Friend of a Friend (FOAF) IRC Chat Logs for 2004-02-25

00:12:53 <danbri> thanks; replied.

00:47:19 <ChrisDodo> http://undergroundlondon.com/antimega/archives/000991.html # please don't take the tech too seriously/rip it apart :)

00:50:59 <sbp> wow, that's pretty cool

00:51:10 <sbp> I was thinking about a similar thing recently, but still developing it...

00:51:30 <ChrisDodo> it's kind of high on the freakazoid scale, but in my mind it's all doable

01:12:29 <IsoosI> hmm

01:12:36 <IsoosI> if you could compress XML somehow effectively

01:12:54 <IsoosI> you you fit a lot into the barcode like that

01:15:55 <xavier> yup.

01:16:05 <xavier> i'm posting a comment on that point right now.

01:19:28 <IsoosI> XML tells you a lot about the data

01:19:33 <IsoosI> which makes compressing really easy

01:20:05 * xavier nods

01:20:29 <xavier> the thought of gzipped rdf encoded as 2D barcodes makes me very cheerful :-)

01:21:47 <kasei> wow... just finished reading. that's /very/ cool.

01:22:29 <xavier> indeed.

01:22:34 <kasei> having support for that in a camera would be fantastic. especially in cases where gps wouldn't work anyway (indoors, underground, ...)

01:22:37 <xavier> Anyone up for hacking out some barcode libraries?

01:23:51 <IsoosI> xavier: hmm

01:24:02 <IsoosI> I'm not sure gzip would quite give you that much compression if it's small

01:24:43 * IsoosI imagines the uni with RDF barcodes on lecturers doors, who they are, what they're interested in, what lectures they take, what their office hours are...

01:25:02 <xavier> gzipping a 16k foaf file brought it to 8k.

01:25:09 <IsoosI> true

01:25:20 <IsoosI> but when he was halving the size of the barcode by not having his email

01:25:36 <IsoosI> it assumes he either has a stupidly large email address, or a tiny foaf file

01:25:48 <xavier> (i think he just had a lot of email addresses)

01:26:42 <IsoosI> ahh

01:26:45 <IsoosI> that could explain it

01:27:03 <xavier> gzipping a 4k FOAF file left it about the same.

01:27:21 <xavier> bz2 as well.

01:27:22 <xavier> hm.

01:27:22 <IsoosI> have you seen Xerox's encoding of information in an image using vertical or diaginal pixels?

01:27:37 <IsoosI> embedded metadata! :)

01:27:38 <xavier> don't believe so..

01:28:26 <xavier> oh, you mean using the low-order bits of the pixels?

01:28:45 <kasei> if space is at a premium, perhaps some serialization besides xml would be better?

01:29:16 <xavier> kasei: that was my comment. RDF can be encoded in other ways, optomized for the medium.

01:29:45 <xavier> hm. i'll play with it on the train ride home, methinks :-)

01:29:46 <kasei> ah, sorry... missed that :)

01:29:55 <xavier> kasei: not your fault. i meant on the blog post.

01:30:53 <xavier> the trickiest parts are not the data, but the format, though.

01:30:55 <IsoosI>http://www.parc.com/research/asd/projects/dataglyphs/

01:32:26 <xavier> i was playing with a tool that understood 2d barcodes (on US licenses and such) and it took forever to get an image that it accepted. The hard part is getting a clear enough photo - a physical characteristic, not a software one.

01:32:52 <xavier> of course, if the japanese cell phones can do it, so can we :-)

01:35:31 <Freso> Well... they *are* Japanese...

01:35:35 <Freso> ;P

01:45:13 <IsoosI> hmm

01:45:18 <IsoosI> how much information can you fit in an RF tag?

01:53:35 <kasei> googling for rf tag capacity yields "In most cases the amount of data storage on a passive tag is fairly limited - capacity often being measured in bits as opposed to bytes"

01:53:59 <kasei> plus a serial number

01:59:34 <crschmidt> whoa, talkiness.

02:00:30 <crschmidt> oh, it was all a while ago.

02:00:45 <kasei> heh

02:03:30 <IsoosI> :)

02:04:58 <Freso> Quick link to sha1 sum generator?

02:05:21 <crschmidt> Digest:SHA1! ;)

02:05:42 <Freso> Uhm. Assuming that is Perl, I do not have Perl installed ;)

02:05:52 <Freso> There's probably a Python function to do it though....

02:06:00 * Freso wanders off to Python documentation

02:06:45 <Freso> Whoa.

02:06:59 <Freso> Python 2.3.3 release is taken away from the news box at the index page.

02:07:54 <IsoosI> import sha

02:08:00 <IsoosI> sha.sha("FOAF").hexdigest()

02:08:05 <sbp> away from the news box: eek!

02:08:11 <sbp> sha.new, isn't it?

02:08:27 <Freso> For md5 it's md5.net.hex

02:08:30 <IsoosI> well, it depends on how quick you want the hack to be :)

02:08:37 <Freso> And sha is standard mod?

02:09:26 <IsoosI> I believe it's a standard sha1 hash

02:09:49 <sbp> odd. sha.sha isn't documented...

02:09:52 * sbp reads the source

02:09:56 <Freso> Well, yes, but I was asking if 'sha' is in the standard Python distribution... :p

02:10:07 <IsoosI> I don't recall installing it

02:10:10 <Freso> built-in mod, yes.

02:10:19 <Freso>http://www.python.org/doc/current/lib/module-sha.html

02:10:44 <Freso> "It is used in the same way as the md5 module: use new() to create an sha object"

02:11:24 <IsoosI> you can use new to create an object

02:11:37 <IsoosI> or use sha() to create an object and preseed it with the string argument

02:11:41 <IsoosI> saving a step

02:12:00 <sbp> you can preseed with new too

02:12:31 <IsoosI> kinda makes it redundant having two functions then :)

02:12:48 <IsoosI> someone musta been on drugs at the time

02:12:51 <sbp> yeah. they don't compare equal either... hmm

02:13:41 <IsoosI> it's presumably so yo ucan do "from sha import *" and "from md5 import *"

02:13:42 <Freso> ...

02:13:45 <IsoosI> and not worry that "new" is overloaded

02:13:47 <Freso> strange.

02:13:54 <IsoosI> since you can use "sha" or "md5"

02:14:06 <Freso> The result of the encryption is not the same as what it has already been encrypted as...

02:14:31 <sbp> IsoosI: that makes sense

02:14:49 <sbp> except "from module import *" is evil and should be discouraged

02:16:29 <Freso> Ah!

02:17:07 <Freso> Should mailto: be included in the encrypted string?

02:17:13 <IsoosI> yes

02:17:57 <sbp> if you're using it in foaf:mbox_sha1sum...

02:19:06 <Freso> I am :p

02:20:05 <IsoosI> someone on #foaf asking about sha1sum's and /not/ using it for mbox_sha1sum? <grin>

02:20:11 <sbp> hehe

02:20:36 <sbp> stranger things have happ... well, I guess not :-)

02:21:13 * Freso loves his Cygwin/Python combo

02:21:33 <IsoosI> I 3> python as well

02:21:50 <Freso> =)

02:22:17 * Freso drags IsoosI offchan

02:23:37 <sbp> sbp 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 # there's only one social network (and it's made of agents, not computers) # The FOAF World... It's where Strange Things happen

02:26:40 * crschmidt plays with mortenf's foaf explorer.

02:27:32 <IsoosI> are there any projects that take a .jpg and extract the Exif data into RDF?

02:29:38 <crschmidt> Hm.

02:29:41 <crschmidt> I don't know how to use foafnaut.

02:50:40 <Stevos> Stevos is now known as Stevos|MadAtZos

02:52:04 <Stevos|MadAtZos> Stevos|MadAtZos is now known as Stevos

03:29:30 <golbeck> golbeck is now known as golbekcCA

03:29:33 <golbekcCA> golbekcCA is now known as golbeckCA

07:54:48 * jdougan is away: Later, eh.

08:18:13 <BouncyKevin> BouncyKevin is now known as BOingKevin

08:18:13 <BOingKevin> BOingKevin is now known as BouncyKevin

08:27:32 <BouncyKevin> BouncyKevin is now known as KevinMarks

08:28:10 * crschmidt does the edd dance.

08:30:55 <crschmidt> Anyone in here use Dashboard/Gotten it to work?

08:36:14 <KevinMarks> KevinMarks is now known as Tigger

08:37:29 <esigler> New LJ bot policy: http://www.livejournal.com/community/lj_dev/621620.html

08:48:16 <Jibbler> seems reasonable

08:49:01 <crschmidt> I've been asking for one of those for a year and a half now ;)

08:49:17 <verbosus> verbosus is now known as verbosus-via

08:49:39 <Jibbler> seems quite positive towards bots, really

08:50:55 * crschmidt nods.

08:50:59 <crschmidt> He's really interested in FOAF.

08:51:12 <crschmidt> I kind of wish he didn't seem to have a strong dislike for me ;)

08:52:53 <Jibbler> lol

08:54:19 <crschmidt> It's what I get for being a propononet of good development methods :P

08:55:31 * Jibbler goes to uni to skip past the picket lines again. bye!

08:55:44 <crschmidt> have fun :)

08:57:24 * crschmidt posts link in ljfoaf community to ljdev entry.

08:57:48 * crschmidt goes to bed.

08:58:01 <crschmidt> more cathedral+bazaar reading to do, whee.

08:58:03 <crschmidt> See ya'll tomorrow.

08:58:15 <esigler> night

09:00:54 * crschmidt !! Thank you, again, for support in ljfoaf :)

09:23:56 <verbosus-via> verbosus-via is now known as verbosus

09:24:30 <libby> <IsoosI> are there any projects that take a .jpg and extract the Exif data into RDF?

09:24:38 <libby> yep, at least 3 or 4

09:25:04 <libby> there were some efforts to try and combine the schemas used, which should be ongoing at some point

09:25:58 <Remosi> ahh cool

09:26:01 <libby> see http://lists.w3.org/Archives/Public/www-rdf-interest/2003Aug/0021.html

09:26:03 <Remosi> <- IsoosI

09:26:07 <libby> but work has happened since then

09:26:16 <libby> heh, hello

09:26:30 <ldodds> morning

09:26:41 <Remosi> I usually go by the nick "Isomer", however someone else on this network has registered that nick.

09:26:41 <libby> heya leigh

09:26:52 <libby> ah

09:26:56 <libby> more info here: http://esw.w3.org/topic/ImageDescription

09:27:19 <libby> and here http://esw.w3.org/topic/ImageDescriptionList (though no lists as yet)

09:29:16 <Remosi> heh

09:31:11 <kota> I found the exif-rdf schema here http://www.w3.org/2003/12/exif/ , is the namespace gonna be used for exif data description?

09:31:30 <kota> (i mean, we're not gonna use kanzaki.com's namespace)

09:35:07 <libby> I don;t know kota, maybe ask danbri later? I think there will be one on the w3c site, although I don;t know if it is better to create a new one or just confusing

09:35:42 <libby> oh interesting

09:35:51 <kota> hmm, ok, so still not sure.

09:35:57 * ldodds waves morning to Jim

09:35:57 <libby> yes maybe that wil be it then, I didn;t realize masahide had been doing that

09:37:03 <libby> kota, looks like it is yep: http://www.ilrt.bris.ac.uk/discovery/chatlogs/rdfig/2004-02-18.html#T03-36-06

09:37:13 <libby> karl and masahide have been collaborating

09:37:20 <kota> yeah, i saw his memo that he didn't know what's goin on (http://www.kanzaki.com/memo/2004/02/19-1

09:38:01 <libby> ah is that what that says...

09:39:12 <kota> yeah, anyway, hope they're goona do a good work. they're so cool :)

09:39:40 <libby> oh I'm sure they will :)

09:39:53 <kota> :)

09:42:15 <libby> they're multiplying!

09:43:27 <JibberJim3> yeah, bit of an unstable connection...

09:43:53 <JibberJim3> but the original JibberJim one is okay...

09:44:23 <JibberJim3> Hey Kirsch...

09:45:11 <JibberJim3> oops.

09:49:45 <JibberJim3> 02:29:41 <crschmidt> I don't know how to use foafnaut.

09:50:13 <JibberJim3> If you're about, you click on blubs and watch sproings appear - HTH

10:23:37 <Tigger> Tigger is now known as KevinMarks

12:20:13 <ldodds> hi danny

12:20:25 <JibberJim3> hey all

12:34:04 <swh> swh is now known as swh_lunch

12:37:53 <danja_> hi folks - forgot this thing was on

12:38:35 <Remosi> :)

12:39:08 * ldodds is looking for evidence that ISBNs get reused, or whether they can be safely treated as an IFP

12:39:43 <danbri> I heard they're re-used, but I can't back that up with evidence.

12:39:50 <ldodds> nor me

12:40:18 <ldodds> the only quote I could find was something on isbn.nu that said some people may have reused ISBNs from out of print books

12:40:22 <danbri> BTW there is a reading of the OWL specs that'd permit a bib:isbn property to still be called an IFP

12:40:32 <ldodds> the ISBN manual clearly states that they shouldn't be reused

12:40:45 <dajobe> in http://www.readerware.com/help/rwISBN.html

12:40:47 <danbri> ...so long as it was never simultaneously true (in the same doc/graph) that some ISBN was the isbn of two distinct things.

12:41:28 * danbri wonders if chris' plogs.net LJ installation will acquire foaf powers sometime

12:41:34 <dajobe> here's a specific example with publisher and books http://www.ioba.org/newsletter/V11/IOBANL-ISBNlookup-5-03.php

12:41:35 <ldodds> "a publisher could reuse ISBN's from out of print titles, or simply make a mistake"

12:41:53 <danbri> the former is the scary bit

12:41:57 <dajobe> "often the publishers themselves lost track of what ISBNs they had already used."

12:42:01 <danbri> what sense of 'could'

12:42:20 <danbri> how about: we think, what's most likely, ISBN practice gets cleaned up, vs replaced.

12:42:30 <danbri> I suggest cleanup most likely, and that we treat it as IFP

12:42:45 <danbri> but have no sense how bad the problem is

12:43:27 <Remosi> apparently large bookstores use their own numbering schemes as they frequently find collisions

12:43:36 <Remosi> although that may include ISBN <-> non-book-barcodes.

12:43:52 <Remosi> and I have no reference for that.

12:45:06 * ldodds is inclined to state it's an IFP

12:45:18 <ldodds> backed up by spec, if not by common usage.

12:45:21 <dajobe> it'll get you the book you want if you are lucky

12:45:22 <danbri> yeah, it might help find some errors in data

12:45:56 <ldodds> I want to be able to smush on _something_ for book data, and that's the best there is

12:46:33 <ldodds> Amazon homepage might do, but that's an ASIN based link which is really just ISBN; plus there are multiple homepages (.com, .co.uk, etc)

12:48:31 <ldodds> the ISBN URN registration seems to assume they're unique too

12:49:02 <danbri> I have my barcode scanner still to use on my bookshelves...

12:49:36 * ldodds is going to do his by hand

12:49:51 <danbri> you're welcome to borrow my scanner

12:50:10 <ldodds> I might whip up a spreadsheet to capture stuff, then write some code to process it and generate RDF

12:50:51 <ldodds> danbri: ta, I'll think about it, but thought it might be an excuse to write another fun tool :)

12:51:28 <danbri> typing in book codes isn't my idea of fun, but whatever does it for you...

12:51:49 * ldodds grins

12:52:05 <danbri> I like this rdf-as-barcodes thing, http://undergroundlondon.com/antimega/archives/000991.html

12:52:08 * xavier hands danbri a CueCat

12:52:15 <ldodds> yes, thats cool

12:52:21 <ldodds> then I can have it on a T-shirt :)

12:52:30 <danbri> I bought some genreic scanner thing from maplins

12:53:20 * ldodds goes to look at scanners on maplins

12:53:51 <ldodds> hmm, maplins.co.uk isn't the right site, but hey, a sword umbrella might be cool :)

12:56:13 <ldodds> hand typing is cheaper :)

13:00:01 <JibberJim3> but worse for RSI

13:44:09 <swh_lunch> swh_lunch is now known as swh

13:46:48 <CaptSolo> ldodds: are you making a reading-list-in-RDF application?

13:47:02 <CaptSolo> what's a CueCat?

13:47:21 * CaptSolo also has a profile on allconsuming.net

13:47:30 <CaptSolo> though i have only one book on it :>

13:50:44 <JibberJim3> A cuecat is a breed of pet cats, that has a large stick poking out of its head, much like a unicorn, but looking a little more like a snooker cue.

14:01:38 <JibberJim3> hey libby

14:03:20 <libby> heya jibberjim3

14:03:58 <JibberJim3> Have you BRS lot looked at your calendars for a foafmeet?

14:04:14 <libby> hm

14:04:16 <JibberJim3> and anyone free for an SVG meet on March 9th - I'll be talking about RDF and SVG...

14:04:17 <libby> not really

14:04:35 <libby> re foafmeet. danbri's the tricky one - he's away a lot next few months

14:04:37 <dajobe> 10th bad for ILRTers

14:04:58 <libby> is 9th tuesday?

14:05:04 <dajobe> yes

14:05:11 <dajobe> I'm away then

14:06:05 <libby> I migth make it then jibs, see hoiw much cacsh I have

14:06:12 <libby> this is in the evning right?

14:06:34 <JibberJim3> yeah evening.

14:06:39 * JibberJim3 will buy beer.

14:06:44 <libby> :)

14:07:03 <libby> wn:beer is my foaf:friend

14:07:21 * JibberJim3 actually has already offered other people dinner for attending, so I'm possibly bid uppable to that :=)

14:07:29 <libby> where is the meet?

14:07:30 <libby> heh

14:07:41 <JibberJim3> Central London

14:07:43 <libby> I'll have to go back same evening, but that should be ok

14:09:05 <JibberJim3>http://lists.w3.org/Archives/Public/www-svg/2004Feb/0064.html

14:09:59 <libby> neat. that's very central, cool

14:14:42 <verbosus_> verbosus_ is now known as verbosus

14:27:45 * ldodds catches up

14:28:01 <ldodds> CaptSolo: http://www.ldodds.com/blog/archives/000126.html, vocab linked from comments

14:30:18 <CaptSolo> ldodds: thanks - the idea is good

14:30:30 <CaptSolo> ldodds: about ISSB numbers

14:31:07 <CaptSolo> i had a thought one can make an electronic listing of his home library by scanning in the book barcodes

14:32:16 <ldodds> yep, that's what danbri was talking about

14:32:21 * CaptSolo has a barcode scanner, but it won't work so good for me - most of my library is in latvian or russian and they do not have barcodes on them.

14:33:15 <CaptSolo> need to get more new books :)

14:50:23 * Jibbler adds lots of new postcodes to http://www.jibble.org/ukpostcodes/

15:13:28 <masaka> exit

15:26:29 <crschmidt> Hey, I know most of you probably know RDF-Dev mailing list protocol better than me - jbond just sent me a list of possible suggestions to the LiveJournal FOAF file. Better to respond to him, or to the list?

15:28:08 <danbri> list is good

15:28:11 <danbri> rdfweb-dev

15:28:25 <danbri> (rdf-dev was another list years ago, precursor to www-rdf-interest, the RDF Interest Group list)

15:28:31 * crschmidt nods, yah, that ;)

15:28:33 <crschmidt> Okay, thanks.

15:31:08 <JibberJim3> the rdfweb-dev protocol is you get kicked off every month, and then can't get mailman reg strings which work to let you back on :-(

15:32:34 <danbri> I got some way towards upgrading mailman, jim. new version compiled, some work on migration needed, the install instructions make it sound suspiciously easy (but also fiddly and confusing)

15:59:36 <chrisMobile> rdf resource can be a url, right? Jbond suggested we add errorReportsTo, and that user an rdf resource, so we could link to

15:59:46 <chrisMobile> Our web support area?

16:00:35 <JibberJim3> that would be good, or an email address to.

16:01:39 <chrisMobile> Email address on lj for support is mostly not used, it opens a request, but then it just goes to the web support board anyway.

16:02:54 <JibberJim3> yep, whatever's most appropriate. We need a robot scutter vocab, 'cos you could maybe have an email address that accepts reports from scutters about invalid data, aswell as a human errorReportsTo...

16:04:06 <DaniB> someone knows where can I find help with Jena API???

16:04:18 <dajobe> DaniB: try #rdfig

16:04:19 <danbri> the jena-dev list is quite supportive

16:04:28 <danbri> also yes #rdfig for IRC discussion

16:04:33 <DaniB> ok.. thanks dajobe

16:04:37 <chrisMobile> I'll leave that up to you guys to build, i'm just using what you give me. :)

16:04:42 <DaniB> thanks danbri!!!!

16:04:44 <dajobe> I don't see any jena develoeprs there right now though.

16:04:57 <ldodds> DaniBL I've used Jena a reasonable amount, although not an expert

16:05:21 <JibberJim3> Any requests for foafnaut hacking, Ihave 6 hours (+train delays) put aside this weekend on it.

16:05:29 <ldodds> conferences?

16:05:31 <ldodds> :)

16:05:49 <JibberJim3> yes, yours is in, but I reckon that should be quicker that.

16:05:56 <DaniB> I just want to know how to get the root concepts of an OWL ontology!!!

16:06:11 <ldodds> ah, that's the bit I'm not an expert on :)

16:06:27 <chrisMobile> Jim, i couldn't really figure out how to get it to work... Followed the link from foaf explorer, and nothing came up.

16:06:52 <chrisMobile> Have you tried an lj foaf file at all?

16:08:02 <chrisMobile> (or is there some part of it that i'm not doing right? :))

16:08:11 <ldodds> Jim: what will the extent of the changes be for conferences? Just adding the attendee relationship, or altering the blubs too? I wasn't 100% sure where to hack things to look up an alternate blub.

16:09:13 <JibberJim3> Looking up an Alternate bug is already in.

16:09:19 <JibberJim3> s/bug/blub/

16:09:40 <JibberJim3> the data for a blub just needs to contain the relevant type.

16:09:49 <ldodds> cool

16:10:09 <JibberJim3> ah, the problem there ChrisMobile is that foafnaut is based on old data, so that was a person not in foafnauts db.

16:10:17 <ldodds> I'm going to do a "books peple are reading" vis. too

16:10:19 <JibberJim3> and you need the blub design in the XML file.

16:11:04 <chrisMobile> Jim, is there a way to expand that information? Is it your own scutter bot?

16:11:14 <JibberJim3> Yes, it's my own scutter.

16:12:24 <chrisMobile> Jim, alright. Does it do any scuttering based on the incoming nodes? Can i go to a url and say "spider me next"?

16:12:51 <JibberJim3> In theory yes you can go to http://jibbering.com/foaf/update.html

16:16:42 <JibberJim3> but in practice you can beg me, as there's too much foaf in the world for my box.

16:16:42 <chrisMobile> Jim, it says to pester you. ;)

16:16:50 <chrisMobile> Heh.

16:17:52 <chrisMobile> I need some kind of nifty, pretty, easy to use app to show people in general.

16:18:38 <chrisMobile> Foaf explorer is kind of neat, but not pretty enough. I don't know if foaf walk works on our data.

16:19:15 <chrisMobile> I can never get it to work on my machine.

16:21:11 <JibberJim3> you could run your own foafnaut pretty easily.

16:21:36 <ldodds> yep, generating the cache files isn't that hard

16:21:47 * ldodds will write up what he did

16:21:48 <DaniB> chrisMobile, try http://www.semaview.com/foafwalk/viewer.html?q=http://www.inf.puc-rio.br/~dani/foaf/foaf.rdf

16:21:55 <chrisMobile> Interesting, maybe i'll look into that.

16:22:08 <chrisMobile> Dani, i'm on my cell phone.

16:22:15 <DaniB> hmmmm

16:22:25 <chrisMobile> No graphical browser, only my ssh session. ;)

16:22:51 <chrisMobile> Crschmidt can hold the url for me, if you msg it to him.

16:22:53 <DaniB> ok!

16:23:13 * chrisMobile is crschmidt, so everyone knows.

16:27:43 <chrisMobile> vcard uses locality, region, country. In the states, could that map to city, state, country?

16:30:02 <kasei> as long as you don't mind 'region' varying in size by a great amount

16:30:39 <chrisMobile> I don't mind, i just wonder if tools that use the info mind.

16:30:58 <kasei> ah, dunno

16:31:27 <chrisMobile> People want location information, and i want to provide it.

16:31:39 <kasei> a quick check shows that apple AddressBook doesn't use any of them

16:31:45 <chrisMobile> But i'm not sure what namespace, if any, best applies.

16:32:12 <JibberJim3> the geo one?

16:32:12 <CaptSolo> jibber: how may one know if he's in the scutter-list or not?

16:32:26 * CaptSolo assumes he's not and needs to bother JibberJim

16:32:42 * kasei &

16:32:43 <chrisMobile> We can't calc geo information on the fly, unless there's a really light weight way to do it.

16:33:07 <JibberJim3> hmm, there was a page you could go to ask the foafnaut db what it knew about a url.

16:33:28 <chrisMobile> Now we do all our post code to location mapping when the information is entered.

16:35:25 * chrisMobile sighs at livejournal's lack of realname.

16:35:44 <chrisMobile> We provide a field for it, but no one user it for that.

16:35:56 <chrisMobile> Uses.

16:36:11 <chrisMobile> And it's what everyone wants. :p

16:46:13 <chrisMobile> Can two fields together identify an ifp, or does it have to be just one?

16:47:45 <danbri> they can in practice, but it is beyond the expressivity of RDFS/OWL to tell that to machines

16:48:03 <chrisMobile> Okay.

16:48:23 <chrisMobile> I'm trying to hack up an rdf spec for class information.

16:48:52 <chrisMobile> And having trouble defining an ifp that would work more than where i am.

16:49:24 <chrisMobile> I can use call numbers at uiuc, but i don't know if most schools have them, etc.

16:49:34 * chrisMobile needs to do more thinking.

17:11:52 <DaniB> DaniB is now known as DaniBR

17:12:20 <JibberJim3> DaniBR a dyslexic danbri?

17:12:43 <chrisMobile> It appears so!

17:13:26 <chrisMobile> Jim, what's the backend of foafnaut?

17:14:26 <danbri> foafnaut hacking req: a distribution that'd encourage others to bolt it onto their DBs and crawlers

17:14:45 <ldodds> IIRC there's a scutter/smushing/database component which actually gather up the data, then there's the front end. The glue inbetween is a series of "cache" files, with a relatively fixed format.

17:14:58 <ldodds> The cache files are generated by the database, but could be generated by A.N. Other system

17:15:09 <ldodds> I've written something to take a model in Jena and spit out cache files for foafnaut

17:15:23 <ldodds> Aiming to publish a demo Real Soon Now

17:15:33 <ldodds> will also release my cache generator

17:17:26 * chrisMobile will look into building a foafnaut type tool for lj.

17:18:24 <ldodds> it's really simple to generate the required files

17:19:04 <chrisMobile> Crawling our data from outside is hard, but building stuff and working from it might work... We already run a cached 6 degrees

17:19:06 <chrisMobile> Tool.

17:19:29 <chrisMobile> So might be able to work from that.

17:21:15 <JibberJim3> the cache files are an option, they don't need to be cstatic files.

17:21:48 <JibberJim3> adn if you have 1 way only foafnaut (no isKnownBy) then it's actually just a translation of the current foaf files for LJ

17:22:22 <JibberJim3> an XSLT could probably do the job

17:22:24 <chrisMobile> There's an is known by field in foaf?

17:22:45 <ldodds> no it's a foafnaut thing

17:22:45 <JibberJim3> just in foafNnaut, it's in the jim Namepsace

17:22:51 <chrisMobile> Oh, okay.

17:23:05 <chrisMobile> People were asking me to include that data.

17:23:25 <chrisMobile> Which is why i asked. :)

17:27:52 <chrisMobile> Can't foafnaut do things like semaview, with the expanding the graph, or is that my imagination?

17:28:50 * danbri doesn't think isknownby needed

17:29:16 * JibberJim3 is willing for patches from danbri

17:29:30 <JibberJim3> what do you mean by expanding graph?

17:30:27 <chrisMobile> I don't know, i thought it drew a graph of knows, but i think i'm getting it confused with semaview.

17:31:16 <JibberJim3> libby's path generateor could produce a link with a list of people which would all get explodied showing the links.

17:31:26 <JibberJim3> it can also highlight a path.

17:31:58 <JibberJim3> (maybe, taht might be disabled in the current UI)

18:00:45 <DaniBR> Oh... are you talking about my nick?! ;)

18:00:58 <DaniBR> Dani = Daniela and BR = Brazil

18:01:21 <JibberJim3> ahah.

18:02:01 <crschmidt> Apparently semaview doesn't work with LiveJournal foaf data.

18:02:08 <crschmidt> probably because we don't have foaf:name support.

18:02:16 <crschmidt> anyonw know who i'd contact about that?

18:04:28 <xavier> why not just put in their "real names"?

18:05:27 <crschmidt> Because no one uses it for that.

18:05:44 <crschmidt>http://www.livejournal.com/support/highscores.bml

18:05:45 <xavier> so?

18:05:53 <crschmidt> So I don't want to provide incorrect information :P

18:06:13 <crschmidt> "pleased to meet you...hope you can guess my name" isn't much of a "foaf:name" ;)

18:06:14 <xavier> it's correct in its own domain

18:06:37 * crschmidt shrugs. That's arguable.

18:06:48 <crschmidt> For a long time, changing your "name" was the only way to change the <title> of your journal.

18:06:58 <crschmidt> Now, it's still used like that, even though it's not needed.

18:07:20 <crschmidt> I doubt Jay wants the web to know his name as being "Vice Admiral Asshat" ;)

18:07:46 * xavier chuckles

18:10:21 <crschmidt> hm. that doesn't seem to be what's causing the problem either.

18:10:27 <crschmidt> meh. lunchtime. i'll poke at it when I get back.

18:13:43 <crschmidt> man, i'm going to need to install most of GNOME to get Dashboard to work :P

18:15:12 <xavier> that's why i've avoided it..

18:15:30 * crschmidt shrugs. I want to have screen shots for all those people who say FOAF is useless ;)

18:15:58 <crschmidt> "FOAF isn't useless - look at how friggin cool this thing that edd wrote is!"

18:16:09 <xavier> yeah.

18:16:44 * crschmidt does the edd dance, then goes to lunch. really.

19:01:53 <mortenf> does anyone have specifics on http://www.hellomachine.net/ - i've seen it mentioned a couple of times near foaf...

19:24:00 * crschmidt should have asked for a dependancy list for dashboard before he started building it.

19:24:17 <crschmidt> i build gtk-sharp before gtkhtml, so it didn't get built with gtkhtml suppor.t

19:35:35 <bryce> bryce is now known as bryce_mp

19:59:35 <crschmidt> Anyone here know if Dashboard supports Mozilla/Fire*?

20:02:49 <jaseb> evening all

20:03:39 <crschmidt> hi, jaseb.

20:09:21 <jaseb> hi JibberJim

20:09:33 <JibberJim> Hey

20:09:48 <libby> hey jibs

20:09:51 * JibberJim is grabbing some very cold wifi as the trains at woking are so crap!

20:10:05 <libby> hm, that reminds me...

20:10:06 <libby> brrr

20:10:22 <mortenf> hi jim

20:10:23 <libby> jibs, you know anything about getting a good sim for a gsm modem?

20:10:48 <JibberJim> The Orange 25 quid a month unlimited (limited to 100mb) is the best deal AIUI,

20:11:05 <libby> hm

20:11:15 * libby was thinking of something very cheap at this stage :)

20:11:17 <xavier> is that 25 quid for just data or voice too?

20:11:37 <JibberJim> voice is extra.

20:11:50 <JibberJim> I think all of them are about the same a few quid a mb.

20:11:51 <libby> ah, I mean, I just want text for sending text mesaages

20:11:59 <xavier> and "unlimited" is limited to 100mb? heh.

20:12:01 <JibberJim> try mobitopian's they no more.

20:12:14 <JibberJim> or they know more at any rate.

20:12:18 <libby> for http://www.isis.de/members/~s.frings/smstools/

20:12:19 <JibberJim> just text messages?

20:12:26 <libby> ah good plan, thanks jibberjim

20:12:39 <libby> yeah I was going to make little playtools with sms

20:12:58 <libby> so on the spur of the moment I bought a gsm modem thing

20:13:18 <JibberJim> not with gsm modem though, just through net.

20:13:39 <libby> darn that prolly would have been fine

20:14:27 <libby> I really wanted to mms but apparantly that's pretty awful so I thought I'd have fun with texts instead

20:15:16 <JibberJim> SMPP is the Simple Message Protocol, SMS people talk it.

20:15:21 <JibberJim> or something like that.

20:15:47 <libby> heh, cool, thanks

20:16:48 <danbri> http://mayhem-chaos.net/blog/archives/000479.html re foaf:tipjar

20:18:55 <JibberJim> f'me it's cold

20:19:37 <xavier> 6°C here. it's nearly tropical :-)

20:20:00 * JibberJim 's only thermometer is his phone and that always seems to think it's 30C

20:20:29 * crschmidt sighs at dashboard.

20:20:31 * JibberJim was in the pub, but everyone left... I did get people to agree to come to the svg meet though.

20:20:44 <crschmidt> Too difficult to use, sadly.

20:21:05 <crschmidt> Anyone know who I'd poke about semaview not working with LiveJournal FOAF data.

20:21:12 <jaseb> shame, it actally looks promising

20:21:13 <JibberJim> I need to finish my Bluetooth thing like Edd's that was in SVG much better than dashboard

20:21:19 <xavier> crschmidt: plink doesn't either, fyi.

20:21:29 <crschmidt> no? Plink parsed the data i gave it...

20:21:42 <xavier> oh, it parses it. it just doesn't display it.

20:21:51 <xavier> from what I saw, anyhow.

20:22:10 <crschmidt> What would it display?

20:22:15 <crschmidt> it showed up with my contact info and all.

20:22:30 * JibberJim would like to send foafnaut against some LJ data - do you have a few urls you'd like, I could maybe do a limited scuttering?

20:22:30 <mortenf> hi leigh, i just pressed send on a reply re isbn

20:22:44 <JibberJim> it's too cold!

20:23:13 * ldodds checks email

20:23:24 <JibberJim> I got your zip ldodds#

20:23:35 <ldodds> cool

20:23:36 <mortenf> is that a threat?

20:23:38 * JibberJim needs gloves

20:23:39 <mortenf> :)

20:23:47 * ldodds tries to escape

20:23:54 <xavier> crschmidt: oh, right. it has no way to search for the user save by email address. and you can't walk it, as it doesn't use nicknames instead of real names.

20:24:00 * crschmidt nods.

20:24:04 <crschmidt> Yeah, that's an issue with a lot of tools.

20:24:11 <crschmidt> They all depend on names, and don't fall back to nicknames.

20:24:16 <ldodds> mortenf: glad you think it's reasonable that ISBNs are IFPs

20:24:20 <xavier> crschmidt: plink is really good at consolidating all info from any foaf file about a given person.

20:24:46 <mortenf> yep, let's just go for it, leave the problems to anyone else... :)

20:24:48 * JibberJim is frozen, gotta go - bye!

20:24:50 <ldodds> I didn't think plink smushed, or is that fixed now?

20:24:51 <crschmidt> JibberJim: You could start with mine, if you'd like: i've got 150 foaf:knows (about) http://www.livejournal.com/users/crschmidt/data/foaf

20:24:56 * ldodds grins

20:24:57 <mortenf> foaf explorer supports foaf:nick!

20:25:03 * xavier tosses JibberJim a heatpack

20:25:27 <crschmidt> i know!

20:25:29 <xavier> ldodds: "smush"?

20:25:31 <crschmidt> I was playing with that last night.

20:25:39 <crschmidt> xavier: take all information about a perosn and consolidate it.

20:25:44 * xavier nods

20:25:48 <crschmidt> so that crschmidt.net/foaf.xml and crschmidt.livejournal.com/data/foaf is both me.

20:26:01 * crschmidt can't get plink to load anyway ;P

20:26:16 <xavier> mortenf: foaf explorer supports bloody RSS feeds. it's def. the best out there.

20:26:30 <mortenf> heh, thanks, but i disagree!

20:27:05 <xavier> it's the only one that's liked my rdf:Alt tags..

20:28:07 * crschmidt tries the foaf explorer book marklet on an lj page...

20:29:02 <mortenf> i tried that earlier as well, worked

20:30:09 <crschmidt>http://beta.plink.org/profile.php?id=bbff51a6d70630daafe242c186a6e27fda3e99c7

20:30:12 <crschmidt> that's my plink profile.

20:30:33 <mortenf> leigh, what's the licensing terms for your allconsuming xsl - i'd like to "steal" and "improve" it?

20:30:54 <crschmidt> http://eikeon.com/foaf/?mbox_sha1sum=bbff51a6d70630daafe242c186a6e27fda3e99c7 hm.

20:31:26 <ldodds> public domain, do what you will with it.

20:31:34 <mortenf> great, thx

20:31:40 <ldodds> I fixed the errors you reported earlier, but forgot to upload it when I left work

20:32:01 <ldodds> what are you planning to change?

20:32:06 <mortenf> ah, ok, will fix (and remove trailing dot as well)

20:32:20 <mortenf> ^ at least, don't remember what else, probably nothing...

20:32:47 <mortenf> oh, of course, incorporate your book/reading vocab

20:32:51 <ldodds> the main thing still to do is patch over some inconsistencies in the rest interface, but that won't take long

20:32:54 <ldodds> yep, that's in too

20:33:06 <mortenf> in the one you haven't uploaded?

20:33:09 <ldodds> yep

20:33:14 <mortenf> ok, will wait...

20:33:29 <mortenf> what inconsistencies?

20:33:41 <crschmidt> bah.

20:33:47 <ldodds> sometimes comments are as an attribute, sometimes they're sub-elements

20:33:48 <crschmidt> people aggregator doesn't supoprt nicks either!

20:33:55 <mortenf> oh.

20:33:58 <ldodds> minor things really

20:34:12 <ldodds> when the Amazon API hiccups behind the scene the output gets mangled too

20:34:16 * mortenf signs up for allconsuming

20:34:35 <ldodds> what we need next is an "allconsuming" for musicbrainz

20:34:42 <mortenf> yeah, i know, as i said, i've got it as a cron job, but it's failed almost every day the last 2 weeks...

20:35:36 <mortenf> ah yeah, that'd be nice (alongside some ownership stuff)

20:36:22 <ldodds> Ownership is next on my list -- going to try to flesh out some docs for wiki page on Friday (train journey coming up)

20:36:31 <mortenf> great

20:36:50 <mortenf> btw, did you see my log comment re lowercasing the /NET/ part of the purl?

20:36:55 <ldodds> yep, did that too

20:37:01 <mortenf> ah :)

20:38:29 * crschmidt sends email to peopleaggregator re supporting foaf:nick

20:39:46 * crschmidt then realizes he's going to be late to class, and leaves.

20:39:48 <crschmidt> BBL.

20:39:57 <xavier> crschmidt: you should also be suggesting that people also support rdfs:seeAlso while you're at it. make it so the social connections get linked as well.

21:19:06 <CaptSolo> ldodds: what's the url for book rdf thing you just made?

21:19:17 <CaptSolo> i don't have it at home....

21:19:40 <mortenf> see http://www.ldodds.com/blog/archives/000126.html

21:32:03 <CaptSolo> thanks

21:33:07 <CaptSolo> i took a look at del.icio.us but find my current bookmarking is quite good

21:33:40 <CaptSolo> if bookmarks could be made like RSS feeds for blogs... where everybody bookmarks and del.icio.us aggregates them...

21:36:20 <mattmcc> You'd end up with something resembling Annotea Bookmarks.

21:38:04 <CaptSolo> matt: what do i need to use Annotea Bookmarks? does it integrate with mozilla?

21:38:58 <CaptSolo> what i imagine is - devote one blog to bookmarks and let such bookmarks site (lie Del.icio.us) feed from this blog's rss feed

21:49:24 <chrisMobile> what about people who bookmark 800 things a day? Wouldn't info be hard to find?

21:50:19 <mortenf> that's where rdf comes into play :)

21:50:38 <CaptSolo> morten precisely! :)

21:51:12 <CaptSolo> and that information about be displayed into hierarchy [is one of the views on this info]

21:51:35 <chrisMobile> I know my dad used to email his bookmarks from work to home every day.

21:51:56 <chrisMobile> He has a several megabyte bookmark file.

21:52:07 <CaptSolo> based on who reported it or what topic is associated with it (referencing some ontologies)

21:52:17 <chrisMobile> This was all years ago, of course.

21:52:55 <CaptSolo> chris: part of my bookmarks i keep in mozilla bookmarks file. and actually send it back & forth. to my web server from which i can take it as needed.

21:53:05 <CaptSolo> the other part is wrote on the weblog

21:53:21 * chrisMobile nods.

21:54:08 <CaptSolo> but when i browse and accumulate 10-20 open tabs in a browser window (use multizilla extension for mozilla for tabbed browsing) it is much easier for me to hit "Bookmark this group of tabs" than to place each url in the blog

21:54:18 <CaptSolo> (or in Del.icio.us)

21:54:54 <chrisMobile> Completely agree.

21:55:31 <chrisMobile> Is there an api for del.Ico?

21:55:37 <xavier> yup.

21:55:44 <CaptSolo> i think its home page mentions one

21:59:12 * crschmidt nods.

21:59:29 <crschmidt> Could you write some kind of mozilla plugin that would catch your mozilla bookmarks and put them in your del.ico.us?

22:06:08 * crschmidt waves to chrisMobile.

22:16:12 <CaptSolo> crschmidt: somebody could do that

22:16:30 <CaptSolo> then i would not mind Del.icio.us catching my bookmarks.

22:16:40 <CaptSolo> just that i want to do it the easy way :)

22:19:26 * crschmidt nods.

22:19:41 <crschmidt> I've never used del.ico.us, but it seems like it would easily allow you to do something like that. I think.

22:20:30 <crschmidt> s/ico/icio in everything I've said ;)

22:20:43 <CaptSolo> ok ;>

22:21:31 <OliB> wow this is totally hyper-cool http://www.yergler.net/projects/mozcc/install.html

22:21:34 * crschmidt creates an account.

22:25:27 <CaptSolo> crschmidt - an account? on del-icio-us?

22:29:19 <crschmidt> yep.

22:33:19 * crschmidt wonders if mortenf is still around.

22:33:30 <mortenf> somewhat...

22:34:08 <crschmidt> Complaint from a user that FOAF explorer chokes on his file.

22:34:10 <crschmidt>http://xml.mfd-consult.dk/foaf/explorer/?foaf=http%3A%2F%2Fwww.livejournal.com%2Fusers%2Fjackola%2Fdata%2Ffoaf

22:34:19 <crschmidt> Is there a size limit to FOAF files?

22:34:38 <mortenf> well, a practical one (and 60 seconds i think)

22:34:53 * crschmidt nods.

22:35:09 <crschmidt> I was just wondering what to tell him about limits - he's willing to edit his foaf file, just wondering what to shrink it down to.

22:35:26 <crschmidt> odd.

22:35:29 <crschmidt> Content-length: 8821

22:35:33 <crschmidt> that shouldn't be too big...

22:35:38 <crschmidt> oh, he may be using an external one now.

22:35:54 <crschmidt> hm, no.

22:36:05 <crschmidt> that's as big as his foaf file is apparently. That shouldn't break anything, should it?

22:36:50 <mortenf> it's 131465 bytes...

22:37:09 <crschmidt> right, but it comes through as gzipped...

22:37:22 <mortenf> and 2400 statements...

22:37:39 <crschmidt> So would it choke foaf explorer?

22:37:39 <mortenf> well, the point is that i still have to parse all the bytes as xml

22:37:43 * crschmidt nods.

22:37:47 <mortenf> yeah, that's what happens

22:37:50 <crschmidt> Okay.

22:37:58 <crschmidt> Any suggestions for how small it should be?

22:38:25 <mortenf> 30K perhaps

22:38:32 <crschmidt> otay.

22:38:37 <mortenf> but that'd still have you wait 30 seconds or so...

22:38:39 <IsoosI> can't he split it up into lotsa smaller files?

22:38:48 <mortenf> that's what i've done...

22:38:58 <crschmidt> So he would just do that with a series of seeAlsos?

22:39:03 <mortenf> yep

22:39:06 <mortenf> see mine

22:40:32 * crschmidt doesn't know where mortenf keeps his foaf file.

22:40:47 <mortenf>http://xml.mfd-consult.dk/foaf/explorer/?foaf=http://xml.mfd-consult.dk/foaf/morten.rdf

22:41:50 <crschmidt> cool.

22:47:24 <crschmidt> heh. I have a blog I've never used, and someone found a link to it - I've got 3 blog spams on there. http://peanut.sytes.net/blog/archives/000010.html

22:50:45 <CaptSolo> morten: re. your comment in ldodds blog that you do the cron yourself for Amazon wishlist.

22:51:04 <CaptSolo> do you run cron or you 'simulate cron' by doing conversion by hand?

22:51:30 <mortenf> see http://www.wasab.dk/morten/2004/01/amazon-pager/

22:54:31 <CaptSolo> rtfm all the time :>

22:54:35 <mortenf> heh

23:01:21 * crschmidt sighs.

23:01:40 <crschmidt> Okay, Location data and depictions of some kind seem to be the most widely demanded features for LiveJournal FOAF.

23:02:04 <mortenf> i think you're on the right track with foaf:logo or x:avatar

23:02:15 <crschmidt> okay.

23:02:28 <crschmidt> jbond's email to me said to just use foaf:img, but I really really don't like overusing that part of the spec.

23:02:36 <mortenf> agreed.

23:03:19 <crschmidt> Are there any current uses of foaf:logo?

23:03:38 <crschmidt> I'm assuming that most foaf:person's don't have a logo so much as an avatar, so it might be an accurate use of that part of the spec.

23:04:05 <mortenf> example: http://xml.mfd-consult.dk/foaf/explorer/?foaf=http://www.mfd-consult.dk/index.rdf

23:04:49 <crschmidt> The thin in the upper right is a logo?

23:04:59 <mortenf> there's no foaf:avatar, but it sounds like a good idea

23:05:00 <mortenf> yeah

23:05:21 <CaptSolo> morten - what does amazon-pager needs to run?

23:05:28 <crschmidt> Which is used in the same way as a foaf:img or foaf:depiction, isn't it?

23:05:36 <mortenf> just Perl

23:05:37 <crschmidt> If a FOAF file includes it, isn't that where it would show up on the explorer?

23:05:39 <CaptSolo> i do not have much control over the site my web is running

23:05:50 <mortenf> yep and yep

23:06:19 <crschmidt> hm.

23:06:33 <crschmidt> Is there anything like an rdf namespace that would include avatar yet?

23:06:42 <mortenf> haven't seen any...

23:07:00 <CaptSolo> but just running perl returned Can't locate LWP/UserAgent.pm

23:07:20 <crschmidt> you have perl and no LWP? You poor poor soul.

23:07:29 <mortenf> oh, i see. well, yeah, LWP::Useragent is needed...

23:08:03 <CaptSolo> anything else?

23:08:39 <mortenf> no, except Pod::Usage, but that should be in there

23:08:40 <CaptSolo> crshmidt: yes, if it was my box, i'd have everything needed. but on this server - there is what admin needs. apparently not LWP:Useragent

23:08:56 <crschmidt> it looks like it needs Data::Dumper too.

23:09:17 * crschmidt waves to Freso.

23:09:28 <mortenf> not really, i see i left it in there after debugging...

23:09:33 * Freso waves back to chris

23:09:37 <mortenf> am fixing it atm...

23:10:18 * Freso presented FOAF to a geek friend of his earlier today :D

23:10:44 * crschmidt has personally convinced 8 or 9 people of the usefullness of FOAF.

23:10:53 <Freso> Hehe :)

23:10:58 <crschmidt> Every time I describe it, I come up with better examples of what you might be able to do with it.

23:11:10 <Freso> Has anyone made an addressbook with FOAF in-/output yet btw?

23:13:10 <CaptSolo> crsch: have you put down the uses somewhere?

23:13:33 <CaptSolo> freso: would be a nice thing.

23:13:38 <CaptSolo> is there already?

23:13:58 <crschmidt> CaptSolo: That's what he was asking ;)

23:14:09 <crschmidt> CaptSolo: I'm going to try and do that over the next few days.

23:14:16 <crschmidt> CaptSolo: Schoolwork keeps getting in the way ;)

23:14:17 <CaptSolo> crsch: i know :P

23:16:09 <crschmidt> A lot of my uses for FOAF are kind of "This is what could happen when people write tools that do it."

23:16:16 <crschmidt> Rather than "this is what we can do with information we already have."

23:17:00 <crschmidt> Partially because of the data that is out there being limited, in the case of livejournal, and partially just because the tools to do the interpretation aren't done yet.

23:18:39 * mortenf received his scsi raid controller, is waiting for more disks, then it's scutter time...

23:20:05 <CaptSolo> i guess ldodds blog (a recent post in it) point to FOAF Application Ideas in rdfweb.org wiki

23:20:16 <CaptSolo> usage ideas might be different from application ideas

23:20:39 <CaptSolo> but you might either update the FOAF application page or create a new for FOAF Usage

23:21:00 <crschmidt> Well, I'll blog into my journal first.

23:21:11 <crschmidt> The people who read it there aren't Sem Webbers.

23:21:17 <crschmidt> And they're the kind of people we need to start convincing.

23:21:30 <crschmidt> Smart, geek-types - who don't understand what foaf is or how it can work for them.

23:22:19 <crschmidt> The best tool that FOAF could use right now is a tool that allows you to build a quick web of your friends, including all their FOAF information (and maybe Friends Of Friends, too) and then search through it easily.

23:22:40 <crschmidt> So I could do something like "Search friends and friends of friends for people with foaf:interest "HTML" " or whatever.

23:22:55 <crschmidt> With something like that, you start to build a network of information, which is what the SemWeb is all about.

23:23:34 <crschmidt> Big web tools are fine up to a point, but when you get to 2 million+ foaf files, you're going to be SOL when you try to crawl the whole mess.

23:24:33 <crschmidt> anyway, I'm going to go grab some dinner, then probably come back and do some semweb writing in my journal.

23:26:51 <CaptSolo> morten: i try to look at http://kaste.lv/~captsolo/semweb/foaf-captsolo.rdf in foaf explorer, but :(

23:27:13 <CaptSolo> it says cant parse FoaF - that it could be invalid

23:27:32 <CaptSolo> though it validates ok in rdf validator

23:27:39 <CaptSolo> Raw XSLT error message: Sablotron XSLT transformation error on line 13: conflicting variable bindings 'foaf'.

23:27:52 <xavier> CaptSolo: missing namespace?

23:28:05 <Freso> CaptSolo: I'm getting that as well.

23:28:39 * mortenf takes a look

23:28:44 <crschmidt> It doens' tparse through the validator.

23:28:46 <crschmidt> Exception parsing: {E301} The reference to entity "xmlfile" must end with the ';' delimiter.

23:29:24 <crschmidt> you didn't properly escape your rdf:about

23:29:31 <crschmidt> & should be &amp;

23:29:39 <mortenf> yep, that's it

23:29:55 <Freso> Chris: http://freso.dcginternet.co.uk/foaf.rdf - how about that?

23:30:07 <Freso> I'v elooked it through, can't find anything wrong with it.

23:30:25 <mortenf> try the validator...

23:30:29 <crschmidt> Why are you using xml 1.1?

23:30:33 <Freso> But... Explorer wouldn't parse it when I tried to show Konstantin... hm...

23:30:34 <crschmidt> can't validate 1.1, mortenf.

23:30:46 <mortenf> that's my point...

23:30:49 <Freso> Chris: Because I like staying ahead? :P

23:31:00 <crschmidt> Freso: What are you using from 1.1 that you can't from 1.0?

23:31:18 <crschmidt> The validator can't do xml 1.1, so try dropping it back down to 1.0 and running it through the validator.

23:31:27 <Freso> Nothing. But if I did 1.0 it would be exactly the same as 1.1.

23:31:28 <CaptSolo> crsch: right, my mistake

23:31:40 <Freso> Well, it's paresed it as 1.1 before.

23:31:44 <Freso> *parsed

23:31:56 <crschmidt> xception parsing: {E301} XML version "1.1" is not supported, only XML 1.0 is supported.

23:32:05 <Freso> Hm. Weird.

23:32:09 <Freso> Oh well.

23:32:17 <CaptSolo> cr: where should &amp go? in the url for the reading list?

23:32:19 <mortenf> why would you want 1.1?

23:32:34 <mortenf> CaptSolo: replace single & with &amp;

23:32:39 <crschmidt> what he said ;)

23:32:55 <crschmidt> freso: you're using "" around your version number, try ''

23:33:09 <xavier> crschmidt: both are valid.

23:33:17 <crschmidt> yes, but it seems that one confuses the validator ;)

23:33:55 * crschmidt shrugs. I'm going to dinner, poke me when i get back if you havne't fixed it.

23:35:12 * mortenf still doesn't get why 1.1 is wanted...

23:37:03 * Freso shrugs

23:37:18 <xavier> 1.1 pretty much just upgrades some bits to unicode, right?

23:37:26 <mortenf> no...

23:37:26 <Freso> Well...

23:37:42 <mortenf> http://www.wasab.dk/morten/2004/01/amazon-pager/ is now at 0.2

23:37:43 <Freso> 1.0 has a certain range of characters you _are_ allowed to use.

23:37:53 <Freso> 1.1 has a range which you're _not_.

23:38:07 <CaptSolo> now its ok :)

23:38:10 <CaptSolo> at last!

23:38:13 <Freso> 1.1 can therefore expand theoretically infinitely.

23:38:38 <Freso> Allowed-characters-wise.

23:38:59 <CaptSolo> aha - and foaf exploring now 'eats' it too :)

23:39:53 <xavier> Freso: ah, i see.

23:40:35 <Freso> xavier: At least, that's how I've understood it :)

23:44:43 <CaptSolo> morten, crsh - thanks

23:44:51 <CaptSolo> now i shall sleep :)

23:44:54 <CaptSolo> take care

23:45:12 <xavier> ChrisDodo: have you had any luck finding barcode generators?

23:45:42 <ChrisDodo> not free ones. in english anyway.

23:45:50 <xavier> curses.

23:46:11 <xavier> i found a site that would do it, PDF417 anyhow. but nothing downloadable.

23:46:17 <ChrisDodo> it can't be that hard to write one. someone ;)

23:46:33 <ChrisDodo> yeah, the websites are pretty fickle about what they encode

23:46:38 <xavier> the harder part will be the reader.

23:47:37 <ChrisDodo> i have been told, by different people, it really isn't that hard. now to convince them to write it :)

23:48:44 <xavier> this place has one for PDF417: http://www.turbulence.org/Works/swipe/main.html

23:48:49 <xavier> but only a specific data density.

23:49:01 <xavier> (i believe)

23:51:14 <crsMobile> i just figured out that my battery was slightly loose, causeing my phone to seemingly randomly turn off.

23:51:38 <CaptSolo> xavier - 2d barcodes? (pdf417 is 2d if i remember correctly)

23:52:21 <CaptSolo> the reader - yep, that might be harder

23:52:23 <xavier> CaptSolo: yup. ChrisDodo's idea was to encode RDF into 2d barcodes.

23:52:32 <ChrisDodo> but as the japanese QR Code

23:52:40 * xavier nods

23:52:49 <CaptSolo> the ones i knew used OCR and were expensive

23:53:33 <crsMobile> I guess normal barcodes are considered 1d?

23:53:34 <xavier> ok, gotta go. server is being upgraded

23:53:42 <ChrisDodo> cameraphones can do it. i have seen it done. most jap phones do it. c'est ne pas un problem :)

23:54:13 * crsMobile has camera phone!

23:54:34 <CaptSolo> chris: nice :)

23:54:39 <crsMobile> Remember, though, that japaneses image processing is about 3 to 5 years ahead of the us.

23:54:44 <CaptSolo> what resolution does it need?

23:54:46 <CaptSolo> crs: yup

23:55:13 * crsMobile gets 640x480.

23:55:13 <CaptSolo> crs: true, and their mobile technologies as well

23:55:26 <crsMobile> I just like it cause i can run irc from it.

23:55:32 <CaptSolo> crs: what camera? nokia 7650 does 640x480

23:55:47 <crsMobile> The us version of that.

23:55:49 <crsMobile> 3650.

23:55:54 <CaptSolo> the reader i used was http://cbnco.com/id/idproducts.html

23:56:10 <CaptSolo> i thought 3650 has 320x240 or like that

23:56:29 <CaptSolo> but this reader was bigger than phone and for validating passports and id cards anyway

23:58:07 <CaptSolo> yes, vga resolution...

23:58:40 <ChrisDodo> i've seen it done on a p800. series 60 shouldn't be a problem too.

23:59:05 <CaptSolo> Chris: it done = recogniszing 2d barcodes?

23:59:38 <ChrisDodo> yes.