Semantic Web Interest Group IRC Chat Logs for 2003-01-09

00:22:39 <icepick-away> icepick-away is now known as icepick

02:57:14 <mdupont-zzzz> mdupont-zzzz is now known as mdupont

03:19:49 <mdupont> morning all

06:44:29 <mdupont_> mdupont_ is now known as mdupont-ZZZ

09:29:03 * bijan waves from WebOnt f2f

09:31:59 <squizz> squizz is now known as mattwhe

10:41:11 <libby> morning bijan

10:51:49 <bijan> Hey libby

11:19:04 <libby> how's the webont meeting going?

11:23:01 <aharth> hi

11:24:23 <libby> morning

11:57:02 <dajobe>http://www.w3.org/2001/sw/WebOnt/

11:57:12 <dajobe> hmm

11:58:01 <libby> hmm?

11:58:29 <dajobe> no chumpbot

11:58:49 <libby> are they using your logger?

11:58:56 <dajobe> who?

11:59:03 <libby> oh

11:59:10 <dajobe> i'm running my logger at webont, which is what I was about to chump

11:59:10 * libby confused, ignore me

11:59:15 <libby> right

11:59:22 <libby> but the chump here is dead

11:59:35 <dajobe> I just restarted it for you

11:59:35 <libby> hooray!

11:59:53 <dajobe>http://www.w3.org/2001/sw/WebOnt/

11:59:53 <dc_rdfig> A: http://www.w3.org/2001/sw/WebOnt/ from dajobe

12:00:02 <dajobe> A:|WebOnt F2F meeting in Manchester 9-10 Jan

12:00:02 <dc_rdfig> Titled item A.

12:00:18 <dajobe> A:official irc logs are here, but I've got [http://ilrt.org/discovery/chatlogs/webont/2003-01-09.html|live logs for today] that you can browse

12:00:18 <dc_rdfig> Added comment A1.

12:01:27 <dajobe> A:seems the official logs are live too, which is news to me. So pick one ;)

12:01:27 <dc_rdfig> Added comment A2.

12:16:52 * danbri waves to all, hopes WebOnt meeting is going well

12:22:11 <dajobe> ok, the chump stuff now all points to irc.freenode.net

13:30:43 <dajobe>http://www.guardian.co.uk/online/story/0,3605,870848,00.html

13:30:43 <dc_rdfig> B: http://www.guardian.co.uk/online/story/0,3605,870848,00.html from dajobe

13:31:15 <dajobe> B:|"Click to the clique", Ben Hammersley, Guardian

13:31:15 <dc_rdfig> Titled item B.

13:31:31 <dajobe> B:includes mention of FOAF and some foaf tools

13:31:31 <dc_rdfig> Added comment B1.

14:18:33 <danbri>http://www-6.ibm.com/jp/developerworks/xml/020906/j_x-foaf.html

14:18:33 <dc_rdfig> C: http://www-6.ibm.com/jp/developerworks/xml/020906/j_x-foaf.html from danbri

14:18:59 <danbri> C:|Edd's ibm developerworks article on Friend-of-a-Friend, in Japanese

14:19:00 <dc_rdfig> Titled item C.

14:19:20 <danbri> C:And [http://www-900.ibm.com/developerWorks/cn/xml/x-watch/part3/index.shtml|chinese].

14:19:20 <dc_rdfig> Added comment C1.

14:19:45 <danbri> C:Writing for them seems a good idea, if you get free translations of your article...

14:19:45 <dc_rdfig> Added comment C2.

14:19:53 <JibberJim> and English?

14:24:30 <libby> nah

14:26:06 <icepick> icepick is now known as icepick-away

14:33:31 <danbri> C:And the [http://www-106.ibm.com/developerworks/xml/library/x-foaf.html|English original].

14:33:31 <dc_rdfig> Added comment C3.

14:42:03 <_joshua> DanC, you around?

14:42:12 <DanC> sorta

14:42:20 <_joshua> Ah. you asked me some questions yesterday about geourl

14:42:27 <_joshua> or the day before

14:42:38 <DanC> I saw it got slashdotted... cool!

14:42:47 <DanC> I looked at some details, and was pretty happy with what I found.

14:42:47 <_joshua> yeah, my poor little machine

14:42:55 <_joshua> anyway, i recall the questiosn were

14:43:16 <_joshua> "do they use meta tags" - yes, other people should index them

14:43:50 <_joshua> "why DC.title" - so people can use something different on the <title> and their listing. I'll use title if dc.title is missing. I also want people to start using dublincore

14:44:46 <_joshua> what else was there?

14:45:02 <_joshua> am I doing anything wrong? do you have any suggestions? i have tried to do this as openly and as cleanly as possible

14:53:24 * danbri wonders how the slashdotting of geourl went... did you get lots more data?

14:57:18 <_joshua> danbri: yes, a few thousand entries

14:57:24 <_joshua> one melted computer

14:57:34 <_joshua> geourl.org/status/ says what's going on

14:57:58 <JibberJim> ERROR!

14:58:12 <_joshua> FUCK

15:00:35 <danbri> Magic number checking on storable string failed at blib/lib/Storable.pm

15:00:38 <danbri> sounds painful

15:00:59 <danbri> It's only a matter of time before one of the foaf demos goes the same way

15:01:06 * danbri volunteers Jim for the dubious honour...

15:10:24 <_joshua> it's just a bad cache object i think

15:12:21 <JibberJim> Cheers danbri, maybe I'll just take foafnaut down, and donate the code to whoever hosts foafnaut.org :-)

15:15:24 <danbri> foafnaut.org is (as Damian just pointed out) run on the worst computer we have in active service!

15:16:30 <JibberJim> and you think jibbering.com is a good machine?

15:18:58 <_joshua> http://geourl.org/status/ works now

15:18:58 <dc_rdfig> D: http://geourl.org/status/ from _joshua

15:21:29 <_joshua> damnit sorry

15:21:57 <danbri> D:|GeoURL status page

15:21:57 <dc_rdfig> Titled item D.

15:22:10 <DanC> joshua, I'm generally quite liberal with my criticism. I looked at your system for more than 10 minutes and didn't find any real gripes.

15:22:15 <DanC> well done!

15:22:27 <danbri> D:See [http://geourl.org/|GeoURL home page] for more details.

15:22:27 <dc_rdfig> Added comment D1.

15:22:31 <sethl> DanC, w/ RDF/XML's new nodeID, does cwm need to create that existential quantifier "log:forSome" thing? or is there another reason that is around?

15:22:34 <DanC> using <title> if dc.title isn't there is the Right Thing. You seem to be doing The Right Thing in lots of cases.

15:23:02 <DanC> no, log:forSome is bogus, broken, and, with rdf:nodeID, unnecessary.

15:24:02 <sethl> great! I've taken it out and added in nodeID support to everything but RDF/XML output (input works now). plus, when the sink runs into anonymous nodes, log:forSome isn't generated anymore

15:24:34 <_joshua> DanC: please let me know if I could be doing other things more right, then

15:25:13 <DanC> sethl, you fixed cwm's RDF parser?

15:25:59 <sethl> danc: I wouldn't say fixed, I would say "hacked it more so I don't see log:forSome and bNodes work better". I want to run it through whatever tests it has to see what I broke

15:26:20 <DanC> re more things right: you could encourage the use of <head profile="http://geourl.org/somethingHere"> to ground the name ICBM in URI-space

15:26:23 <sethl> all I wanted was to input n3 and get back kosher rdf/xml :)

15:26:57 <DanC> yes, plese do run the tests, and submit a patch (to timbl, me, and www-archive+n3bugs@w3.org).

15:26:58 <JibberJim> joshua, an interactive SVG interface to the data would be great!

15:26:59 <danbri> joshua, we have an RDF advocacy issue w.r.t. fact that we want lots of sites to deploy RDF/XML rather than the old META tags, yet we don't want them deploying invalid HTML. Plain old embedding of RDF/XML in the HEAD of a Web page can break validation...

15:27:16 <DanC> he's not putting RDF/XML in HEAD, danbri.

15:27:38 <danbri> ...so using meta tags, with their meaning clarfied (as DanC suggests) with profile, is imho fine.

15:27:41 <DanC> I don't think META tags are all that bad.

15:27:42 * danbri realises that

15:27:53 * danbri neither, when their meaning is grounded via a profile

15:27:53 <DanC> ah; ok.

15:28:36 <_joshua> so using META with html is fine, and i should index RDF in XHTML?

15:28:48 <_joshua> All we need now is an RDFS for coordinates

15:28:49 <danbri> when the metadata is more complex than simple attrribute/value pairs, eg. if you want to include the names and contact details of several authors, using RDF/XML (by pointing to it with LINK REL or embedding and ignoring that it breaks validation) become more attractive options

15:28:51 * _joshua looks around expectantly

15:29:05 <danbri> One thing you could do, is re-expose the aggregated data as an RDF/XML dump...

15:29:27 <danbri> Yes, a lot of folk seem interested in that. I've lost track of what the state of the various vocabs is...

15:29:39 <JibberJim> we have lots of people interested in RDFS for co-ordinates we really should sort it out, maybe arrange an RDF Calendaring meeting like thing here some time?

15:30:16 <danbri> Jim, yes, that sounds like a good thing to schedule.

15:30:31 <_joshua> it'd be so much easier if someone would just tell me what to do :)

15:30:40 <_joshua> expose data as RDF? Hm

15:30:50 <danbri> Maybe we could use the same timeslot as the cal meeting, so we could get used to the channel being topical on Weds afternoons...

15:31:27 <danbri> Joshua, it almost works the other way around: if enough folk like you tell us what you've done (are doing...) we can shape an RDF vocab that reflects real world apps.

15:31:44 <danbri> And then we tell you how it looks in RDF/XML angle brackets, namespace URIs etc., by way of thank you.

15:31:49 * JibberJim can't make weds 18:00 though...

15:32:14 <_joshua> Ok

15:32:20 <_joshua> This isn't a real world app.

15:32:26 <_joshua> Actually there's already a GeoXML std

15:32:30 <danbri> But it got slashdotted ;-)

15:33:09 <_joshua> heh heh

15:33:11 <JibberJim> slashdotted is the real world...

15:33:12 <danbri> Maybe a session where we pull together as many urls and drafts and sites on the topic woudl be a good start?

15:33:19 <_joshua> Yeah suddenly people are coming out fo the woodwork to bother me about this

15:33:31 <_joshua> my big hope is that someone cooler starts indexing :)

15:33:49 <danbri> re 'expose data as RDF? Hm', would you be itnerested in trying that...?

15:34:04 <danbri> then maybe we can talk Jim and others into plotting it in SVG, etc etc.

15:34:44 <danbri> if indexers/aggregators expose what they find, makes it easier for application-oriented folk to get stuck into the part of the problem they're focussed on, instead of writing crawlers

15:35:11 <danbri>http://www.memepool.com/Subject/Cartography/

15:35:11 <dc_rdfig> E: http://www.memepool.com/Subject/Cartography/ from danbri

15:35:38 <danbri> E:|Memepool's Cartography section (22 articles)

15:35:38 <dc_rdfig> Titled item E.

15:35:40 <_joshua> Well, I already emit RSS

15:36:06 <_joshua> the whole http://geourl.org/rss091/ etc

15:36:49 <danbri> cool! (though RSS 1.0 is cooler ;)

15:37:54 <icepick-away> icepick-away is now known as icepick

15:38:16 <_joshua> One sec

15:39:48 <_joshua> danbri: get a rss091 link, change the 091 to 10 and see if it looks acceptable?

15:40:51 <_joshua> Too bad XML::RSS sucks so harshly

15:42:09 <JibberJim> _joshua, if you could create a genuine RDF version, that looked a bit like http://jibbering.com/rdf/airports.1?LHR;ORD for each entry I could create an interactive SVG version of it in a couple of minutes.

15:42:24 <danbri> tried http://geourl.org/rss10/?lat=51&lon=-3.00000000000001 but error

15:43:25 <_joshua> fixed

15:43:48 <danbri> cool

15:44:00 <danbri> the XSLT styling doesn't work, but that's not a suprise...

15:44:11 <danbri> (there are XSLTs around for rSS1.0 you could borrow...)

15:44:21 <_joshua> hrr.

15:44:50 <_joshua> FUCK

15:44:50 <_joshua> demo:377: error: Input is not proper UTF-8, indicate encoding !

15:46:53 <danbri> urk

15:47:27 <_joshua> try now?

15:48:53 <danbri> I get the rss showing as text; guess <?xml-stylesheet type="text/xsl" href="/rss10/rss10.xsl"?> isn't working for some reason

15:48:55 <danbri> this in mozilla

15:49:11 <_joshua> oh. sorry. i have no fucking clue what i'm doing, incidentally

15:49:42 <danbri> :)

15:49:52 <_joshua> making it up as i go along

15:50:19 <_joshua> last 4 days of traffic btw

15:50:20 <_joshua> 4184 06/Jan/2003

15:50:20 <_joshua> 15960 07/Jan/2003

15:50:20 <_joshua> 112796 08/Jan/2003

15:50:20 <_joshua> 27269 09/Jan/2003

15:52:55 <danbri> I don't know XSLT, but looking at the XSLT as Mozilla shows it to me: <xsl:value-of select="/rdf:RDF/channel/title"/>

15:53:11 <danbri> ....suggests that 'channel' and 'title' are in no namespace, rather than the RSS one.

15:54:00 * danbri googles for an example to copy from

15:57:13 <_joshua> does it not inherit the parent tag's namespace?

15:58:45 <danbri> In XSLT, I dunno... I'm asking wiser folk than I... back in a bit.

16:00:18 <danbri> logger, pointer?

16:00:18 <danbri> See http://ilrt.org/discovery/chatlogs/rdfig/2003-01-09#T16-00-18

16:15:17 <danbri> cool, http://geourl.org/rss10/?lat=51&lon=-3.00000000000001 works now in Mozilla

16:15:36 <danbri> thanks Joshua (and MaxF for the XSLT fixup)

16:15:44 <_joshua> sweet

16:15:53 <_joshua> ok what's next on the agenda

16:16:20 <danbri> hmm, what about crosslinks from FoafNaut to your page about each location? (Jim?)

16:16:39 <_joshua> first we gotta start indexing FOAFs

16:16:49 <danbri> Exposing some of the geo-info back out in the RSS feeds? (and then a datadump could just be a larger RSS feed with (literally) global scope...

16:17:17 <danbri> Jim's got an index of FOAFs; I meant, in his SVG map, make it so you can link across to your site...

16:19:05 <danbri> DanC, do you have a w3.org RDF/XML namespace for geo stuff, lat/long etc?

16:22:26 <DanC> sorta... I have a postoffice thingy and an xplanet thingy...

16:22:42 <DanC> has the benfoaf thing been chumped?

16:22:44 <danbri> don't think i've seen the postoffice one

16:22:51 <danbri> dc_rdfig:view

16:22:51 <dc_rdfig> A: WebOnt F2F meeting in Manchester 9-10 Jan (http://www.w3.org/2001/sw/WebOnt/)

16:22:52 <dc_rdfig> B: "Click to the clique", Ben Hammersley, Guardian (http://www.guardian.co.uk/online/story/0,3605,870848,00.html)

16:22:53 <dc_rdfig> C: Edd's ibm developerworks article on Friend-of-a-Friend, in Japanese (http://www-6.ibm.com/jp/developerworks/xml/020906/j_x-foaf.html)

16:22:54 <dc_rdfig> D: GeoURL status page (http://geourl.org/status/)

16:22:55 <dc_rdfig> E: Memepool's Cartography section (22 articles) (http://www.memepool.com/Subject/Cartography/)

16:22:58 <danbri> yes, it's B:

16:23:13 <DanC> earthMap: http://www.w3.org/2000/10/swap/pim/earthMap.n3

16:23:24 <DanC> post office: http://www.w3.org/2000/10/swap/pim/usps.n3

16:23:32 <danbri> Jim isn't reporting slashdot-like levels of Web hits from Guardian readers, fwiw.

16:25:04 <_joshua> heh heh

16:25:37 <DanC> crud; I neglected to cite the code that groks usps data (and prints mailingn labels) from usps.n3

16:25:43 <DanC> I cited the relevant code from earthMap.n3

16:26:25 <DanC> I use this to print mailing labels for a group my wife belongs to: http://www.w3.org/2000/10/swap/pim/jobieLabels.pl

16:27:03 <DanC> and I have some rules files to go from city names to lat/long...

16:27:07 * danbri most interested in the earthmap one, which has a some earthmapisms but 4 main properties that I guess we need

16:27:38 <DanC> rules: http://www.w3.org/2000/10/swap/pim/usPlace2LatLong.n3

16:28:23 * danbri notes that http://www.w3.org/2000/10/swap/pim/earthMap and the same url+.n3 both work, so assumes it could return RDF/XML by content negotiation if we fiddled around with server config directives

16:29:07 <danbri> BLURB:RDF geo vocab -- some existing vocab

16:29:07 <dc_rdfig> F: RDF geo vocab -- some existing vocab from danbri

16:29:13 * DanC tries to remember to publish .rdf too, but often neglects it

16:30:54 <danbri> F:DanC's [http://www.w3.org/2000/10/swap/pim/earthMap.n3|EarthMap] vocab (Xplanet-based); also [http://www.w3.org/2000/10/swap/pim/usps.n3|US post offices] and [http://www.w3.org/2000/10/swap/pim/usPlace2LatLong.n3|rdf rules for city names to lat/long conversion].

16:30:54 <dc_rdfig> Added comment F1.

16:33:08 <icepick> icepick is now known as icepick-away

16:33:39 <_joshua> blah, what we really need is jsut a latitude and a longitude with an ability to specify datum

16:33:52 <_joshua> and an altitude

16:34:14 <_joshua> three floats

16:34:50 <_joshua> labels and whatever should be handled by some other rdf whatever

16:34:56 <danbri> F:Also Jo Walsh's [http://space.frot.org/|spacenamespace] work (see [http://space.frot.org/etcon.html|forthcoming paper] at O'Reilly's Emerging Tech conference, 2003. See [http://space.frot.org/mudengland.html |model of england] (sample: [http://space.frot.org/a_place/Milton_Keynes|Milton Keynes lookup]).

16:34:56 <dc_rdfig> Added comment F2.

16:35:17 <danbri> Here's Jo's markup:

16:35:19 <danbri> <rdf:Description rdf:about="http://space.frot.org/a_place/Milton_Keynes">

16:35:20 <danbri>

16:35:20 <danbri> <place:knowhere_url rdf:resource="http://knowhere.co.uk/412.html"/>

16:35:20 <danbri>

16:35:20 <danbri> <place:name>Milton Keynes</place:name>

16:35:20 <danbri> <place:dd_long>-0.7000000</place:dd_long>

16:35:22 <danbri> <place:dd_lat>52.0333333</place:dd_lat>

16:35:24 <danbri> <place:type>Populated_Place</place:type>

16:35:26 <danbri>

16:35:28 <danbri> <place:is_contained_by rdf:resource="http://space.frot.org/a_place/Buckinghamshire"/>

16:35:30 <danbri>

16:35:33 <danbri> </rdf:Description>

16:35:46 <danbri> so yeah, a simple place RDF vocab that could be extended by others, seems plausible to me.

16:36:50 <_joshua> right, but a place might not haev a name

16:37:01 <_joshua> some rdf blob is pinned at location

16:37:11 <_joshua> Populated_place sounds like TGN

16:37:12 <_joshua> Clearly there needs to be a robots.txt equivalent - I propose retards.txt

16:37:12 <_joshua> that prevents idiots from attempting to be communicating with me

16:37:25 <_joshua> I'm getting some seriously dumb email

16:37:38 <danbri> heh, what did they ask?

16:37:51 <_joshua> > I think tagging the pages is a mistake. The geographical

16:37:51 <_joshua> > location from which a page is served is not a property of

16:37:51 <_joshua> > the page content. So it would be better to store the

16:37:51 <_joshua> > location information into a separate file, say,

16:37:51 <_joshua> > location.txt.

16:37:55 * danbri noticed a French Weblog appearing down near Antarctica, btw.

16:38:01 <_joshua> yeah

16:38:07 <_joshua> just a few people in antarctica and mongolioa

16:38:29 <danbri> Your page avoids that, I thought. Says 'people prolly not going to care where the server is'

16:38:37 <_joshua> Yes

16:38:52 <_joshua> You would not fail the retards.txt test

16:38:55 * danbri shouldn't care where http://www.w3.org/people/danbri/ is, as its in 3 places (us, france, japan)

16:39:02 <danbri> don't bank on it!

16:39:07 <_joshua> The question is, where is danbri?

16:39:17 <_joshua> that is infinitely more interesting than where the MACHINE is

16:39:29 * danbri nods

16:39:49 <danbri> I liked DanC's use of nearestAirport as a balance between interesting data vs privacy protection

16:40:10 <danbri> I have gps for my visor, so could have a bunch more precise data than nearestAirport... but even that is interesting.

16:40:30 <danbri> I found out that Ben Hammersley had moved to Sweden, for eg., by seeing him on the SVG foafmap.

16:45:41 <_joshua> ehhh. people who want the privacy can just by not including it

16:46:35 <JibberJim> but it's different level of info saying I live near London Heathrow Airport, and I live at 0N 0E...

16:46:52 <danbri> Yup, especially when you chuck in calendar sharing at the same time

16:47:12 <JibberJim> and missile guidance systems.

16:47:47 <_joshua> nod

16:47:57 <_joshua> even so

16:48:16 <danbri> heh yeah, see JanG's 'burglar bill' RDF calendaring use case, in http://www.w3.org/2001/sw/Europe/reports/dev_workshop_report_2/

16:48:17 <danbri> [[

16:48:18 <danbri> Jan Grant's usecase - trust and security of information

16:48:19 <danbri> Jan made a point about security and trust, by suggesting a query: ' find two people in a house worth 200,000 pounds where they're both away for a week'. Even knowing where you are might be valuable information, for example for a CEO

16:48:19 <danbri> ]]

16:48:32 <danbri> But I agree that nearest airport info on its own is pretty impoverished

16:48:52 <danbri> Marking up local services (shops etc) would be a good app for more precise info.

16:49:26 <danbri> Libby and I started to photograph and GPS-catalogue shops in nearest shopping street in Bristol...

16:50:20 <_joshua> Right. If you're concerned, don't markup or be vague, etc

16:50:23 <danbri> not sure which namespace(s) to use for that though. Just lat/long is enough, any fiddle extras (basedNear etc) could shove into FOAF or other scruffier apps

16:50:55 <danbri> If we catalogue a dozen shops in a row, I'm not sure whether to put individual GPS readings on them (pretty time consuming) or just say they're all near some common marker spot.

16:51:12 <_joshua> nod

16:51:21 <_joshua> i think you would use other schemas for other kinds of information

16:51:35 <_joshua> just really let you specify a coordinate on a given thingy

16:52:53 <_joshua> so you could phrase

16:53:06 <_joshua> person -near-> airport -at-> latlong

16:53:08 <danbri> So how about we make a w3.org namespace that really has just lat/long/alt and nothing else ever? as a common basis for experimentation, since more fancy namespaces can have RDF sub-properties that map to the base 3.

16:53:11 <_joshua> person -at-> latlong

16:53:11 * danbri nods

16:53:28 <_joshua> store -at-> latlong

16:53:28 <JibberJim> we want elavation too.

16:53:29 <_joshua> etc

16:53:33 <danbri> events will ikely be a common one (cos they don't move around much)

16:53:33 <_joshua> yeah, whichever

16:53:38 <JibberJim> which means we need to decide on a datum.

16:53:44 <_joshua> nah, specify a datum

16:53:48 <danbri> can you explain for

16:53:51 <_joshua> er, allow user to specify a datum

16:53:51 <danbri> yeah what he said

16:54:10 * JibberJim prefers to encourage one, if we're just going to have a simple one.

16:54:25 <danbri> er, can you explain a bit more?

16:54:39 <_joshua> geodetic reference datum

16:54:41 <danbri> is this some subtlety of geo information?

16:54:50 <_joshua> basically a) the earth isn't a sphere but lat/long assumes a sphere

16:54:55 <_joshua> b) you need an exactly measured origin

16:54:56 * danbri nods

16:55:06 <_joshua> take your GPS and it will let you pick different datum

16:55:14 <danbri> what's the most common practice?

16:55:20 <JibberJim> Exactly, so i

16:55:31 <JibberJim> what is the danger in saying use WGS84?

16:55:32 <_joshua> WGS84 in the world but all the us govt data us NAD83

16:55:37 <danbri> logger, pointer?

16:55:37 <danbri> See http://ilrt.org/discovery/chatlogs/rdfig/2003-01-09#T16-55-37

16:56:05 <danbri> F:See [http://ilrt.org/discovery/chatlogs/rdfig/2003-01-09#T16-55-37|rdfig chat on geo stuff] for more context.

16:56:05 <dc_rdfig> Added comment F3.

16:56:15 <_joshua> "WGS84 43.123123123 -73.123123123"

16:56:26 <_joshua> i dunno. there's lots of formats

16:56:54 <danbri> How big a difference does it make? hmm this complicates the model a bucnh already

16:57:00 <_joshua> guess i'll work on it

16:57:03 <JibberJim> Yes, which is why I'd like a simple predicate version that we encourage use of being lat/lon in WGS 84 and a seperate predicates where you can also specify the datum.

16:57:17 <_joshua> TIGER/Line claims there's no huge difference

16:57:25 <_joshua> I concur with Jim

16:57:44 <danbri> are they all mechanically interconvertable?

16:57:59 <JibberJim> I think so, if you have the info to do it.

16:58:08 <JibberJim> (and the maths)

16:58:15 <danbri> in which case, picking a likely leader for the three-property namespace seems simplest, best.

16:58:17 <_joshua> It can be done with math

16:58:41 <JibberJim> Is there any ISO stuff?

16:59:01 <_joshua> Dunno

16:59:06 <_joshua> there's a GeoXML standard

16:59:39 <_joshua>http://map.sdsu.edu/gisbook/ch7.htm

16:59:39 <dc_rdfig> G: http://map.sdsu.edu/gisbook/ch7.htm from _joshua

17:01:00 <_joshua> there are content standards

17:02:42 <danbri> how much do we need to know to make the three-property namespace?

17:03:00 * danbri hoping not much more, and the smaller/simpler it is, the less we annoy other folk whose specs I've not read properly...

17:03:40 <_joshua> well there's specs for how stuff is represented

17:03:48 <_joshua> 47deg 123' 131413" etc

17:03:52 <_joshua> or floating point?

17:04:01 <_joshua> I guess just normalize it?

17:04:48 <_joshua> you know

17:04:55 <_joshua> nobody's done an A/S/L into GeoURL

17:06:05 <_joshua> good job

17:06:13 <danb_lap> thank you!

17:06:36 <_joshua> I am waiting for a "lonely 23/f seeks m for lwitp" posted somewhere

17:06:54 <_joshua> "get on the manboat"

17:06:58 <danb_lap> c/o geourl?

17:07:24 <_joshua> yeah

17:07:30 <_joshua> nobody's done ti but i expect it

17:07:43 <_joshua> "set sail for dick"

17:07:49 <danb_lap> several such things are only a matter of time w/ foaf too...

17:07:56 <_joshua> encourage it

17:07:59 <danb_lap> sleptWith etc...

17:09:29 <_joshua>http://www.attrition.org/hosted/sexchart/

17:09:29 <dc_rdfig> H: http://www.attrition.org/hosted/sexchart/ from _joshua

17:10:18 <_joshua> H: friend of a friend with "privileges"

17:10:18 <dc_rdfig> Added comment H1.

17:10:37 <danb_lap> H:The sexchart

17:10:38 <dc_rdfig> Added comment H2.

17:10:47 <danb_lap> H:AsciiArt as graph syntax...

17:10:47 <dc_rdfig> Added comment H3.

17:11:00 <_joshua> clearly you need to have a FOAF for various levels of body fluid interaction

17:12:31 <_joshua> semen, mucus, urine,e tc

17:13:47 <danb_lap> *someone* might, I'm not finding the use cases right now... ;)

17:14:20 <JibberJim> You've obviously not appreciated the dating use cases...

17:14:26 <danb_lap> i don't doubt that it'll happen, i'm just pre-distancing myself from the blame

17:14:41 <danb_lap> the path-finding algorithms are pretty re-usable too

17:15:22 <_joshua> heh

17:15:24 <danb_lap>http://www.apocalypse.org/pub/u/danbri/fc/

17:15:25 <dc_rdfig> I: http://www.apocalypse.org/pub/u/danbri/fc/ from danb_lap

17:15:30 <JibberJim> Ah research into STD transmission methods?

17:15:38 <danb_lap> I:|We still don't talk about semanticwebclub

17:15:38 <dc_rdfig> Titled item I.

17:16:04 <danb_lap> I:who-fought-who scripts, c/o ruby transliteration of java co-depiction tool

17:16:05 <dc_rdfig> Added comment I1.

17:16:29 <danb_lap> all the scripts you need are there, fwiw

17:17:54 <_joshua> I can probably hack to gether a BFS tree anyway

17:18:13 <danb_lap> BFS?

17:18:44 <_joshua> breadth first search

17:19:00 * danb_lap nods; i just copied damian's code

17:19:35 <_joshua> oh bah, rally?

17:20:52 <danb_lap> ...http://www.apocalypse.org/pub/u/danbri/fc/lib/pathfinder.rb

17:20:52 <danb_lap> # Mindlessly-transliterated Ruby version

17:20:52 <danb_lap> # and associated bugs/glitches by Dan.

17:21:40 <JibberJim> Mindlessly transliterated javascript version in foafnaut

17:22:52 <_joshua> Pfft

17:23:07 <_joshua> it's like 3 lines of code

17:28:31 <danb_lap> that's good, right?

17:28:38 * danb_lap heads off for the evening

17:29:07 <_joshua> yep

18:15:57 <icepick-away> icepick-away is now known as icepick

18:24:43 <icepick> icepick is now known as icepick-away

19:02:21 <icepick-away> icepick-away is now known as icepick

19:27:46 <icepick> icepick is now known as icepick-away

19:42:48 <rhoads> What is up with RDF Schema? Last time I checked (~1 year ago) the RDF Schema spec was 50 pages long with several examples. Now its just a list of terms with no examples. Is the IG moving away from this? Should I be looking at OWL?

19:43:39 <dajobe> the examples are in the RDF primer

19:44:16 <dajobe> it is possible you may want some OWL things, that's depends on your application

19:44:52 <dajobe> there are several OWL drafts that could aid you in deciding that, probably the OWL Guide would be the first stop

19:47:56 <rhoads> As you may guess, I'm a newbie. That said, some additional questions: (a) Does OWL build on RDFSchema, or is it a replacement? (b) Will the concepts/constructs in RDFSchema be with us "forever"?

19:48:21 <dajobe> a - yes sort of; it does add restrictions, as higher languages do. read the owl docs for details

19:48:56 <dajobe> b - forever is a rather long time

20:38:26 <LotR> I just read the logs of the ical meet-up, and there's mention of something called sherlock in it (with url sherlock://icalshare.com/sherlock/channel.xml). Does anyone here know what this is, and where I can find more info on it?

20:40:00 <thelsdj> hrm sherlock is mac os x find command i think

20:44:02 <LotR> oh

20:44:51 <Morbus> right.

20:45:00 <LotR> so it would contain info on what kind of data was available on the (webdav enabled part of) the site I guess

20:45:24 <Morbus> based on the URL, i'd think it was a searching facility for the icalshare.com site.

20:45:41 <Morbus> sherlock allows you to search for airlines, get movie showtimes, look for people, books, etc.

20:45:51 <Morbus>http://www.apple.com/macosx/jaguar/sherlock.html

20:45:52 <dc_rdfig> J: http://www.apple.com/macosx/jaguar/sherlock.html from Morbus

20:47:46 * LotR hunts for technical information on it

20:48:24 <LotR> google++ :)

20:48:27 <Morbus> you want an sdk?

20:48:31 <Morbus> i've got the link for it.

20:49:23 <LotR> if it works on linux

20:49:33 <Morbus> nope.

20:49:38 <LotR> I don't actually own an apple :)

20:49:47 <Morbus> i half suspected as much ;)

20:50:29 <LotR> it sounds like an interesting technology to play with

20:52:07 <LotR> oh, it's just a search engine format description language it seems

20:52:13 <Morbus> not anymore.

20:52:15 <Morbus> that was sherlock 2.

20:52:22 <Morbus> sherlock 3 has little to do with sherlock 2.

20:52:45 <LotR> oh

20:52:58 * LotR looks for more recent data

20:53:39 <Morbus> see http://developer.apple.com/macosx/sherlock/

20:54:07 <Morbus> and maybe http://www.sherlockers.com/

20:54:44 <LotR> thanks (and google++ for giving me the first too :)

20:57:36 * LotR wonders if someone is already working on making mozilla grok that

21:05:14 <Morbus> there's a mozilla plugin to grok sherlock 2 files.

21:05:20 <Morbus> but nothing i've seen yet for sherlock 3

21:16:59 <icepick-away> icepick-away is now known as icepick

21:58:54 <icepick> icepick is now known as icepick-away