobin> for rendering to a png yeah batik will be best

09:34:00 <Jibbler> no, i meant just to make the SVG in the first place

09:34:03 <darobin> mattb: asv6 is "very very likely" to be on linux

09:34:08 <darobin> oh

09:34:13 <darobin> use any XML tool ;)

09:34:19 <Jibbler> cos i'm drawing everything on a Graphics2D object (or subclass thereof)

09:34:21 <darobin> or sodipodi

09:34:22 * mattb happy to hear :)

09:34:29 * darobin too

09:34:32 <Jibbler> i know a fair few people that don't have SVG viewing capability "out of the box" so i'm reluctant to use it

09:34:39 <Jibbler> i know practically everyone can see png

09:34:42 <mattb> "

09:34:42 <mattb> The Graphics2D SVG Generator is a Java class file that enables Java applications to export graphics to Scalable Vector Graphics (SVG) format. It requires Java 2, version 1.3 release candidate 3 or better. "

09:34:54 <mattb> says http://www.sun.com/xml/developers/2d-svg/

09:35:02 <darobin> Jibbler: ah ok, didn't know that. I guess I could've guessed from the colour scheme. Well using Batik should be doable for that yes

09:35:29 <mattb> hmm, but that page is gone, replaced by a 302 to apache batik

09:35:36 <darobin> Jibbler: thing is, you can support both with pretty much the same code

09:35:43 <Jibbler> yes, that was my plan. like i use EpsGraphics2D (which extends Graphics2D) to draw it as EPS, i was expecting to just stick an SVGGraphics2D object in its place

09:35:54 <darobin> yup

09:35:59 <mattb> aha, http://xml.apache.org/batik/svggen.html

09:36:01 <Jibbler> darobin: guessed from the colour scheme? eh? :)

09:36:51 <Jibbler> not sure what you meant by that

09:36:52 <darobin> well those shades of light purple (which I like in this instance mind) seem to be what always come out of java graphic tools, no matter what you input :)

09:36:57 <Jibbler> lol!

09:37:04 <Jibbler> i chose all those colours myself

09:37:09 <darobin> hehehe

09:37:19 <Jibbler> it's blue, technically

09:37:27 <Jibbler> i think

09:37:34 <Jibbler> with an alpha channel

09:37:56 <darobin> ah well, it looks lightly purplish to me, but then my eyesight isn't all that good

09:38:48 <Jibbler> hmm, where are the javadocs for SVGGraphics2D lurking...

09:39:20 * danbri joins http://groups.yahoo.com/group/FriendsterRevolution/ mailing list

09:40:01 <mattb> at http://xml.apache.org/batik/javadoc/org/apache/batik/svggen/package-summary.html ?

09:41:52 <Jibbler> looks like a good bet :)

09:42:18 <Jibbler> interesting to see they've made an abstract sublass between Graphics2D and SVGGraphics2D

09:42:57 <Jibbler> other than that, it all looks pretty similar to EpsGraphics2D

09:43:08 <Jibbler> except it actually implements everything

09:44:42 <Jibbler> anyway, something to play with later

09:44:53 <Mutiny> hehe friendster

09:45:12 <Mutiny> "FOAF; less dating oriented, so your girlfriend won't flip out"

09:47:03 <libby> as a friend pointed out last night, we don;t even have gender, so s/less/not/. for now....

09:47:20 <danbri> where'd it say that?

09:47:59 <danbri> should the rdfs:domain of foaf:gender be Person or Agent?

09:48:15 <danbri> we don't have Mammal or LivingThing yet, leave such details to Cyc and TAP and Wordnet...

09:48:20 <danbri> Agent I think

09:48:30 * danbri remembers Bender from Futurama as a use case

09:48:38 <libby> to leave room for robot love

09:48:45 <libby> or Data

09:48:50 <libby> (fully functional)

09:49:00 <libby> is a group and agent??

09:49:08 <libby> a/and/an/

09:49:17 * danbri adding 'foaf:gender'

09:49:29 <danbri> all those opposed speak out now

09:49:33 <danbri> you have 10 seconds

09:49:34 <danbri> 9

09:49:34 <danbri> 8

09:49:34 <danbri> 7

09:49:35 <danbri> 6

09:49:35 <danbri> 5

09:49:36 <danbri> 4

09:49:37 <danbri> 3

09:49:38 <mattb> hehe

09:49:39 <danbri> 2

09:49:44 <mattb> presumably a literal?

09:49:44 <danbri> 1.5

09:49:48 <danbri> yes.

09:49:58 * mattb would object to controlled vocab

09:49:59 <danbri> darn, now someone's gonna object and demand URIs

09:50:04 <Jibbler> options these days would have to include male, female, neither and both

09:50:06 * danbri too

09:50:19 <danbri> "neither" is simple ommission, more or less.

09:50:26 * mattb remembers reading of medical databases and such that have something like 7 genders

09:50:32 <danbri> oh, interesting, do we make it a functional property?

09:50:35 <mattb> wish i could find something to cite on that

09:50:42 <danbri> whatever one's gender is, one has just one of it.

09:50:54 <danbri> that seems more reasonable than policing its values

09:51:03 <danbri> hmm do medical dbs call it 'gender' or 'sex'

09:51:06 <libby> not nec over time though

09:51:12 <libby> sex change?

09:51:29 <libby> we can;t call it 'sex'

09:51:30 <mattb> there are transitional states

09:52:35 <danbri> rdfs:comment="The gender of this Agent (typically but not necessarily 'male' or 'female')."

09:52:40 <danbri> change is different

09:52:44 <Mutiny> "was a male, now female" "was a female, now male" "was a male, now.. other"

09:52:47 <danbri> most foaf properties can change.

09:54:11 <danbri> [[

09:54:12 <danbri> <rdf:Property rdf:about="http://xmlns.com/foaf/0.1/gender" vs:term_status="testing"

09:54:12 <danbri> rdfs:label="gender"

09:54:12 <danbri> rdfs:comment="The gender of this Agent (typically but not necessarily 'male' or 'female').">

09:54:12 <danbri> <rdf:type rdf:resource="http://www.w3.org/2002/07/owl#FunctionalProperty"/>

09:54:12 <danbri> <rdfs:domain rdf:resource="http://xmlns.com/foaf/0.1/Agent"/>

09:54:14 <danbri> <rdfs:range rdf:resource="http://www.w3.org/2000/01/rdf-schema#Literal"/>

09:54:16 <danbri> <rdfs:isDefinedBy rdf:resource="http://xmlns.com/foaf/0.1/"/>

09:54:18 <danbri> <!-- whatever one's gender is, and we are liberal in leaving room for more options

09:54:20 <danbri> than 'male' and 'female', we model this so that an agent has only one gender. -->

09:54:22 <danbri> </rdf:Property>

09:54:26 <danbri> ]]

09:54:28 <danbri> that look ok?

09:55:05 <mattb> think so

09:55:17 <libby> danbri, this means a group can have a gender

09:55:27 <danbri> danbri@fireball:~/s-foaf$ rapper file:index.rdf

09:55:27 <danbri> rapper: relocation error: rapper: undefined symbol: raptor_version_string

09:55:27 <mattb> in cases like this where the vocab is not controlled, but largely expected to come from a fixed list, would you define a normalisation?

09:55:27 <libby> [[

09:55:28 <libby> The foaf:Group class represents a collection of individual agents (and may itself play the role of a foaf:Agent, ie. something that can perform actions).

09:55:28 <libby> ]]

09:55:33 <mattb> {Male,m,man} => "male", etc

09:55:35 <danbri> hmm, debian'd redland rapper is hosed.

09:55:55 <danbri> normalisation: in prose, at least

09:56:05 <mattb> danbri: rapper works for me, what version?

09:56:09 * mattb just updated this morning

09:56:32 <danbri> libby, yup good point. However just because a property is applicable in general, doesn't mean it is always applicable.

09:56:43 <danbri> Some but not all Agents have genders

09:57:09 <Jibbler> do foaf subjects *have* to be humans?

09:57:11 <danbri> the rdfs:domain on foaf:gender of foaf:Agent means just this and no more: If you see ?x foaf:gender foo, you can infer ?x is an Agent.

09:57:31 <Jibbler> i mean, can my cat have a foaf file?

09:57:40 * danbri agrees it is a little awkward and that a more suitable, squishy biological class would make sense

09:57:54 <mattb> jibbler: people have also asked if bots can have a foaf file

09:58:00 <Jibbler> haha

09:58:10 <mattb> social network of bots certainly a useful thing

09:58:10 <Mutiny> hm.

09:58:13 <mattb> if one bot uses another's services

09:58:15 <Jibbler> are there any written rules/suggestions on that matter?

09:58:33 <Mutiny> having gender be literal i can imagine would be tricky. :/

10:00:36 <mattb> otoh do you want to get into the debate on classification of genders into URIs?

10:01:19 <danbri> bots -- sure

10:01:34 <Jibbler> in a way, a bot is backed by a person or persons

10:01:46 <Mutiny> i think having 'male', 'female', and 'other' would be sufficient.

10:01:50 <Jibbler> much like a bot is a form of irc client run by a user

10:01:53 <danbri> there may be restricted types of foaf file where the person who invents that restricted subset says 'no fakes', but in general foaf files can describe anything

10:01:57 <danbri> foaf supports lying

10:02:05 <Jibbler> my bots are male :)

10:03:35 <darobin> "Tonight Lois McDermott, UW psychology professor, will surprise most of her audience with the theory that there are five sexes, seven genders and 400 sexual ... " -- http://archives.thedaily.washington.edu/ 1997/042997/sex.042997.html

10:03:42 <darobin> unfortunately, host unreachable

10:04:59 <Mutiny> google cache has it: http://google.com/search?&q=cache:http://archives.thedaily.washington.edu/1997/042997/sex.042997.html

10:07:10 * darobin slaps self

10:07:21 <darobin> ENEEDCOFFEE

10:07:28 <Mutiny> in my experience if you give a literal field for them to put their gender all it really does it make it hard for you to find the males and females

10:11:25 <danbri> the data format doesn't determine UI. A combo box or dropdown would do for many sites.

10:12:11 <danbri> "In her lecture she will explain that five different sexes exist: the congruent female, the congruent male and the inter-sex condition, including the hermaphrodite and two sexes defined by genes that do not match their visible traits. She will also explain the many genders: the feminine, masculine, androgynous, transsexual, cross-dresser and culturally specific genders such as eunuchs."

10:12:16 <danbri> hmm

10:12:47 <danbri> I don't see 'cross dresser' as a gender, but maybe i'm some kind of bigot...

10:13:09 <danbri> anyway, foaf:gender should get some debate...

10:13:25 <Jibbler> let's not hope it gets mass debate (boom boom)

10:14:18 <darobin> except she only cites six of the supposed seven

10:14:58 <Mutiny> yeak leaving it up to the programmer who decides the ui isn't any better. like he said somebody over here calls it 'm' somebody calls it 'guy' somebody calls it 'male' etc

10:17:47 <libby> problem is gender is cultural

10:17:49 <libby> right?

10:18:19 <libby> or socially contructed or something

10:18:40 * libby vaguely remembers something about this stuff from a mate from social geography

10:19:12 <Mutiny> maybe 'sex' would be a more appropriate property name?

10:19:17 <libby> anyway it's not physical sexual characteristics

10:19:31 <libby> yeah, but people will just put 'yes (please)' ;)

10:19:39 <Mutiny> dictionary tells me that gender is cultural, while sex is just 'male, female'

10:19:58 <Mutiny> hehe even more reason not to have it just be a free literal ;)

10:20:13 <libby> heh. it's traditional.

10:21:12 <Mutiny> "This usage is supported by the practice of many anthropologists, who reserve sex for reference to biological categories, while using gender to refer to social or cultural categories."

10:21:31 <libby> blimey, tricky.

10:21:59 <JimH> I'm doing some Foaf house cleaning and I wonder if there's a "correct" way for me to refer to another Foaf file (an ecademy one) that is also me, suggestions?

10:22:12 <libby> hey JimH

10:22:18 <JimH> hi Libby

10:22:34 <danbri> "sex" is too clinical.

10:22:41 <libby> I guess the best we have is rdfs:seeAlso, which we use to refer to another machine-readable file

10:22:56 <JimH> that's what I was doing

10:23:06 <libby> there's no way of saying: this is another foaf file about me I dont think

10:23:22 <libby> alsthough good to say something about foaf:made at the top refering to yourself

10:23:26 <Mutiny> 'sex' and 'gender' seem two different things.

10:23:49 <darobin> yes

10:23:55 <darobin> but sex is definitely clinical

10:24:07 <danbri> [

10:24:09 <danbri> <p>

10:24:09 <danbri> Note that FOAF's notion of gender isn't defined biologically, which would be tricky since

10:24:09 <danbri> we have a broad notion that applies to all <code>foaf:Agent</code>s (including robots

10:24:09 <danbri> etc.). However FOAF's notion of gender doesn't attempt to encompass the full range of

10:24:09 <danbri> concepts associated with the human gender, biology and sexuality. As such it is a (perhaps

10:24:10 <danbri> awkward) compromise between the clinical and the social/psychological. In general, a Person

10:24:12 <danbri> will be the best authority on their <code>foaf:gender</code>. Feedback on this design is

10:24:14 <danbri> particularly welcome welcome (via the FOAF mailing list,

10:24:16 <danbri> <a href="http://rdfweb.org/mailman/listinfo/rdfweb-dev">rdfweb-dev</a>).

10:24:18 <danbri> </p>

10:24:20 <danbri> ]

10:24:22 <danbri> how's that work for you folks?

10:24:25 <darobin> and besides, most of the inter-sex people that are not hermaphrodites don't know about it because only DNA analysis reveals it

10:24:29 <libby> jimH - you could see if http://swordfish.rdfweb.org/discovery/2003/08/validation/ is any use to you

10:24:53 * danbri notes that the foaf namespace scripts wouldn't let him rebuild the foaf spec until he wrote the gender.en documentation file :)

10:25:01 <libby> :)

10:25:05 <darobin> danbri: that sounds good to me

10:25:12 <libby> it sounds good as a start for discussion

10:25:21 <libby> is the idea that we go for literal?

10:25:22 <darobin> it would seem that gender/sex is about as easy to define as address

10:25:27 <libby> heh

10:25:41 <danbri> I'm trying to be respectful of diversity without trying to catalogue / enumerate the options

10:26:05 <libby> "particularly welcome welcome"

10:26:10 <libby> :)

10:26:27 <JimH> gender sounds sane to me, is there a neuter choice? Or will it only be valid for foaf:Person?

10:26:56 <danbri> neuter ... for example?

10:26:58 <Mutiny> i don't know.. seems like if you did wanted to extend foaf into dating having a way to concretely define gender would be useful.

10:26:59 <danbri> eg. neuter'd cat?

10:27:17 <darobin> eunuch?

10:27:32 <darobin> no that doesn't fit

10:27:47 <JimH> or a 'bot i.e. a foaf:Agent that is not a person

10:28:06 <libby> it's for foaf:Agent JimH, which could incude groups, bots and whatnot

10:28:07 <Mutiny> how about a foaf:sex 'male' 'female' 'other' and a foaf:gender literal?

10:29:09 <darobin> can it be stated multiple times? I guess hermaphrodite would be fine for humans, but for a robot it might need to be "usb & pcmcia & bluetooth"

10:29:52 <danbri> Bender from Futurama is male.

10:29:58 <danbri> and an Agent.

10:30:06 <sbp> heh. you can just imagine robots in the pub. "she had *eight* Firewire 800 ports!"

10:30:09 <danbri> multiple: no , it is functional.

10:30:12 <darobin> lol

10:31:56 <Mutiny> an "equipment" property would apply equally well for Agent and Person

10:31:58 <Mutiny> ;)

10:32:02 <darobin> hehehe

10:32:26 <darobin> it seems we're fast going to define the Web Sex Description Language

10:32:34 <sbp> heh!

10:32:50 <Jibbler> i can imagine the "FOAF Sex Classification"

10:33:27 * darobin starts work on Ubercool Date Discovery Interface

10:33:38 <Jibbler> i can imagine it on jerry springer.... "there's something i have to tell you.... i used to be a foaf:sex male"

10:33:55 <sbp> it's interesting how picky we're being, though, given that hermaphrodites etc. must constantly come across forms with a male/female choice on it. I'd say male/female/other where other is subclassable and none of those are declared as being disjoint would be an adequate model

10:34:05 <libby> shellac suggests lists of chromasomes

10:34:26 <libby> (can exclude YY for dating ;)

10:34:46 <darobin> libby: that's sex as clinical, gives us five known permutation

10:34:58 <darobin> sbp: true

10:35:20 <danbri> [[

10:35:21 <danbri> <p>

10:35:21 <danbri> This may also be a good point for a periodic reminder: as with all FOAF properties,

10:35:21 <danbri> documents that use it will on occassion be innacurate, misleading or outright false. FOAF,

10:35:21 <danbri> like all open means of communication, supports <em>lying</em>. Application authors using

10:35:21 <danbri> FOAF data should be cautious in their presentation of all unverified information, but be

10:35:23 <danbri> particularly sensitive to issues and risks surrounding sex and gender. Users of information

10:35:25 <danbri> conveyed via FOAF (as via information conveyed through mobile phone text messages, email,

10:35:27 <danbri> Internet chat, HTML pages etc.) should be skeptical of unverified information.

10:35:29 <danbri> </p>

10:35:31 <danbri> ]]

10:36:14 <libby> good idea

10:36:16 <dajobe> you could send out monthly reminders like mailman on rand(30) day of month

10:36:26 <Mutiny> ala: don't go cybering people based on just what's in their FOAF file, you might find yourself questioning your sexuality on IRC

10:36:31 <dajobe> or rand(28) for the very picky

10:37:08 <libby> Mutiny: heh

10:37:50 <libby> danbri, we could ask ilrters, lots of biologists there

10:38:01 * danbri assumes most folk have read http://www.juliandibbell.com/texts/bungle.html by now

10:38:02 <Mutiny> foaf:sex with 'male', 'female' and 'other' values and a foaf:gender literal, I say.

10:38:04 <libby> maybe cc them in a mail?

10:54:24 <danbri> ok check out http://xmlns.com/foaf/0.1/#term_gender

10:58:07 <danbri> loggerf, pointer?

10:58:07 <danbri> See http://ilrt.org/discovery/chatlogs/foaf/2003-09-11#T10-58-07

11:01:44 <danbri> [At any point in time, an agent has at most one value for <code>foaf:gender</code>. FOAF

11:01:45 <danbri> does not treat <code>foaf:gender</code> as a <em>static</em> property; the same individual

11:01:45 <danbri> may have different values for this property at different times.

11:01:45 <danbri> ]

11:02:39 <JimH> as in Beckham played football like a girl last night...

11:03:58 * libby has female friends who play footie very well, thank you very much

11:04:26 <sbp> hence "a girl", not "all girls" maybe :-)

11:05:11 <JimH> sorry Libby, I was trying to think of cases where gender might be static and failed

11:05:19 <libby> sok :)

11:05:41 <sbp> and I was trying to make a quantification joke, but failed and looked like a git

11:05:46 <libby> heheh

11:06:41 * danbri revises http://xmlns.com/foaf/0.1/#term_gender

11:06:46 <danbri> quick read-thru anyone?

11:06:51 <danbri> have I been sensitive enough yet?

11:07:29 <danbri> I don't want to mess this up, but at same time want something as simple as possible. I hope the current design (and its writeup) is a reasonable balance.

11:08:46 <libby> I like it

11:09:05 <libby> danbri is a sensitive flower

11:09:24 <dajobe> my only flip comment was, wasn't Bender legally a female robot? in the fictional universe. in the future.

11:09:30 <dajobe> other than that, ok

11:09:43 <libby> Data is defintely a male robot

11:09:47 <Jibbler> can my cat be an agent, then?

11:09:56 <Jibbler> my important question remains unanswered

11:10:04 <danbri> yeah, i wondered if there might've been a Bender blot twist

11:10:22 <JimH> as it's supposed to be very simple an example would be nice for the hard of thinking like me - I assume it's <foaf:gender>male</foaf:gender> or <foaf:gender>female</foaf:gender> etc. ?

11:10:22 <danbri> that just adds colour to the example :)

11:10:22 <sbp> my worry about using literals and leaving it open is that someone comes along as says [ a foaf:Person; foaf:gender "something" ] and another person comes along and says [ a foaf:Person; foaf:gender "something" ] and you don't really know whether they mean the same thing by "something" or not. then again, you could say the same for "male" or whatever... hmm

11:10:49 <libby> I think at the moment, agent is defined so a cat could be an agent

11:11:08 <danbri> sbp, we have definedish meaning for 'male', 'female'. other values could be proposed...

11:11:10 <sbp> Agent Pussypaw, the crime fighting wondercat

11:11:25 <danbri> by using literals it does centralise things a little

11:11:29 <sbp> so you'd have to "register" the meaning somehow? and then it'd be incorporated in the schema?

11:11:32 <sbp> yeah

11:11:43 <danbri> but i don't want to over-encourage diversification; this is the wrong property to try to squeeze all that variety into

11:11:47 <libby> intersting that person best judge of their own gender...thinking of the intersex kids, biologicallmale, brought up as female

11:11:50 <danbri> instead people should make their own namespace

11:12:04 <sbp> myns:gender?

11:12:06 <danbri> own - yeah I thought that was important

11:12:19 <danbri> sbp, i mean a whole bunch of classes etc. and a separate property

11:12:27 <libby> corner cases where people could be incorrect, sorta

11:12:30 <danbri> it'd be a massive undertaking to do 'properly'

11:12:54 <sbp> yes. but on the other hand it makes me wonder what the value of foaf:gender is. what is it to be used for?

11:13:11 <libby> I think that's key

11:13:20 <danbri> defining rules for uncle/aunt/brother/sister/grandparent...

11:13:21 <shellac> it's dating. just admit it.

11:13:29 <danbri> well yes, and dating.

11:13:32 <sbp> (I mean value as in worth, not RDF object!)

11:13:32 * libby thinks that's where it's heading....

11:13:33 <sbp> hehheh

11:13:47 <danbri> but also the inference rules will help nail down the semantics

11:13:50 <shellac> ---------------------------------------

11:13:56 <libby> yeees danbri ;)

11:14:10 * shellac marks point when foaf became a dating service

11:14:20 <danbri> loggerf, pointer?

11:14:20 <danbri> See http://ilrt.org/discovery/chatlogs/foaf/2003-09-11#T11-14-20

11:14:33 <shellac> historic moment :-)

11:14:40 <sbp> don't forget to announce to rdfweb-dev

11:14:42 <danbri> OK I'm gonna send announcement to rdfweb-dev.

11:14:45 <danbri> heh

11:14:46 <sbp> heh

11:14:53 <danbri> one last readthru pls people...

11:14:59 <danbri> that url again http://xmlns.com/foaf/0.1/#term_gender

11:15:47 <libby> is there wn:male?

11:15:48 <sbp> perhaps you could add a little about its worth? the rationale for including it, what it can be used for? might help to prevent misdeployment

11:16:17 <danbri> generally we don't enumerate uses... hmm I'll have a think about that.

11:16:51 <danbri> i think its so common a concept as to barely need explaining in those terms...

11:16:57 <sbp> okay

11:17:15 <danbri> why might you want to know the gender of someone?

11:17:17 <libby> well, normally it's used for marketing purposes right?

11:18:01 <libby> maybe so companies can address you be your 'correct' title also

11:18:16 <libby> (though I get called 'Mr' quite often)

11:18:22 <shellac> foaf:interest wn:female?

11:18:29 <darobin> haha

11:18:32 <libby> heheh

11:18:35 <sbp> hehe

11:18:48 <danbri> yeah we could write rules in FoafAxioms wiki

11:19:03 <libby> what's wrong with foF;gender wn:female? etc

11:19:13 <sbp> would { [ foaf:title "Mr"; foaf:gender "female" ] } => raise an inconsistency?

11:19:14 <libby> except for wn:other.....

11:19:53 <danbri> { foaf:gender ?x 'male' } :- { foaf:title ?x 'Mr' } # I forget N3 syntax

11:19:57 <shellac> in that case you could use rdf:type wn:male

11:20:08 <danbri> sbp, heh i was just tpying same eg :)

11:20:22 <danbri> no, wordnet's Male/Female is too biological.

11:20:24 <sbp> cool

11:20:37 <danbri> yes, that would be inconsistent.

11:20:44 <sbp> FOAF doesn't prevent one from doing rdf:type wn:Whatever anyway, no?

11:20:53 <danbri> agreed

11:22:02 <shellac> so gender is a property, you're not in the class of men (say)

11:22:23 <shellac> maybe that should be an inference?

11:22:42 <shellac> owl restriction? be a nice example

11:23:16 <shellac> (pace issues with wn:male)

11:24:27 * danbri adds gender eg to http://rdfweb.org/topic/FoafAxioms

11:28:04 <shellac> { foaf:potentiallyInterested ?x ?y } :- {foaf:interest ?x ?z } & {foaf:gender ?y ?z}

11:28:22 <danbri> cute

11:30:57 <danbri> announced: http://rdfweb.org/pipermail/rdfweb-dev/2003-September/011917.html

11:40:26 * JimH has foaf:gender in the wild in http://feetup.org/jimfoaf.rdf

11:49:27 <danbri> IimH++

11:49:36 <danbri> er JimH++

11:50:19 * JimH wonders if he got even half close to getting it right though

11:51:43 <danbri> yup

11:52:03 <JimH> cool

11:52:35 <JimH> Rosco helped sort out a few things, so Libby++ too

12:18:38 <JibberJim> The chatlog robot protector has decided I'm a robot :-(

12:21:11 <sbp> heh, heh. it's a bit over-sensitive about that. it's happened before

12:21:55 * JibberJim would just change his IP, but this is the only one in my block that can talk to the SMSC...

12:22:19 <sbp> might be worth an email to dajobe, then. or perhaps he's around on IRC

12:23:00 <JibberJim> I'll wait 'til I see him, no point hassling when it's not urgent, I can get at them on another machine.

12:23:21 <sbp> good plan

12:29:10 <oPless^> heh

12:29:19 <oPless^> bad dobby

12:29:37 <oPless^> you spidered his chat logs?

12:32:14 <JibberJim> nope...

12:33:31 <oPless^> heh

12:36:00 <danbri> yeah, same thing happened here re chatlogs

12:37:39 <JibberJim> I used to read them a lot more when I wasn't at home too, but I guess I was on different IPs all the time then.

12:41:20 <JibberJim> Libby?

12:42:12 <libby> ello

12:42:32 <JibberJim> I was just talking with JimH, and he'd wanted to add title/desc to his sha1sums, so had done - but that made them resources, not literals of course - is there any chance of Rosco warning on literals in resources?

12:43:28 <JibberJim> sort of if the resource doesn't contain # or :, if the base uri doesn't end in a / ?

12:43:29 <libby> ah...I expect so

12:43:41 <libby> yeah

12:43:44 <libby> good idea

12:43:55 <JimH> sort of idiot proofing...

12:44:05 <libby> it's a good point

12:44:19 <libby> roscoe is unfinished abyway - doesn;t do domains

12:44:30 <libby> I'll have a look today

12:44:47 <danbri> rdf's data model is simple, but there are gotchas everywhere in the syntax, easy to slip up. Same goes for XML actually, although RDF gets more flack for it.

12:45:19 <JimH> ah, the Foaf explorer doesn't like my Foaf at all now :-(

12:45:21 <libby> easy to miss stuff

12:45:38 <danbri> thanks for http://feetup.org/blog/Foaf/FoafMobile.html btw, handy to have stuff collected up :)

12:45:38 <libby> hm foaf explorer uses xslt

12:45:54 <libby> be good to specify a schematron schema for foaf explorer

12:45:59 <libby> and have an online tester

12:46:15 * danbri hoping morten will port it to pure rdf...

12:46:25 <danbri> I believe he plans to, just busy

12:46:30 <libby> right yep

12:46:52 <libby> I will bully him when he comes to visit

12:47:11 <JibberJim> We could lock him in a room until he comes out with it finished?

12:47:20 <libby> "no mortenf, you can;t go to the pub until you've finidhed your explore to rdf port"

12:47:31 <libby> yeah

12:47:32 * JibberJim realises he shouldn't suggest that as I'll be locked in until foafnaut works...

12:47:37 <libby> hehehhe

12:48:08 <JimH> JibberJim: make sure you're locked in a room in a pub

12:48:27 * JibberJim doesn't know any BRS pubs with wireless...

12:49:08 <libby> we had a chat with our local landlord and he seemed keen, but we didn't follow it up

12:49:22 <libby> hare on the hill used to, as verbosus lived next door :)

12:50:15 <JibberJim> you shouldn't've let him go back to Italy...

12:50:45 <libby> yeah. nice pub too

12:51:18 <JimH> Libby I think nodeID has defeated Foaf Explorer, if I change them back to rdf:ID it's happy again :-(

12:51:43 <JibberJim> at least it should be a simple fix for Morten.

12:52:20 <libby> bum

12:52:27 <libby> nodeID is relatively new

12:52:47 <libby> I'm sure ID would eb ok till morten can fix it

12:53:15 <JibberJim> the ID police can be held off 'til then you think?

12:53:43 <libby> I reckon

12:53:59 <JimH> well they've probably got more to complain about than my Foaf anyway :-)

12:59:14 <dajobe> JibberJim: you are a robot?

12:59:41 <JibberJim> Apparently, who am I to argue?

13:00:00 <dajobe> IP 217.155.143.66

13:00:06 <JibberJim> yep.

13:00:22 <dajobe> fixed

13:00:31 <dajobe> but it auto-adds you when you do >50 requests in a short period

13:00:50 <dajobe> which seems robot-y enough for me

13:00:57 <JibberJim> Hmm, unlikely that I've done that, what's "a short period".

13:01:06 <JibberJim> certainly recently, not done more than 1 a day.

13:02:45 <dajobe> seems so

13:03:36 <dajobe> it might have been banned for some time then

13:04:01 <JibberJim> ah yeah, I was on 68 until I had to change back actually...

13:08:19 <chrisc_> chrisc_ is now known as chrisc

13:10:56 <libby> re the validation, jims, doesn't the rdfxml parser pick it up when literals have resources?

13:11:33 <JibberJim> no, it just creates a resource like http://example.org/foaf.rdfwww.example.com -

13:11:42 <libby> ok

13:12:00 <libby> ta

13:14:37 <libby> sorry to be dense, jim or jim, but do you have an e.g of the syntax used for resources in literals?

13:14:57 <JibberJim> <foaf:homepage="www.jibbering.com"/>

13:15:07 <JibberJim> <foaf:homepage rdf:resource="www.jibbering.com"/>

13:15:37 <dajobe> this isn't even XML: <foaf:homepage="www.jibbering.com"/>

13:15:49 <JibberJim> yes... I got it a bit wrong...

13:15:51 <libby> ah gotya

13:16:05 <dajobe> that's a relative URI I assume

13:17:02 <JibberJim> that's the problem, people intend it to not be relevant... I was hoping Rosco could do A "Are you sure you meant http://example.org/foaf.rdfwww.jibbering.com"

13:17:48 <libby> might have to have a think about how tot do that

13:18:40 <JibberJim> Does it have to be URI-encoded?

13:19:20 <libby> eh?

13:20:13 <JibberJim> in an RDF/XML file, does the rdf:resource="foo" have to be encoded? 'cos lots of people have things like "Fred Bloggs" in resources too, which would get picked up.

13:21:25 <libby> my code is a very simple wrapper around a parser. not sure how much information is preserved

13:23:51 <MartinR|oell> Hello everyone

13:23:53 <MartinR|oell> foafbot, amd's name

13:23:53 <foafbot> Can't identify 'amd', sorry.

13:24:08 <MartinR|oell> foafbot, danbri's name

13:24:09 <foafbot> Sorry, I don't know danbri's name

13:24:18 <JibberJim> looks like foafbot is ill!

13:24:22 <libby> foafbot, edd's name

13:24:23 <foafbot> edd's name is Edd Dumbill, according to Dan Brickley, Dave Beckett, Edd Dumbill, Jo Walsh, Matt Biddulph; and anonymous sources Anon212, Anon24, Anon34, Anon51, Anon548, Anon79,

13:24:23 <foafbot> Anon82, Anon9, Anon93

13:24:29 <JibberJim> foafbot, status

13:24:29 <MartinR|oell> foafbot, rainerw's name

13:24:29 <foafbot> Hi, I'm your friendly local FOAFbot. It may take a little while for me to compute my status, so hold on.

13:24:30 <foafbot> DB status: 536 URIs OK, 149 in error, and 0 yet to retrieve.

13:24:30 <foafbot> Spider status: 0 spiders active (max. 1), 0 URLs in queue.

13:24:30 <foafbot> Can't identify 'rainerw', sorry.

13:24:38 <libby> ill but not dead

13:24:46 <JibberJim> he's only visited a tiny subset of the world.

13:24:48 <libby> foafbot, picture of danbri

13:24:49 <foafbot> Sorry, couldn't find any matching pictures.

13:25:03 <libby> foafbot, picture of jibberjim

13:25:03 <foafbot> Can't identify 'jibberjim', sorry.

13:25:08 <MartinR|oell> foafbot, picture of libby

13:25:09 <foafbot> Found 30 pictures, just showing 3.

13:25:10 <foafbot> Picture at http://www.picdiary.com/barcelona/conf/dean_libby.jpg -- according to Anon51

13:25:10 <foafbot> Picture at http://rdfweb.org/people/danbri/rdfweb/libby.gif -- according to Anon170, Anon201, Anon218

13:25:10 <foafbot> Caption: A photo of Libby Miller, taken while at Oxford University

13:25:10 <foafbot> Picture at http://photos.heddley.com/albums/www2002/hawaii_014.sized.jpg -- according to Edd Dumbill

13:25:10 <foafbot> Caption: Dean Jackson, Amy ??? from W3C, Libby Miller

13:25:14 <libby> heheh

13:25:18 <JimH> foafbot, gender of JimH

13:25:31 <JimH> well, I can try :-)

13:25:38 <MartinR|oell> wow, cool

13:25:53 <JibberJim> he's normally not so ill, and knows a lot more folk.

13:26:08 <MartinR|oell> how does he learn about new people?

13:26:24 <libby> heh, jimh

13:26:38 <libby> I think edd periodically sends him off spidering

13:26:51 <MartinR|oell> foafbot, edd's name

13:26:52 <foafbot> edd's name is Edd Dumbill, according to Dan Brickley, Dave Beckett, Edd Dumbill, Jo Walsh, Matt Biddulph; and anonymous sources Anon212, Anon24, Anon34, Anon51, Anon548, Anon79,

13:26:52 <foafbot> Anon82, Anon9, Anon93

13:27:11 <JibberJim> he finds foaf files, I think his owner needs to him send off spidering, we used to be able to make him reload from a file, and find new-ones that way, but I don't think that works at the moment.

13:27:21 <libby> - he will find stuff from seealsos linked from various scutterplans, including one from a search on google I think

13:28:04 <MartinR|oell> well, i will come back the next days and see when he knows me. :)

13:29:38 <MartinR|oell> foafbot, libby's homepage

13:29:38 <foafbot> libby's homepage is http://ilrt.org/people/cmlm/, according to Morten Frederiksen

13:31:00 <JibberJim> where's your foaf file MartinR|oell?

13:31:36 <MartinR|oell> it's at http://www.roell.net/foaf.rdf

13:37:54 <JibberJim> erk foafnaut's update page doesn't work :-(

13:38:05 <JibberJim> at least it stored the urls I suppose.

13:52:31 <Jibbler> javadoc generalpath

13:52:35 <Jibbler> arg, wrong window

13:59:55 <rainerw> transfer $1.000.000

14:00:06 <rainerw> arg wrong window ;-)

14:00:18 <MartinR|oell> LOL!

14:00:20 <shellac> :-)

14:00:51 <MartinR|oell> foaf:mafiaConnection

14:03:25 <JibberJim> MartinR|oell, do you have your gpg key in the mit servers?

14:03:48 <MartinR|oell> yes, i think so

14:04:29 <JibberJim> Must be my script then, as I didn't manage to get it :-(

14:04:53 <MartinR|oell> wait a second, let me check...

14:05:19 <MartinR|oell> yes, it's there: http://pgp.mit.edu:11371/pks/lookup?search=roell&op=index

14:05:46 <MartinR|oell> second one from the top. Martin Roell <martin@roell.net>

14:06:04 <JibberJim> yep, same ID as you signed your file with, stupid programmer...

14:07:18 <JibberJim> Aww, you're disconnected from the world :-(

14:07:33 <JibberJim> http://jibbering.com/foaf/foafnaut.svg?sha1=512327da576318b02679cd91ce449c02304964f6 (if you've got SVG)

14:09:01 <MartinR|oell> true. i'm new to foaf. and it seems to me that in the german internet-community it is not very well known.

14:19:44 <rainerw> yesterday I discovered a starting point for the german speaking world (switzerland): http://xml.mfd-consult.dk/foaf/explorer/?foaf=http://blog.bitflux.ch/foaf/chregu.rdf

14:20:58 <MartinR|oell> oh, that's nice! i know his blog

14:21:29 <JibberJim> The foaf-people-map has a person in Stuttgart and one in cologne

14:22:17 <JimH> a couple of the #mobitopia guys are based in Germany

14:23:05 <MartinR|oell> where is the foaf-people-map?

14:23:28 <JibberJim> http://jibbering.com/foaf/foaf-people-map.svg although it's slightly broken, it doesn't find everyone :-(

14:25:59 <MartinR|oell> thanks

14:27:05 <MartinR|oell> http://jibbering.com/foaf/foafnaut.svg?0a9d600329b781dcdd710bf97c8ad16c44cf8dc9 may be an interesting starting point. many connections there.

14:28:48 <MartinR|oell> yippeh! i just found the first person i know (besides rainerw) that has a FOAF-file! :-)

14:28:50 <MartinR|oell>http://weblog.garyturner.net/foaf.rdf

14:29:31 <libby> :)

14:29:38 <JibberJim> hey foafnaut's explosions are turned off :-(

14:31:29 <rainerw> not really: http://xml.mfd-consult.dk/foaf/explorer/?foaf=http://joi.ito.com

14:32:28 <JibberJim> hey jbond

14:36:06 <fidothe> nice trick

14:46:00 <MartinR|oell> 3 more people. great. :)

14:53:36 <rainerw> hey, I only have 4 degrees to tim berners lee ;-)

14:53:53 <JibberJim> which means 6 to JFK!

14:55:43 <rainerw> no i only have 2 to clinton.

14:55:54 <ndw> TBL is 2 from JFK?

14:56:02 <dajobe> via presidents

14:56:14 <dajobe> US presidents that is

14:56:22 <ndw> I discovered by chance a couple of days ago that I'm only three from Clinton

14:56:30 <dajobe> remember the clinton JFK photo?

14:56:43 <ndw> no

14:56:54 <libby> did that turn up? it got lost

14:57:12 <JimH> was TBL in that photo?

14:57:13 <libby> wha's the path ndw?

14:57:16 <ndw> Oh, so the assertion is TBL to Clinton to JFK?

14:57:21 <JibberJim> yes

14:57:22 <dajobe> there: http://www.geocities.com/CapitolHill/4035/jfkwjc.jpg

14:57:25 <dajobe> google + 2 clicks

14:57:29 <ndw> That puts me 2 from Clinton. Heh. I beat my better half! :-)

14:57:29 <libby> ta

14:57:34 * JimH gets very confused as to how important people get connected with US presidents

14:58:01 <ndw> Deb knows a poet at UMass who's a friend of Clinton's, which got me there in three.

14:58:11 <JibberJim> US presidents like to surround themselves with important people, Bush is probably clamouring to meet norm at the moment.

14:58:14 <rainerw> perhaps there's a shortcut via "lover-of"

14:58:19 <ndw> lol

14:58:43 <libby> did you see the codepiction paths thing ndw? http://swordfish.rdfweb.org/discovery/2002/02/paths/

14:58:45 <ndw> How rude is it to decline the invitation, I wonder. And do I care?

14:58:51 <libby> there used to be JFK in there

14:59:09 <libby> heh

14:59:13 <dngor> I'm only one degree farther away from everybody you know!

14:59:16 <ndw> I think I saw the paths once, libby.

14:59:32 <libby> actually it's a bit out of date

15:00:06 <ndw> dngor, "at most one degree farther away" :-)

15:00:32 <dngor> Yeah, yeah, but that's less "fanboy" and more "maths major".

15:00:42 <ndw> he

15:00:47 <ndw> heh, even

15:02:20 <danbri> clinton pic example is at http://rdfweb.org/2002/01/photo/

15:02:33 * ndw waits for "list of people in database" to complete. Is starting to expect a *big* result :-)

15:02:33 <danbri> links me to frank sinatra...

15:03:08 <libby> ndw if that's my stuff some is v slow :(

15:03:44 <ndw> There's the Clinton/JFK picture, too.

15:04:03 <JibberJim> You should see my stuff Lib, people using live foafnaut, means my server isn't up to sending SCP files across the local network to me above 3k/sec

15:04:20 <libby> ouch!

15:04:48 <libby> I get these sudden lengthy pauses when using my box a sa desktop. wish I could fix it up

15:05:15 <libby> all I can do is cut doen the really slow queries

15:05:38 <JimH> Libby that's some old data you've got there!

15:05:57 <JibberJim> it's quite an old slow box now... I'm thinking of switching to all C++ for my stuff, as I've been playing with libmysql++ and it's quite nice.

15:06:12 <libby> is it JimH?

15:06:24 <libby> haven;t scutterd for a bit

15:06:34 <libby> really jim...? wow

15:06:43 <JimH> the stuff in http://swordfish.rdfweb.org/discovery/2001/08/codepict/allpeople.jsp must be at least 3-6 months old

15:06:47 <libby> mine's only a year or so old.

15:06:55 <libby> really!

15:07:01 * libby checks what's up

15:07:04 <JimH> and it's missing a lot of people I know of

15:07:12 <libby> I've moved adtabases recently

15:16:27 <libby> ok, now takes about 2 minues (boo) but finds total: 2549 people (yay!)

15:21:33 <danbri> crontabbable?

15:22:04 <libby> yeah. maybe that's the best way

15:22:33 <libby> I think jim said that I needed more indexes too

15:31:52 <MartinR|oell> i have added myself to the http://swordfish.rdfweb.org/rweb/who database.

15:32:12 <libby> I think I already added you :)

15:32:41 <MartinR|oell> oh. :-)

15:33:57 <libby> doesn;t make any difference...

15:34:20 <libby> my database doesn't smush currently, so can be rather disconnected

15:36:34 <MartinR|oell> Thank you for your help! This was a really fascinating day for me, learning so many new things about foaf. I will be back more often.

15:37:14 <MartinR|oell> have a nice day... cu

15:37:19 <JibberJim> bye

15:37:33 <libby> bye!

16:13:47 <danbri> Last update:

16:13:47 <danbri> $Date: 2003/09/11 16:08:26 $

16:13:47 <danbri> Revision:

16:13:47 <danbri> $Revision: 1.45 $

16:13:48 <danbri> dohhhhhhhh

16:13:59 <danbri> Was wondering why the foaf spec pgp signing wasn't working

16:14:14 <danbri> whenver i cvs commit it, the file is rewritten with date and version info

16:14:15 <danbri> doh

16:14:55 <libby> heheh

16:15:01 <libby> aw, poor danbri

16:15:12 <danbri> danbri@fireball:~/s-foaf/0.1$ GET -H 'Accept: application/rdf+xml' http://xmlns.com/foaf/0.1/ | gpg --verify ../foafsig -

16:15:12 <danbri> gpg: Signature made Thu Sep 11 17:12:49 2003 BST using DSA key ID 73228FE4

16:15:12 <danbri> gpg: Good signature from "Dan Brickley <danbri@w3.org>"

16:15:12 <danbri> gpg: aka "Dan Brickley <daniel.brickley@bristol.ac.uk>"

16:15:12 <danbri> gpg: WARNING: This key is not certified with a trusted signature!

16:15:14 <danbri> gpg: There is no indication that the signature belongs to the owner.

16:15:16 <danbri> Primary key fingerprint: FA0C 0D5A 2B3F 808D AA28 2B63 3E15 EF2F 7322 8FE4

16:16:12 <dajobe> trust no one, eh?

16:16:41 <danbri> quite :)

16:24:17 * mortenf checks logs, wonders whether he should cancel his return ticket or sign up for houdini 101...

16:24:43 <libby> noooo

16:24:49 <mortenf> ;)

16:24:49 <libby> we didn;t mean it mortenf :)

16:25:01 <JibberJim> houdini 101 sounds a good plan though.

16:25:28 <JimH> or hide in a glass box in central London

16:25:30 <mortenf> actually i (re-)started thinking about a way to do FE2 (since FE3 is postponed)

16:25:38 <mortenf> seems that's been done already? :)

16:25:43 <JibberJim> nah JimH, that'd just get lots of abuse!

16:25:54 <JimH> bit like being in here then :-)

16:26:13 <JibberJim> true, but less this is less eggy.

16:26:26 <mortenf> i want to do a system which can easily be extended, and to allow for contributions (xslt-templates)

16:26:54 <danbri> what was FE3 going to be?

16:26:59 <mortenf> but redland 0.9.14 doesn't seem to work with php :(

16:27:10 <mortenf> the storage bound one...

16:27:39 <mortenf> still researching, testing etc.

16:28:07 <dajobe> eh? I didn't change nuffin

16:28:11 <mortenf> so for now, i want to parse and reserialize (with redland, hopefully)

16:28:25 <mortenf> just installed new apache and php 4.3.3 - no workie :(

16:28:29 <mortenf> dumps core

16:28:49 <dajobe> I published what it worked with for 0.9.13, same for 0.9.14

16:28:57 <mortenf> did you try it? :)

16:30:06 <dajobe> to use it? no. But I don't use java or tcl or csharp ... either

16:30:13 <dajobe> the test code did work

16:30:18 <mortenf> hmm

16:31:13 * danbri wonders how readable foaf is after reserialization by redland

16:34:53 <nachlin_> somewhere, i saw a list of the commonest properties found in foafs on the web. does anyone have a URL for that?

16:36:08 <libby> there's a couple - jibberjim's got one

16:36:12 <JibberJim> http://jibbering.com/foaf/predicate-list.1 is mine nachlin, Libby's got one too somewhere.

16:36:35 <JibberJim> that's not a full foaf crawl, huge FOAF places like ecademy and the BHG are excluded unfortunately.

16:36:51 <libby> here: http://swordfish.rdfweb.org/discovery/2001/08/codepict/stats.html also

16:36:59 <libby> I'll just check that that is up to date

16:37:06 <danbri> wanna link those from wiki anyone?

16:37:10 * danbri all typed out

16:37:40 <nachlin_> thanks. maybe this is a crazy idea, but i'm writing a tool similar to the foafexplorer, and i want to be sure it recognizes the most commonly used stuff.

16:37:51 <libby> hm, mine's pointing at the wrong database...ficxing

16:39:27 <JibberJim> Sounds good nachlin_! - remember tools are also a good way of getting people to add stuff, if you support something interesting people may way add properties so they appear!

16:40:54 <nachlin_> one of the ideas behind my project is that a small group may want to support properties that the larger community might not be interested in. like, for instance, somethisng specific to my CS department.

16:41:35 <opless2> I canextract a list too

16:42:22 <danbri> dajobe, can http://www.redland.opensource.ac.uk/docs/pod/RDF/Redland/Serializer.html give me a string, or just write to files?

16:43:40 <dajobe> to files only

16:43:49 <danbri> <- feature request

16:43:49 <mortenf> string would be nice...

16:43:52 <mortenf> :)

16:43:53 <dajobe> sure

16:43:56 <danbri> ta!

16:43:59 <dajobe> I only tend to add after requests

16:44:09 <danbri> good policy

16:44:10 <mattb> parse from string, output to string both very useful

16:44:56 <dajobe> it might take some time since most of the serialization code assumes files or C FILE*

16:44:59 <danbri> <rdf:Description rdf:nodeID="dan">

16:44:59 <danbri> <ns0:mbox xmlns:ns0="http://xmlns.com/foaf/0.1/" rdf:resource="mailto:danbri@apocalypse.org"/>

16:44:59 <danbri> </rdf:Description>

16:44:59 <danbri> <rdf:Description rdf:nodeID="dan">

16:45:00 <danbri> <ns0:mbox xmlns:ns0="http://xmlns.com/foaf/0.1/" rdf:resource="mailto:danbri@w3.org"/>

16:45:02 <danbri> </rdf:Description>

16:45:12 <danbri> ...it breaks things out into separate elements, is that intended behaviour?

16:45:27 <mattb> looks like a redland serialisation to me

16:45:28 <mortenf> it's simple...

16:45:28 <dajobe> yes, stateless between triples

16:45:32 <mattb> nice and simple

16:45:48 <dajobe> printf("ns0:%s", qname) etc.

16:45:59 <mortenf> optional sort and group by subject would also be nice

16:46:08 <danbri> fine by me, i just worry about borking foaf explorer with markup in this style...

16:46:11 * dajobe points to the source

16:46:57 <mortenf> heh, got it, when you add to-string, i'll add the optional sort

16:48:39 <libby> if people are trying the swordfsih stats page, you may have to wait about 5-10 minutes!

16:48:42 <libby> crud

16:48:59 <libby> sorry about that. another candidate for pre generating that page

16:49:52 <libby> or it may just time out :(

16:54:33 <nachlin_> jibberjim, this is strange. in your list of predicates, http://jibbering.com/foaf/jim.rdf#isKnownBy is fourth most common. ???

16:54:49 <danbri> he generates those!

16:54:56 <danbri> an artifact of how foafnaut works

16:55:08 <JibberJim> Yeah, sorry nachlin, it's just an artefact of the database.

16:55:12 <mortenf> biab

16:55:47 <JibberJim> to make foafnaut queries more efficient it creates an inverse knows triple for every foaf:knows.

16:56:22 <danbri> (which could also be called foaf:knows, I still argue...)

16:56:37 <danbri> (with provenance dealing with the who said what)

16:57:08 <nachlin_> i should have guessed that since isKnownBy.count() == knows.count();

16:58:04 * JibberJim doesn't agree with danbri's arguments, and as implementor I get to decide :-)

16:59:23 * opless2 chuckles

16:59:30 <opless2> opless2 is now known as oPless

16:59:56 <JibberJim> anyway I need the extra colours in foafnaut..

17:05:03 <danbri> the colours indicate who-said-it not what-was-said, though

17:06:27 <JibberJim> but not enough people sign their documents, so we don't know who said what.

17:07:04 <JibberJim> all we know is that someone said some stuff about A and that A knows B.

17:32:16 <danbri> http://www.projectcensored.org/publications/2004/index.html - Censored 2004: The Top 25 Censored Media Stories of 2002-2003

18:19:11 <Schuyler> danbri: Project Censored is based right here in Sonoma County.

18:19:15 <Schuyler> They're our heroes.

18:31:23 <GNUWookie> GNUWookie is now known as mdupont

18:32:46 <JibberJim> Anyone know any good protocols for sending small amounts of data from a server to a client, with little but acks and an initial login going the other way?

18:33:37 <mdupont> there is the simple way that webchats work by constantly refreshing a page with a cookie..

18:33:56 <mdupont> hi JibberJim

18:35:25 <JibberJim> hi, HTTP ain't what I want. SMPP is the most appropriate, but I don't like the protocol at all.

18:35:56 <mdupont> i kinda thought that after i hit enter

18:36:02 <mdupont> hmmm.... jabber?

18:36:03 <JibberJim> I wonder if Jabber or something might be appropriate.

18:36:06 <JibberJim> snap!

18:36:27 <JibberJim> ERK, it's XML...

18:36:33 <mdupont> what is your server written in?

18:36:54 <JibberJim> neither my server or my client are written... but I imagine they'll be C++

18:38:04 <zoyd> wanted to know if anyone uses the barter schema by Ian Davis.

18:38:37 <JibberJim> do you have the namespace?

18:38:40 <zoyd> foafbot: Ian Davis' weblog

18:39:23 <zoyd> one sec

18:39:59 <JibberJim> entering a predicate into http://jibbering.com/foaf/pred.1?predicate=http://... will turn up anyone I know.

18:40:08 <zoyd>http://purl.org/vocab/barter/0.1/

18:40:30 <JibberJim> It seems like I know no-one :-(

18:41:13 <zoyd> JibberJim: "no-one" who uses that namespace?

18:41:55 <JibberJim> no document found in a large crawl of the foaf world, contains that predicate

18:42:10 <zoyd> aah

18:42:24 <JibberJim> http://jibbering.com/foaf/pred.1?predicate=http://purl.org/vocab/barter/0.1/

18:42:56 <zoyd> reference: http://rdfig.xmlhack.com/2003/09/05/2003-09-05.html#1062749717.978209

18:43:25 <JibberJim> ah, although I don't find anyone using the foaf namespace either :-(

18:43:30 <zoyd> was thinking the barter schema would help for the above

18:44:37 <JibberJim> but even fixing it, I don't find anyone :-(

18:46:11 <mdupont> what about a soap messaging system?

18:46:34 <dngor> SOAP-O-GRAM

18:46:47 <zoyd> "soap messaging system"?

18:47:18 <JibberJim> I asked before you arrived zoyd if anyone knew of any short messaging protocols.

18:47:30 <mdupont> you can use smtp for soap

18:47:38 <mdupont> have them use an email system

18:47:44 <mdupont> put the soap message in that

18:48:10 <JibberJim> client basically just recieving short messages from a server. I think that's all too much overhead mdupont, this is likely running off a 9k or at best GPRS connection.

18:48:21 <mdupont> wow

18:49:38 <zoyd> btw, how many people are using foaf from India?

18:49:56 * zoyd needs to look for a standalone SVG viewer

18:50:24 <danbri> you're the only one I know of

18:51:24 <JibberJim> Batik is the only option for that zoyd really.

18:51:34 <JibberJim> Batik's squiggle that is.

18:51:50 <JibberJim> Eeek, zoyd's fallen off the foaf-people map :-(

18:52:03 <zoyd> danbri: how'd you find that out?

18:52:34 <danbri> Last I looked at the foaf map, anyway

18:52:49 <zoyd> aah

18:52:56 <danbri> also just from memory... in terms of people who show up in foaf irc/mail/etc discussions

18:52:56 <JibberJim> there's no one in India in the map...

18:54:05 <zoyd> JibberJim: i remember submitting my URL.

18:54:21 <JibberJim> yeah, I think it's broke, you were there before :-(

18:54:55 * mdupont is on the phone

18:55:09 * danbri finds http://plug.org.in/pipermail/plug-mail/2003-May/008045.html

18:56:21 <zoyd> *grin* didn't help a tiny bit.

18:56:48 * danbri notices you've been busy with Atom stuff... has Atom stabilised at all yet?

18:57:45 <zoyd> it works .. the auth part is still kind of wiggly.

18:57:56 <zoyd> but it's usable.

18:58:49 <zoyd> had problems with a python minidom bug ... couldn't get my cgi implementation working coz of that

18:59:44 * danbri notes there's no Indian version of http://www.ldodds.com/foaf/foaf-a-matic.html yet (hint hint...)

18:59:53 <danbri> any chance you'd be persuadable to help w/ a translation?

19:00:09 * mdupont back

19:00:37 <shellac> what's Indian?

19:00:41 <shellac> Hindi?

19:01:02 <JibberJim> lots of languages...

19:01:08 <zoyd> yeah

19:01:08 * danbri wasn't naming a language, but meaning 'for folks in India'...

19:01:39 * zoyd shall talk to some l10n guys about translating that

19:01:50 <danbri> or maybe English so widely used that translation not needed?

19:02:25 <zoyd> yes it's very widely used.

19:02:37 <danbri> l10n -- that'd be great

19:02:46 <danbri> there are quite a few translations now

19:03:07 <danbri> english + 7 on Leigh's site, and a korean one somewhere too

19:03:41 <zoyd> need some ideas to push foaf/rdf in my user group.

19:04:04 <danbri> who are your user group?

19:04:17 <zoyd>http://www.plug.org.in/

19:05:25 <zoyd> there are quite a few developers .. most are obviously admins.

19:05:50 <danbri> maybe something like http://www.debian.org/devel/developers.loc but more local? how big an area does yr group cover?

19:06:44 <zoyd> 15-20km radius.

19:07:09 <zoyd> the interesting part is the amount of students that come here to study.

19:08:51 * danbri notices the membership form is PDF... is that usual for linux user groups?

19:09:10 <zoyd> dunno.

19:10:48 <zoyd> developers.loc is nice .... but a user group would need things like GPS devices to get that kind of detail.

19:13:16 <danbri> this is a group of geeks, though... but yeah, gps isn't cheap enough by far...

19:14:11 * JibberJim needs to return to India, India was the reason I bought a GPS...

19:18:14 <zoyd> foafbot: homepage of Vinay Pawar

19:18:45 <JibberJim> foafbot, homepage of zoyd

19:18:53 <JibberJim> it needs a comma, not a colon.

19:19:25 * zoyd waves at foafbot hoping it would see him

19:20:00 <JibberJim> foafbot, Edd's name

19:20:00 <foafbot> Can't identify 'Edd', sorry.

19:20:07 <JibberJim> foafbot, zoyd's name

19:20:08 <foafbot> Can't identify 'zoyd', sorry.

19:20:09 <zoyd> hehe

19:33:37 <edd> so...

19:33:40 <edd> e-tech proposal deadline looms

19:33:50 <edd> what's the situation re foaf proposals?

19:33:50 * JibberJim just saw that too...

19:34:11 * danbri hasn't done his yet, will put in for a general foaf overview

19:34:12 * edd isn't talking to you, JibberJim :-p

19:34:27 <danbri> foafwars!

19:34:48 <edd> <foaf:saysMeanThingsBehindEddsBack />

19:34:57 <JibberJim> :-)

19:38:44 <ndw> e-tech?

19:38:48 <JibberJim> FoafMobile is what you should be taking, all the other Foaf tools are mature, not emerging :-)

19:38:58 <ndw> emerging technologies?

19:39:04 * ndw so wants to go to that conference

19:40:53 * edd was prolly going to pitch foafbot, with a mobile aspect to it

19:42:36 <JibberJim> I'd like to make a mobile foafnaut, primarily to show off SVG on a smartphone... but not sure how practical that is.

19:42:47 <JibberJim> releasing it would be the biggest problem.

19:44:39 <oPless> there is svg?

19:44:45 <oPless> or a smart phone?

19:44:50 <oPless> s/or/on/

19:45:02 <JibberJim> yeah, there's at least 2 viewers for Symbian devices

19:45:06 <oPless> oh

19:45:16 <JibberJim> the PPC smart phones presumably have 3

19:45:29 <oPless> not on java then?

19:45:30 <JibberJim> tinyline may well run on many too (that's java based)

19:45:37 <danbri> yeah, mobile would be a good focus

19:45:42 <danbri> might even have something running by then

19:45:42 <oPless> j2me to be precise

19:45:54 * danbri bets he'll have a new phone by then too, gadget junky...

19:46:17 <JibberJim> The Bitflash viewer I've got is really, really good, I'm very impressed with how capable it is.

19:47:01 <JibberJim> but it's not released as a browser, it's for embedding in other progs...

19:47:37 <edd> meanwhile all new smartphones will come with Flash

19:47:48 <edd> ... something people already have authoring tools for

19:47:51 <edd> *sigh*

19:48:05 <JibberJim> Hmm, only the A920 has flash.

19:48:23 <edd> OK, maybe that's my pessimistic projection.

19:48:34 <edd> But I would hardly think it'll be the last.

19:48:35 <JibberJim> It really depends on what Opera will do I think.

19:48:50 <oPless> eww - flash :(

19:48:59 * edd is of the opinion that W3C *really*, umm, "screwed the pooch", over SVG adoption

19:49:26 <JibberJim> anyway, I don't mind flash much any more - Robin Debruill has really shown how well flasf2SVG can be done...

19:49:58 <oPless> I'm sure once M$ include it in windows foghorn 2009 it'll be 'standardized'

19:50:55 * oPless longs for the days of multiple different platforms where there was actul competition

19:53:37 * danbri -> read his midp books and eat

19:57:43 * zoyd is off to sleep

20:07:58 <oPless> interesting point on el reg

20:07:59 <oPless>http://www.theregister.co.uk/content/4/32780.html

20:08:10 <oPless> s/point/article/

20:08:13 <oPless> <-- tired

20:10:02 <Jibbler> i have no respect for The Register

20:10:19 <Jibbler> not after they got my nickname wrong

20:10:19 * JibberJim agrees with Jibbler

20:10:35 <Jibbler> Look at this: http://www.theregister.co.uk/content/55/26200.html

20:10:37 <Jibbler> bastards!

20:11:04 <oPless> lol

20:11:07 <oPless> jibble!

20:13:38 <oPless> gotta love the reg tho

20:14:17 <oPless> I found out a semi-sponsored customer was selling CS servers on my network

20:14:28 <oPless> through el reg

20:14:35 <Jibbler> i find their articles a bit lame to be honest... they seem to be written for shock factor and are often technically incorrect

20:14:41 <Jibbler> but that's just my opinion

20:14:43 <oPless> true

20:14:45 <Jibbler> and they got my nickname wrong

20:14:48 * JibberJim agrees with Jibbler

20:14:51 <Jibbler> but i'm not bitter, oh no

20:15:04 <Jibbler> JibberJim: i see you're supporting the theory that we are the same person :)

20:15:09 <oPless> /. el reg inquirer etc all very "yawn"

20:15:12 * ndw missed the beginning of that thread. Who's articles?

20:15:18 <oPless> el reg

20:15:29 <ndw> ah. yes. funny often, but a bit short.

20:15:43 <oPless> it's all very "tabloid"

20:15:57 <Jibbler> they got my nickname wrong, *and* their claim in that article that the problem was fixed wasn't true.

20:16:05 <oPless> but without the semi-naked women

20:16:37 <JibberJim> my problem is I've virtually never seen an article on anything I understand that didn't include huge inaccuracies, so I have no trust in any of the articles where it's stuff I don't know about.

20:17:26 <oPless> trouble is all the so-called "news" sites all nick stuff off each other

20:17:36 <Esaj_> Esaj_ is now known as Esaj

20:17:39 <Jibbler> although having them link to my website did cause me to shift 500mb of traffic in one afternoon :)

20:17:47 <oPless> heh

20:17:52 <oPless> that all ?

20:17:57 <ndw> Non sequitor. Everytime I see the /topic of this group in xchat, I read it as "...abort, retry, flail" which I invariably find amusing.

20:18:21 <oPless> oPless has changed the topic to: foaf-project.org - rdfweb.org # foaf(X,Z) :- foaf(X,Y), foaf(Y,Z) # public logs: http://ilrt.org/discovery/chatlogs/foaf/latest # You do not have any friends. [a]bort, [r]etry, [f]lail...

20:18:25 <Jibbler> yes, there were nowhere near as many hits as i would have expected.... maybe there aren't that many people that read it

20:18:37 <oPless> better ndw?

20:19:05 <oPless> geez I'm tired

20:19:07 <ndw> rotfl

20:19:09 <oPless> this is so wrong

20:20:44 <edd> s/sequitor/sequitur/

20:21:55 <ndw> Thanks, edd

20:22:21 <JibberJim> handy to have an editor here to make sure our IRCing is tip-top.

20:22:34 * edd justifies his existence

20:26:25 * mortenf still has no luck getting redland php working

20:27:39 <Cardinal> Heh. I got it built, and promptly haven't gotten around to testing it.

20:59:36 <dajobe> not sure what I can say, it works for me and I don't have enough info to diagnose what's wrong for you

21:00:05 <dajobe> the only tricky thing I found was making sure it was using the right librdf.so - but I have lots hanging about

21:00:13 <mortenf> i know, i'll try to see if i can narrow it down

21:00:21 <mortenf> hmm, will check that

21:00:35 <dajobe> run the core in gdb and see the libraries it loads are what you expect

21:04:27 <mortenf> hmm, seems it might be a hosed libxml2

21:07:01 <mortenf> hmm, no, seems to be a pthread issue

21:07:30 * mortenf tries without --with-threads

21:07:34 <oPless> hmmm http://freshmeat.net/branches/41304/

21:08:02 <ndw> On the subject of hosed libraries, I know there's some Debian unstable folks here, anyone want to comment on libc 2.3.2-6?

21:08:11 <dajobe> running it now

21:08:39 <ndw> Ok, I'll let apt-get upgrade run then :-)

21:08:59 <dajobe> redland should be there now

21:09:23 <ndw> oh, cool, a chance to play with redland maybe

21:09:31 <dajobe> I'm just checking

21:09:39 <dajobe> clearly I've got it installed from my local packages

21:09:45 <Jibbler> doh, i missed the new canerbury tales

21:09:51 <Jibbler> but spelt properly

21:09:56 <dajobe> the wife of bath did it

21:15:19 <dajobe> yes, I see librdf-ruby etc.

21:23:57 <ndw_> ndw_ is now known as ndw

21:59:28 <mortenf> seems redland won't check with --with-raptor=system - fails to find raptor.h for rdf_uri_test

21:59:40 <mortenf> (and why build the internal raptor when =system?)

22:01:21 <dajobe> it shouldn't build anything in the raptor dir when --with-raptor=system, it's removed from the list of sub directories

22:01:39 <mortenf> hmm, it seems to do it anyway

22:02:31 <mortenf> still stuck at a thread problem i think, Reading symbols from /lib/i686/libpthread.so.0...done.

22:02:31 <mortenf>

22:02:31 <mortenf> warning: Unable to set global thread event mask: generic error

22:02:31 <mortenf> [New Thread 1024 (LWP 21649)]

22:02:31 <mortenf> Error while reading shared library symbols:

22:02:32 <mortenf> Cannot enable thread event reporting for Thread 1024 (LWP 21649): generic error #0 __pthread_mutex_lock (mutex=0x0) at mutex.c:99

22:02:35 <mortenf> 99 mutex.c: No such file or directory.

22:02:37 <mortenf> in mutex.c

22:02:43 <mortenf> any ideas?

22:02:52 <dajobe> that could be php, libxml2, libcurl or redland

22:03:30 <dajobe> I can't debug anything with threads, gdb just crashes usually

22:04:34 <dajobe> tar xfz; ./configure --with-raptor=system; make; make check; does not enter raptor dir

22:05:08 <dajobe> from redland-0.9.14.tar.gz

22:05:28 <dajobe> that's with raptor-1.0.0 installed

22:05:39 <mortenf> yep, that's what i've got as well

22:06:19 * mortenf extracts from scratch

22:06:26 <dajobe> so "<mortenf> seems redland won't check with --with-raptor=system - fails to find raptor.h for rdf_uri_test" must be something else

22:08:15 <mortenf> well, it does it again, did you try "make check"?

22:08:30 <dajobe> yes

22:08:42 <mortenf> and it passes rdf_uri_test?

22:09:04 <mortenf> (in ./librdf)

22:09:17 <dajobe> yes

22:09:21 <dajobe> all tests worked

22:09:24 <mortenf> not here :(

22:09:33 <mortenf> and php still fails

22:09:43 <mortenf> i think i'll wait untill i get around to upgrading to rh9

22:09:47 <dajobe> ah

22:09:59 <dajobe> what do you have?

22:10:03 <mortenf> 7.1

22:10:06 <mortenf> ++

22:10:09 <dajobe> ah

22:10:14 <dajobe> I just threw my 7.1 cds out

22:10:30 <mortenf> heh, and i dl'ed the 9 isos a couple of weeks ago

22:10:52 <mortenf> anyway, strange thing that it used to work...

22:10:55 <dajobe> better to spend the 1hour doing that than debugging this

22:11:10 <mortenf> well, upgrading would likely take more time, but yes, should do it anyway

22:11:42 <dajobe> I can't really tell why, apart from mutter about threads. That stuff has improved a lot in the many years since 7.1

22:12:08 <mortenf> i imagine so, and no prob, thx for your time

22:12:27 <dajobe> it's still hard to debug even with the latest shiny software

22:13:04 * mortenf looks for that gzip --25-hour-day option...

22:13:12 <dajobe> try bzip2

22:13:17 <mortenf> ah

22:30:18 <mortenf> ot: would a 64bit pci card work in a 2 year old mb, is it backward compat?

22:34:37 <_joshua> 64bit pci card has a different connector

22:34:46 <mortenf> :(

22:39:58 <ndw> ndw is now known as ndw^afk

22:44:04 <danbri> http://rdfweb.org/mt/foaflog/archives/000067.html <- foaf:gender announced

22:46:56 <JibberJim> Am I right in thinking Owl couldn't be used to say things like A rel:siblingOf B and B foaf:gender "male" -> A rel:brotherOf B ?

22:47:19 <JibberJim> the problem being the literal range of foaf:gender?

22:49:51 <mortenf> btw, foaf:thumbnail could be an owl:TransitiveProperty

22:50:39 <JibberJim> hmm, GoogleAd's are beginning to annoy me.

22:50:56 <mortenf> just ignore them

22:52:29 <danbri> Jim, yeah we need N3-ish rules stuff for that. Regardless of literal-vs-resource

22:53:48 <JibberJim> oh right, I thought owl coulda done it if it was a resource.

22:56:13 * JibberJim works on kill google ad stylesheet.

23:00:45 <mortenf> nn

23:01:16 <JibberJim> oh well got rid of most...

23:49:43 <ndw^afk> ndw^afk is now known as ndw

23:58:08 <danbri> morning edd

23:58:28 <edd> hiyas