Friend of a Friend (FOAF) IRC Chat Logs for 2003-10-22

00:04:57 <mattb> nn

00:16:08 <NetHome> NetHome is now known as Netminder

00:38:50 <edd>http://conferences.oreillynet.com/pub/w/28/track_social.html

00:38:59 <edd> social software sched foor ETCON

00:39:10 <edd> notice FOAF gets the real butt end of the schedule

00:42:06 <Schuyler> grr

00:42:58 <edd> i don't get why there's an XQuery session in the ched

00:42:59 <edd> sched.

00:43:31 <edd> and also how come rael gets three talking slots?

00:44:40 <edd> i thought there was going to be a Dashboard session that either Nat F or I would do

00:44:51 <edd> Rael even asked me to submit an abstract cos he wanted one

00:45:08 <edd> and it was pushed into a FOAF 5 minute lightning talk, oddly.

00:45:11 * edd Doesn't Get It

00:46:32 <Schuyler> I thought it was a bit strange that Rael was giving all those talks, yes

00:46:58 <Schuyler> and nothing on Dashboard?

00:46:58 * Schuyler shakes head

00:47:02 <Schuyler> pitifully little on the semweb in contrast to what's been submitted

00:47:17 <edd> Too right. And some rather less than emerging talks.

00:49:20 <edd> Hmm, Rael's sessions are panels

00:49:30 <edd> so I guess I take back the schedule hogging accusation

00:54:25 <edd> Schuyler: hmm, on the at-a-glance i even see empty boxes

00:56:22 <Schuyler> yes. so it's likely that more talks will be added.

00:56:29 <Schuyler> I just got a raft of rejections a moment ago, though.

01:14:18 * JibberJim got accepted for a foaf lightning talk...

01:14:45 <Zack> nice!

01:15:04 <JibberJim> but I doubt I have the cash to drop on it right now... especially as confirm by friday...

01:15:10 <Zack> hey jim - any thing much happening with that geo overlay foaf mapper thing/

01:15:18 <edd> yeah

01:15:18 <Zack> eeep :/

01:15:22 * edd considers boycotting it

01:15:33 <edd> or, less contentiously, just not going :)

01:15:50 * JibberJim just thinks that the cash would go a long way to a digital SLR...

01:16:05 <edd> oh yummy

01:16:14 <edd> i played with nat friedman's the other week. it was so nice.

01:16:17 <JibberJim> Nah Edd, Boycott it! go on...

01:16:22 <edd> the d10 or something

01:16:28 <JibberJim> Yeah I played with Morten's (other than the weight...)

01:17:04 <edd> Righto

01:17:09 <edd> Time for me to expire.

01:17:12 <JibberJim> night Edd.

01:17:20 <edd> One night I'll go to bed without having found something that makes me grumpy.

01:17:27 <edd> I can only aspire!

01:17:28 <edd> nn

01:17:48 <Mutiny> dream a little dream, edd. ;)

01:18:03 <JibberJim> Very little I'm afraid Zack, erm very little is happening with many of my interesting projects.

01:22:31 <JibberJim> hmm, now that's just weird anyone know anything about "thespameater.com" I just got an email, asking a perfectly reasonably question about my javascript RDF parser but the from address is "The Spam Eater . com" <anti.spam@thespameater.com>

01:22:31 <JibberJim>

01:22:57 <JibberJim> I'm going to ignore it, if they can't send me a proper email address, hard luck.

05:10:03 <KevinMarks> KevinMarks is now known as _KevinMarksMI5

06:08:49 <_KevinMarksMI5> _KevinMarksMI5 is now known as KevinMarks

08:49:18 * maxf waves

09:13:05 <maxf> hi libby

09:13:47 <libby> hey maxf

09:14:00 <libby> when are you arriving?

09:14:23 <maxf> flying at 4, so should be around at 5:30

09:17:36 <libby> cool :)

09:21:07 <maxf> want something from France?

09:21:30 <libby> heh

09:21:44 <libby> nice stuff ;)

09:22:06 <maxf> ok :)

09:43:08 <libby> re etcon, my talk was rejected, so I (grumpily) think we should have foafcon at same time (somewhere else)

09:45:24 <dajobe> so the foaf sessions are the last hour on the final day of etcon

09:45:38 <libby> a less personal grumble is that we shoudl have had about 5 sessions, not two.

09:45:50 <libby> yeah

09:46:12 <libby> ble

09:46:58 <libby> we dont need them!

09:47:18 <dajobe> we've got our own foaf's and the triples to prove it

09:50:58 <libby> yeah

09:51:13 <JimH> libby: hwabout foafcon in a pub in England somewhere?

09:51:22 <libby> sounds good :)

09:51:38 <JimH> sounds appropriate if nothing else :-)

09:51:55 <libby> indeed. we can get those wn:beer tshirts made

09:59:28 <maxf> +1 ;)

10:00:37 * mattb votes for a parallel foafcon, can't afford to go to etcon :)

10:01:17 * libby neither probably

10:01:29 <Mutiny> seems pretty exclusive

10:01:55 <libby> ?

10:02:28 <libby> well, I expact that some peoplke will want to go to etcon. I was sorta joking. but we shoudl have a foafcon anyway

10:03:21 <mattb> yeah, lots of good foafy people will be there

10:03:29 <mattb> we should arrange something nearby in time

10:03:34 <libby> yeah

10:09:00 <libby> ooh, maxf, if you see any saucisson, bring that! ;)

10:10:57 <maxf> libby: too late to go to supermarket now, I'll see what I find at the airport

10:11:31 <maxf> outrageously overpriced saucisson for the Rivera jet-set, no doubt.

10:12:02 <libby> heh, don;t worry! just a thought....damian just reminded me of how much a like it that's all

10:13:09 <libby> public transport data people: http://www.fluidtime.net/use/ptt.html

10:13:13 <libby> in turin though

10:13:16 <libby> not bristol

10:14:15 <JibberJim> Hmm, getting tracklogs of the network sounds more interesting in Turin

10:17:51 <libby> the interfaces to teh data look pretty neat: http://www.fluidtime.net/use/interfaces.html

10:49:22 <zool> morning

10:50:46 <JibberJim> mornin'

10:50:54 <mattb> morning zool

10:51:04 <mattb> glasses sorted?

10:51:54 * zool nods, though they are probably a bit impractically small, i liked them; vision will be an exciting experience

10:52:22 * zool scrollbacks... did danbri's fear-of-a-foaf-planet talk get etcon-accepted?

10:52:31 <JibberJim> YEs

10:52:50 <zool> i had my 'botworld' proposal turned down, too :/ too impractical and rambly i suspect

10:53:00 <dajobe> me too

10:53:05 <JimH> ETCon sessions here - http://conferences.oreillynet.com/pub/w/28/etech04_grid.html

10:53:07 <zool> perhaps too far i n the future to be called by rights 'emerging' ;)

10:53:10 <zool> ah cool

10:53:33 <JimH> this one sounds nice - http://conferences.oreillynet.com/cs/et2004/view/e_sess/4852

10:53:37 <JibberJim> Judging by some of them zool, not far in the past enough to be called emerging

10:53:41 <zool> i decided not to send another one that would have clashed topically with dan's, so i'm bloody glad he got a slot

10:53:48 * zool frowns

10:54:04 <dajobe> oh it's not on the last day, it's at the end of Wed

10:54:14 <JimH> FOAF Lightning Talks - 5min lunacy sessions

10:54:26 * edd grumbles a lot

10:54:31 <edd> considering not going at all

10:54:33 <libby> aw

10:54:34 <zool> oooh, matt webb

10:55:26 <dajobe> ?Y?know ? For Kids!?: Social Software for Children - Fiona Romeo, BBCi"

10:55:38 <edd> the bbc seems to have packed in speakers

10:55:51 <JimH> maybe you lot should try to mug Russ for a snippet of his 3.5 hr session on the first day

10:56:02 <JibberJim> If there are lots of FOAF lightning talks - why don't we know who they are?

10:56:18 <edd> Well, danbri and i are meant to be hosting it

10:56:22 <edd> and we don't know who they are yet either

10:56:30 <edd> mortenf is one

10:56:32 <edd> i had another

10:56:40 <JibberJim> and me

10:56:45 <JibberJim> but that makes 3

10:57:03 <dajobe> they are listed at http://conferences.oreillynet.com/cs/et2004/view/e_sess/4852

10:57:18 <libby> damian, liz, josh, edd again

10:57:40 * libby doesn't think damian planning to go

10:57:52 <libby> unfortunately

10:58:06 <JimH> Marc Canter too

10:58:16 <libby> oh yes

10:58:47 * edd doesn't understand how dashboard got crammed into 5 minute FOAF

10:59:23 <zool> ditto re most of that set

10:59:31 <libby> indeed

11:00:01 <dajobe> and 45 minutes on "Echo wiki: lessons learned"

11:00:04 * dajobe boggles

11:00:08 <edd> dajobe: exactly.

11:00:26 <libby> well, probbably be a fun session, the foaf bit. probably can't justify going tho. and skint...

11:00:48 <edd> right, i can't justify going for a 5 minute talk

11:00:57 <libby> yeah

11:01:01 <libby> heyho

11:01:11 <JibberJim> 2 5 minute talks Edd...

11:01:14 <edd> we could always sched a meet for bristol at the same time

11:01:18 <edd> :-)

11:01:30 <libby> yeah, 10 minutes!

11:01:51 <libby> well, we were kinda discussing this earlier, probably more politic to ahve it just after or just before

11:02:04 <libby> but yeah, that'd be fun... :)

11:02:28 <libby> is someone going to get wn:beer is my foaf:friend tshirts done or shall I?

11:07:48 * JibberJim curses the BBC

11:09:48 <edd> libby: you would appear to be the expert :)

11:14:39 * JibberJim wonders who I have to have pictures of in compromising situations to make them broadcast test cricket matches, relying on IRC commentary is just not acceptable :-(

11:15:29 <edd> JibberJim: aye, :((((((((9

11:16:21 <libby> not on the radio?

11:16:48 <Jibbler> someone donated $3 on my website, drinks are on me!

11:16:54 <libby> wow :)

11:17:01 <JibberJim> No, not even on the nice shiny new Digital channels

11:17:38 <dajobe> probably on Sky satellite channels. Blame them.

11:18:00 <JibberJim> TV yeah, but no Radio coverage, Sky don't buy radio rights

11:39:53 <libby> heya shellac

11:42:35 <shellac> hi

11:43:09 * shellac about to tell etcon that I won't be hauling myself halfway around world for 5 min talk

11:45:10 * maxf recommends the Bristol hotel in San Diego. Free network, reasonable prices...

11:45:41 <JibberJim> ephidrina?

11:50:58 <ephidrina> yes?

11:51:52 <JibberJim> etcon, you attending or joining most of the rest of us in not...

11:51:54 * ephidrina also reluctant to fly to san diego for 5 mins

11:52:07 <libby> this is lame

11:52:10 <libby> :(

11:52:11 <ephidrina> is this a mass boycott?

11:52:23 <libby> I hink more of a practical boycott

11:53:26 <ephidrina> so did anyone get more than 5 minutes??

11:53:35 <libby> dan.... :)

11:54:48 * ephidrina chortles

11:57:15 <zool> well, i think i will wind up going anyway... codecon 10 days after it, it'd be nice to be in california

11:57:47 <JibberJim> Yes, but you go to every conference zool

11:58:00 <zool> i can't say no, i realised.

11:59:12 <JibberJim> I could say you should go to CA, but you'll read that as California and not Conferences Anonymous

11:59:59 * zool giggles quietly

12:04:51 <GregElin> Hey all.

12:05:19 <JibberJim> Hey GregElin

12:05:47 * JibberJim wants to schedule another image chat (to get forced to finish off the things I agreed to do after the last ones)

12:08:00 <GregElin> That sounds great! I think we should do it early/mid November?

12:22:19 <JibberJim> I'd guess mid now, but yep.

12:22:56 <GregElin> Cool. Mid Nov. Let's float that and see if it works for people.

12:23:16 * danbri_ flies to japan Nov 12, so 10/11ish prolly best for me.

12:23:39 <danbri_> If we want to do this as another RDF Interest Group scheduled chat, prolly best to migrate this into #rdfig so people not suprised...

12:24:16 <GregElin> Sounds great to me. Want to post it to that group?

12:24:38 <danbri_> maybe scratch out some agenda ideas in #rdfig?

12:24:49 <danbri_> one would be to have someone from Adobe/XMP come along...

12:24:59 <danbri_> eric miller has contacts there that i should follow up

12:25:09 <GregElin> I just joined #rdfsig.

12:25:12 <GregElin> Let's chat there?

13:15:05 <JimH> libby: I'm looking for a Foaf viewer that I thought was on swordfish, it shows an image for a person and two lists of people (knows and known by), any ideas?

13:16:24 <JimH> it's not eikeon's one - although the layout is similar

13:18:30 <JimH> found it - http://swordfish.rdfweb.org/rweb/who

13:29:16 <zool> flash get

13:36:48 <zool> p><STRONG>Something is wrong with the database, pester Jim, or try again later!</STRONG></p>

13:37:40 * darobin thinks zool is choosing option one

13:38:14 <zool> well, i thought i'd give it a punt. unworking-database may have been the lamentable situation for a while

13:38:32 <darobin> hahaah

13:38:36 <darobin> he fled!

13:38:39 <zool> er, also the error page was sent with the wrong mime type :/

13:38:42 <zool> soryr JibberJim

13:38:55 * zool smiles

13:39:03 <JibberJim> What did I do?

13:39:21 <zool> break foafnaut's brane

13:39:29 <darobin> I was hoping zool would pester you for the sheer entertainment value of it

13:39:58 <JibberJim> The foafnaut update error page was sent with wrong mime-type?

13:40:11 <JibberJim> Yeah foafnaut's brain is completely dead

13:40:21 <zool> i put my foaf file at a second url while the first one is down, this despite it having been down two months only now. wanting to get the other into the system, added to the wiki etc

13:40:24 * zool nods

13:40:31 <zool> i should try a scutter run soon

13:40:52 <zool> hey esigler

13:44:47 * JibberJim starts fixing foafnauts brain by getting a fresh scuttering

13:45:25 <esigler> Hola zool.

13:45:45 <esigler> What's up in your neck of the woods?

13:56:29 <zool> ah, this and that. the usual mapping shenanigans, bot writing, yak shaving

14:05:11 <esigler> Did the yak have a say in that?

14:16:05 <libby> JimH, yep, I've got one of those

14:16:56 <libby> - http://swordfish.rdfweb.org/rweb/who

14:17:28 <libby> add stuff to it here: http://swordfish.rdfweb.org/rweb/addurl

14:19:15 <JimH> Thanks Libby, I'd found it, I hope I didn't kill your box by seeing if it'd cope with Thomas Power...

14:19:21 <libby> heheheh

14:19:23 <libby> uhoh

14:19:27 <libby> seems ok

14:19:34 <nmg> morning all

14:19:45 <libby> heya nick! congratulations :)

14:20:08 <libby> did it work JimH - could you see it?

14:20:27 <nmg> ta.

14:20:33 <JimH> yeah it worked fine, took a while to complete though

14:20:42 * nmg is hungover this morning :)

14:20:42 * libby not surprised :)

14:20:47 <libby> heh

14:21:12 <nmg> new public url for aktivespace: http://triplestore.aktors.org/SemanticWebChallenge/

14:21:21 <zool> yay

14:21:57 <libby> what's the power url?

14:23:03 <nmg> power url?

14:23:10 <libby> thomas power

14:23:10 * nmg is feeling a little slow this morning

14:23:16 <libby> networked extroidinaire

14:23:22 <libby> networker I mean

14:23:24 <libby> aw

14:23:35 <nmg> oh, him. the one that knows 24,000 people (or whatever it was)

14:23:54 <JimH> Libby - http://swordfish.rdfweb.org/rweb/who?mboxsha1=5558502bef1b63304d278891cb71e2dfe959327e

14:24:09 <nmg> If I made a point of emailing everyone that registered on my community portal, I'm sure I could rival that...

14:24:52 <libby> ta JimH

14:25:07 <JimH> indeed, in fact he does more than emailing them, he adds these people to his Foaf almost automatically too

14:25:19 <libby> hm it no workie

14:25:41 <libby> ah missed a bit off

14:26:11 <nmg> jiimh: which I would argue is not really in the spirit of the foaf:knows relation

14:26:41 <libby> heh, layout doesn;t work too well with that many!

14:27:14 * libby surprised it works at all!

14:27:53 <mattb> gosh, aaron morton posts to rss-dev about audio streaming and rss

14:27:59 <mattb> he works for the bbc and is interested in rdf

14:28:18 <JimH> nmg: I don't think it's in the spirit of things either, but there's a lot about ecademy that I find baffling

14:28:27 * mattb recommends that license-fee payers tell the bbc about the semantic web

14:28:32 * nmg grins

14:29:04 * JimH reckons license fee payers should get a rebate for this act :-)

15:03:15 <libby> heya ndw

15:05:06 <ndw> Hi Libby!

15:05:59 <libby> how are you doing?

15:06:51 <ndw> I'm ok. Desperately behind after having my laptop out of commission for six days.

15:07:03 <libby> oh man

15:07:06 <libby> that's bad...

15:07:21 <libby> mines being fixed at the moment, but I borrowed another

15:08:48 <ndw> it was dreadful. http://norman.walsh.name/2003/10/20/luck

15:13:09 <libby> oh god! poor you :(

15:14:38 <ndw> Well, my new disk is bigger and faster. Maybe that's something... :-)

15:14:44 <edd> Poor ndw. Are you deluged in friends telling you you ought to run OS X

15:14:50 <edd> Be happy, ignorant and poorer.,

15:15:21 <mattb> does osx have cable-yank protection these days?

15:15:24 <ndw> For a couple of days, I thought the machine was a lot more damaged. If it was replaced, it would probably have been with an OSX box.

15:15:33 <mattb> extra shock proofing around the bytes so they don't break when the disk is rattled

15:15:54 <ndw> Someone told me the new Thinkpads actually have a motion sensor and park the disk if it is jossled

15:16:35 <mattb> reminds me of that new minolta(?) digicam with the camera-shake protection

15:16:41 <mattb> some sort of floating lens contraption

15:16:42 <edd> Oh my. Do they have dribble protection on the keynoard too?

15:23:01 <libby> tea and wine protection woudl be good

15:28:17 * edd glances at the winestain down the side of one of his machines

15:28:20 <edd> damn waste of wine

15:30:39 * libby admires http://norman.walsh.name/2003/10/13/pubsigns

15:31:18 <libby> sorta goes with http://swordfish.rdfweb.org/discovery/2001/08/codepict/querywn.jsp?wn=pub

15:34:15 <nmg> but where's the provenance? which pubs had those signs? ;)

15:36:09 <ndw> If you click on the signs and follow your nose, you should get to resource that identifies the city where the pub occurs. Is that close enough?

15:37:49 <ndw> Hmm. I don't have wn:Pub in there yet though...

15:38:07 <ndw> Part of the problem with wordnet is that it's big enough to have the "you can't find it if you don't know what you're looking for" problem

15:38:11 <nmg> cool, ta

15:38:15 <libby> yeah

15:38:41 <libby> it's amazing what in there though. rambutan for example.

15:39:02 <ndw> It appears that Hoary Marmot is gone from 2.0.

15:39:33 <libby> no!

15:39:41 * libby gutted

15:39:49 <ndw> Apparently. I couldn't find it, anyway.

16:01:51 * JibJimMob waves

16:02:00 <libby> heya JibJimMob

16:03:50 <JimH> GRPS Jim?

16:04:02 <JibJimMob> Yes

16:04:36 <JibJimMob> How do i do a private msg?

16:04:55 <edd> /msg

16:04:59 <JimH> wireless IRC on 3650?

16:06:00 <JibJimMob> Msg didn't work, yes jim

16:06:28 <JimH> one sec I'll ask over in #mobi

16:06:31 * edd wonders if symirc would deploy ok on the 3650?

16:06:38 <edd> it's just java stuffs, innit?

16:08:08 <JibJimMob> Privmsg worked

16:08:47 <JimH> edd: if it's just Java it won't work on the 3650

16:09:22 <JimH> J2ME sockets on the 3650 are useless

16:09:52 <JimH> so it's probably a Personal Java app on the P800

16:13:53 <JibJimMob> Do the other lightning talk folk need to let them know by fri?

16:14:22 <edd> Yeah

16:14:34 <edd> I'm contemplating exactly what form of "shove it up your ****" to use

16:14:48 <mattb> fewer asterisks, more characters

16:17:24 <JibJimMob> Shove it up your fat ***

16:17:30 <JibJimMob> ?

16:17:58 <JibJimMob> More american?

16:20:02 <JimH> chaps, we're English, use the *r** spellig please :-)

16:20:12 <ndw> lol

16:20:17 <JibJimMob> Bye!

16:22:12 <edd> when i'm prime minister all this publicly logged bile will come back to haunt me

16:22:32 <mattb> time to declare your policies on tax in an [off] line

16:23:28 <edd> not much point in [off] when at least 50% of the channel are logging

16:23:40 <edd> besides, if you can say it to 25 people, you've said it to the world

16:24:03 <edd> only reason for [off] is not publishing othe rpeople's secrets i suppose

16:24:06 * libby thinks its a shame about etcon. Maybe they will reconsider. be good to have some of us eurofoafers out there

16:24:07 * ndw sees the headline now "American shames PM: he told them to shove it up their *r**!"

16:24:09 <edd> like nose picking, that sort of thing

16:24:34 <JimH> it's better to be a pm with honest straight forward opinions on the record than the usual lying evasive subterfuge...

16:24:46 * ndw mutters, who the **** sent me a 4.5mb email attachment...

16:25:02 <edd> gotta be a uit

16:25:05 <edd> s

16:25:33 <ndw> It's going to be a picture of some powerboat that some fellow employee is selling. I betcha.

16:34:45 <nmg> nmg is now known as nmg_lunching

16:36:36 <edd> gee ndw, have you not told them about this web thing?

16:45:27 <mattb> time to go

16:52:54 <ndw> edd, i've mentioned it, but it doesn't always click. Uh. Pun intended. Maybe.

16:56:50 <ndw> In this case it was a 4.5Mb wedding invitation. In India. How...sweet.

16:56:56 * ndw isn't too cynical, is he? Nah.

17:01:06 * mortenf concurs with the pre-etcon foafcon someplace closer -- initial cost assesment ended up at 1000++ quid, enough for a nice new lens...

17:01:15 <mortenf> hi jbond

17:01:25 <jbond> g'day

17:02:06 <danbri_> amsterdam was fun...

17:02:21 <mortenf> indeed

17:02:29 <edd> we could have a foaf day around xml europe

17:02:36 <mortenf> and with a few more people, a little organizing etc.

17:02:40 <edd> which will likely be in AMS next year

17:02:41 <mortenf> where/when?

17:02:44 <mortenf> ah.

17:03:56 <mortenf> danbri, do you know the diff between = and == in squish?

17:04:24 <libby> I think they're the same mortenf

17:04:32 <mortenf> ah, even better, thanks libby

17:04:36 <danbri_> no. I've never liked the versions that have extra punctuation beyond rdf relations...

17:04:41 <danbri_> fork! :)

17:04:44 <mortenf> heh

17:05:22 <libby> I just use ~ and sometimes eq

17:05:30 <mortenf> is there an accepted syntax for optionals, i.e. in the select or where clause, punctuation etc.?

17:05:30 <jbond> Was looking for a good primer on rdql. Recommendations?

17:05:41 <libby> ~ is *incredibly* useful and I won;t have a word said against it

17:05:56 <mortenf> libby, do you use ~ for uris as well?

17:06:03 <mortenf> s/for/against/

17:06:03 <libby> mortenf, you could try damian's one

17:06:12 <mortenf> which is that?

17:06:14 <libby> shellac, where's your ruby optionals query stuff?

17:06:36 <shellac> um

17:06:44 <libby> yeah, I don't distunguish mortenf, e.g. for partial url matching

17:06:48 <danbri_> what does ~ mean though? looks like it should mean 'maybe'

17:07:06 <libby> it means case insensitve substring

17:07:06 <mortenf> hmm, i'm translating it into the mysql operator REGEXP

17:07:15 <mortenf> isn't that LIKE?

17:07:28 * libby was using them synonymously

17:07:33 <libby> that's bad probably

17:07:36 <mortenf> ah

17:07:39 <libby> can change....

17:07:57 <shellac>http://rdfweb.org/people/damian/esquish/esquish.tar.gz

17:08:08 <libby> jbond, not sure....is there not one on the jena site?

17:08:09 <mortenf> jbond, i don't think i remember seeing a "real" introduction to rdql or so, but a google on rdql and squish turns up some useful docs

17:08:13 <danbri_> Jbond, re rdql, I recall there being a big section in the jena tutorial...

17:08:14 <mortenf> ta

17:08:27 <danbri_>http://www.hpl.hp.com/semweb/doc/tutorial/RDQL/

17:08:35 <libby> I think shellac just used a ? after the optional select clauses. looked quite nice.

17:08:54 * libby wishes I had time to implemnat optionals, but not right now :(

17:08:58 * shellac must document esquish properly

17:09:01 <libby> yeah

17:09:18 <mortenf> e for extended or?

17:09:34 <libby> mortenf, should we collaborate on a more accurate squish bnf?

17:09:48 <libby> I'm just gearing up for a release of tinkling...

17:09:50 <mortenf> that'd be nice, i think dajobe might have some input as well

17:11:20 <mortenf> btw, i added an optional ORDER BY clause to my test impl. - i think that could be useful

17:11:40 <mortenf> i.e. "order by ?name asc, ?mbox desc"

17:12:12 <mortenf> although it might not be practial to implement for all storage backends (i get it for free with mysql)

17:12:18 <libby> sure

17:12:28 <danbri_> aren't the returned columns of mixed type, sometimes?

17:12:38 <libby> I think as long as a common core useful

17:12:41 <mortenf> could be, i guess

17:12:57 <mortenf> but only if there's an error on input :)

17:13:00 <shellac> mortenf: sorry - chattin' to chaals. yep e for extended

17:13:06 <mortenf> np, thx

17:14:03 <jbond> Sweet. Thanx.

17:14:44 <libby> have you implemented optionals in sql morten? danbri showed me some code from ericp that did that once

17:15:04 <mortenf> hacking on atm

17:15:43 <libby> neat :)

17:20:07 <mortenf> hmm, so in esquish a non-qualified triple matches only once (as compared to squish, where it matches all), while the qualified matches like regexp operators?

17:20:25 <shellac> erm

17:21:01 <mortenf> ah, found query.html, reading...

17:21:02 <shellac> oh - yes - + and * are a bit freaky

17:24:06 * danbri_ wonders about migrating this to #rdfig (slightly larger audience...)

17:25:02 <libby> be nice to do some testing too mortenf. I've been revisting the rdfq test cases...

17:25:05 <mortenf> i interpret plain squish ?p foaf:mbox ?m to return all (unique) triples that match, not just one. thus, the only sensible add-on is "?", to allow for NULL to be returned. is that right?

17:25:35 <mortenf> yeah, will have a look at those. is there a reference model they should be executed against?

17:26:25 <shellac> the use for + and * is that you don't have to juggle results afterwards

17:26:40 <mortenf> why would/should i do that?

17:27:30 <mortenf> isn't + just like without an operator, and * like ?

17:27:31 <libby> reference model?

17:27:47 <mortenf> a reference to be able to test impl.

17:27:59 <mortenf> (i realize there's yet no auto-testing)

17:28:07 <shellac> + and * give bind an array to a variable

17:29:00 <shellac> eg (foaf:mbox ?a mailto:pldms@mac.com) (foaf:knows ?a ?people)*

17:29:02 <mortenf> hmm, i don't think i understand

17:29:16 <shellac> sorry - better example:

17:29:28 <shellac> eg (foaf:mbox ?a ?mbox) (foaf:knows ?a ?people)*

17:30:11 <shellac> each result row contains a mailbox, and all the people that mailbox knows

17:30:20 <mortenf> ah, all in one row?

17:30:24 <shellac> yep

17:30:38 <mortenf> okay, that's the part i didn't get, thinking more...

17:31:06 <mortenf> hmm, that's not very sql-like ;)

17:31:16 <shellac> exactly :-)

17:31:29 <libby> not really a reference model....there are some manifest files with expected numbers of results rows and things

17:31:31 <shellac> but it's useful - one query can do alot of work

17:31:38 <mortenf> i was expecting an additional row for each extra known person

17:31:49 <mortenf> i.e. vertical instead of horisontal

17:32:07 <shellac> but in more complex cases it saves some work

17:32:10 <mortenf> libby, ok, but the input for the queries?

17:32:23 <mortenf> perhaps, at least on the wire

17:33:12 <mortenf> i'm not sure i like it, but i'll at least sleep on it :)

17:33:27 <libby> http://swordfish.rdfweb.org/rdfquery/tests/tests/squishtests2003-04-24/inputs/query-results-manifest.rdf - make sense?

17:33:53 <shellac> I'm not sure I like it, but it was worth playing with - cases where you have to sift cartesian product results

17:33:55 <libby> you can download them here: http://swordfish.rdfweb.org/rdfquery/tests/rdfquerytests2003-05-14.tar.gz

17:33:58 <mortenf> yep, looks managable

17:34:03 <mortenf> thx

17:34:13 <mortenf> oh indeed, worth exploring

17:34:48 <shellac> the optional sub-query case is more interesting, I think

17:34:59 <dajobe> FWIW, I've been using the jena rdql tests to get my query stuff working

17:35:22 <mortenf> yep, sub-queries are always interesting

17:35:45 <mortenf> but i think it can be done with just "?"

17:35:59 <shellac> the use case for that is nearestAirport

17:36:03 <mortenf> (and then keep the relational algebra in sync)

17:36:21 <shellac> try it with just an optional - things go bad :-(

17:37:50 <mortenf> like (spo): select ?n ?icao where ?p foaf:name ?n, ?p contact:nearestAirport ?a, ?a air:icao ?icao .

17:37:59 <mortenf> doh

17:38:20 <mortenf> like (spo): select ?n ?icao where ?p foaf:name ?n, ?p contact:nearestAirport ?a?, ?a air:icao ?icao .

17:38:38 <shellac> yep

17:38:39 <mortenf> so, if it doesn't match an ?a, it follows that there's also no ?icao

17:38:49 <shellac> nope

17:38:53 <mortenf> no?

17:39:01 <dajobe> this is more #rdfig stuff, hmm?

17:39:09 <mortenf> yeah, i guess

17:39:15 <shellac> it gets every air:icao available

17:39:37 <mortenf> hmmm

17:39:43 <shellac> unless you assume the triple order is significant

17:40:09 <shellac> and that's not nice for optimisation

17:40:12 <mortenf> no, because the ?a is null

17:41:10 <shellac> that's when order is significant

17:41:32 <mortenf> well, relative order, yes

17:41:46 <mortenf> hmm, not a total ordering is necessary, that is

17:41:54 <shellac> yep

17:42:50 <mortenf> i'm simply thinking of doing left joins for ?a...

17:43:30 <mortenf> so the order isn't really interesting

17:44:56 <ericP> libby, optionals, this has a sample query and the SQL that it compiles to --> http://dev.w3.org/cvsweb/~checkout~/perl/modules/W3C/Rdf/test/OrderTracking4-alg.sh?content-type=text/plain

17:45:16 <mortenf> thanks, will look

17:45:25 <ericP> so you're working on this too?

17:45:32 <ericP> maybe we could collaborate

17:45:38 <mortenf> yep, and yep :)

17:46:30 <mortenf> ah, yes, your way is exactly what i was thinking

17:47:33 <ericP> i've gotten started (barely) documenting the algorithm i'm using --> http://www.w3.org/2003/01/21-RDF-RDB-access/#ExampleOrdertrackingOuterJoin

17:47:41 <ericP>http://www.w3.org/2003/01/21-RDF-RDB-access/#ExampleOrdertrackingOuterJoin

17:47:48 <ericP> oops, no bot

17:47:55 <dajobe> #rdfig please :)

17:48:05 <mortenf> ok, everybody move NOW.

17:48:23 <shellac> don't give in!

17:48:24 <libby> ooh ta ericp

17:49:54 <ericP> busted!

18:21:02 <nmg_lunching> nmg_lunching is now known as nmg

18:31:26 * danbri finds http://www.happyfeettravels.org/Stencils/MissionDist/ (graffiti / stencil pics)

18:31:41 <danbri> from a search on 'monkey knife fight' at google images, i think it found a stencil i saw

19:14:18 <golbeck> golbeck is now known as golbeckSanibel

20:19:03 <mortenf> a stencil i found: http://www.wasab.dk/morten/2003/10/photos/england/4/image-90.html - i was going to dare you to refind it, but then i realised the metadata has geo-properties...

20:21:01 <sbp> tea in the next photo courtesy of Ceylon

20:21:11 <mortenf> :)

20:21:33 <mortenf> and it only took me three tries to pronounce it right

20:21:39 <sbp> hehe

20:22:27 <sbp> I want to download the whole set one day

20:22:39 <sbp> so if your bandwidth gets maxed out, feel free to blame me

20:22:56 <sbp> though at my rates it'll take a few months...

20:23:13 <mortenf> i can pack you up a nice tgz?

20:23:26 <sbp> ooh, that'd be helpful

20:23:37 <mortenf> all from /england/?

20:23:42 <sbp> yeah, please

20:24:34 * mortenf compresses

20:26:11 <Boki_> Boki_ is now known as Boki

20:29:46 <sbp> wget reckons 38 hours. not bad!

20:29:51 <mortenf> heh

20:30:03 <mortenf> throttled?

20:30:31 <sbp> nope

20:30:31 * JibberJim advertises that url to the world and encourages everyone to download!

20:30:35 <mortenf> :)

20:30:50 <sbp> heh

20:31:05 <sbp> you could get the same effect with while 1: get(uri) on a smaller resource, though

20:31:20 <mortenf> ah, yeah

20:31:32 <sbp> anyway, I've gotta run, but I can leave this in the bg. thanks again, morten!

20:31:38 <mortenf> sure

20:36:27 <JibberJim> They're quite soft your images mortenf, I always feel an urge to sharpen them

20:36:45 <mortenf> yeah, i think it's a canon thing

20:36:56 <mortenf> or simply out-of-focus :)

20:37:43 <JibberJim> Of course it's easy enough to sharpen-them-up in photoshop

20:41:00 <JibberJim> In fact googling around now, seems to suggest they're deliberately leaving it soft for editing

20:41:04 <mortenf> i had a few prints made the other day (for a class!), they look pretty good on paper, .e.g. http://www.wasab.dk/morten/2003/08/photos/1/image-15.html and http://www.wasab.dk/morten/2003/09/photos/3/image-32.html

20:41:09 <mortenf> yep, it's a feature

20:42:03 <mortenf> but a somewhat controversial one

20:45:14 <JibberJim> Of course I'm being overly picky as you tend to become when they're really, really good.

20:45:44 <mortenf> thanks :) and no problem, pick away

20:50:14 <JibberJim> http://www.barrypearson.co.uk/photography/portfolios/ecuador.htm - compare INFO to the 700 images, that's the next skill I need to learn, how to process the images in photoshop

20:50:31 <JibberJim> http://www.barrypearson.co.uk/photography/pictures/eg95/eg95_22_23_3.htm <- That's just stunning.

20:51:03 <mortenf> wauw, yeah...

20:53:14 <mortenf> but, it's cheating...

20:53:43 <JibberJim> Hmm, I dunno, "it's correcting the deficiencies in the camera" maybe?

20:53:58 <mortenf> yeah, that's a pov :)

20:54:03 <JibberJim> Especially with the dSLR's deliberately keeping it soft.

20:54:14 <mortenf> true

20:54:38 <JibberJim> Although his are 35mm (he want's an ist-D same as me!)

20:54:50 <mortenf> heh

20:57:15 <mortenf> from H: in #rdfig: if (!foaflist.contains(str) && !str.endsWith("scutterplan.rdf")) // scudderplan.rdf has broken etag stuff

20:57:24 <mortenf> is that your plan?

20:58:07 <JibberJim> I disabled all my scutterplan.rdf's

20:58:26 <mortenf> ah, ok by "broken" he means "out-of-date"?

20:59:59 <JibberJim> I was finding very dodgy things with ecademy, it seemed ...foo&uid=10 rdf:seeAlso'd ...foo&&uid=10

21:00:36 <mortenf> ouch, do they still do that?

21:00:58 <mortenf> hmm, seems sbp got cut off after 5M :(

21:01:00 <JibberJim> I don't know I stopped scuttering because there were lots of them

21:01:03 <JibberJim> wget -c !

21:01:10 <mortenf> i hope so!

21:01:18 <mortenf> perhaps i should just send him a cd...

21:01:29 <JibberJim> I didn't check where they stopped working

21:43:24 <oPless> for u phone freaks out there

21:43:26 <oPless>http://tinyurl.com/rj0l

21:45:54 <mortenf> oh boy, the resulting references 116 images - that beats marcc's page!

22:12:14 <sbp> 5M: yeah, I had to drop the line

22:12:20 <sbp> hence wget -c :-)

22:12:37 <mortenf> :)

22:13:11 <mortenf> but seriously, i could surface it to you, perhaps even quicker?

22:13:48 <sbp> heh. faster post bandwidth

22:14:02 <mortenf> yep, isn't that what say of ups vs internet2?

22:14:10 <sbp> yeah

22:14:16 * sbp recalls

22:14:33 <sbp> well, it's up to you. insignificant cost?

22:14:40 <sbp> I could repay in CDRs

22:14:42 <mortenf> yep

22:14:58 <mortenf> there'll come a time when i need something the other direction i'm sure

22:15:10 <mortenf> perhaps a python toolkit ;)

22:15:17 <sbp> heh, heh

22:15:36 <sbp> fair enough. arrange by email?

22:15:41 <mortenf> yep

22:15:51 <sbp> okay. ta!

22:16:15 <mortenf> sure

22:16:15 * sbp ctrl+c's wget

22:16:15 <mortenf> heh, i even have a nice safe box

22:49:15 <mortenf> jim, blub.foafnaut.org seems broken (links don't work anymore)?

22:51:45 <edd> evening

22:51:57 <edd> it's cynical'o'clock, i see

22:53:55 * zool checks her watch, i think you'll find it's quarter-to-maudlin

22:54:29 <edd> does that mean i've missed cynical? damn!

22:55:44 <zool> step this way into my time machine, dear edd

22:56:00 * edd puts his left left leg in...

22:59:03 <LTjake> Eh.

22:59:31 <edd> A great introduction. You must be Canadian

22:59:39 <LTjake> I am, actually.

22:59:49 <edd> Greetings. Most of us aren't.

22:59:57 <LTjake> damn you all!

23:00:00 <LTjake> just kiddin.

23:05:52 <LTjake> JibberJim... around?

23:26:06 <edd> well, he should be

23:26:10 <edd> it's not yet 2am here

23:27:08 <LTjake> yeah...