Semantic Web Interest Group IRC Chat Logs for 2004-09-22

02:01:00 <oommoo> aloha

02:02:42 <crschmidt> Hola, oommoo

02:04:09 <Skyline> Skyline is now known as pjenvey

02:10:57 <oommoo> SKOS does look interesting, has anyone tried to do document classification using it's notion of concepts ?

02:15:01 <oommoo> crschmidt: what is you current interest ?

02:15:21 <crschmidt> oommoo: mostly working with Redland to create a usable tool for querying rdf in community situations.

02:15:25 <crschmidt> For example:

02:15:27 <crschmidt> ^name crschmidt

02:15:27 <julie> Chris Schmidt, Christopher Schmidt

02:15:38 <crschmidt> ^sha danbri

02:15:38 <julie> 362ce75324396f0aa2d3e5f1246f40bf3bb44401, f5c0bc399c6faf8a00d93fc2f69c497336790022, 241021fb0e6289f92815fc210f9e9137262c252e

02:16:05 <crschmidt> (If you have a foaf file, you can add it with ^add http://urlhere

02:17:17 <crschmidt> Also, as always, i spend time regularly explaining why LiveJournal's FOAF output is not a security or privacy concern, things like that

02:20:28 <oommoo> crschmidt: is the implementation strongly IRC biased ?

02:20:50 * oommoo is building a current vintage jabberd as we type

02:21:31 <crschmidt> oommoo: Not really. I'm working on an AIM implementation, attempting to use the Perl Redland bindings, but there's a problem with running queries that I need to resolve first

02:21:45 <crschmidt> and a combination of real life timesuck and laziness has prevented me from contacting the people who can help me with it ;)

02:47:24 <oommoo> kewly, I now have a foaf.rdf (thanks to foaf-a-matic) . . . http://bancroft.dlogic.org/rdfig/foaf.rdf

02:47:47 <oommoo> ^add http://bancroft.dlogic.org/rdfig/foaf.rdf

02:48:09 <oommoo> ^oommoo

02:52:43 * oommoo eyes crschmid1 warily and looks back to crschmidt

02:53:04 <oommoo> crschmidt: fan club or alter ego . ? !

02:53:10 <crschmid1> heh

02:53:23 <crschmid1> crschmid1 is now known as crschmidt

02:55:56 <crschmidt> Internet connection woes.

02:56:09 <oommoo> ouch

02:57:34 * oommoo is building packages for a new server

02:57:47 <oommoo> thanks for reminding me to build out the redland bits

04:07:25 <bparsia> bparsia is now known as bijan

09:16:00 <danbri_dna>http://www.technologyreview.com/articles/04/10/frauenfelder1004.asp

09:16:01 <dc_rdfig> A: http://www.technologyreview.com/articles/04/10/frauenfelder1004.asp from danbri_dna

09:16:13 <danbri_dna> A:|TimBL interview in MIT Technology Review

09:16:13 <dc_rdfig> Titled item A.

09:16:54 <danbri_dna> A:The [http://www.technologyreview.com/articles/04/10/frauenfelder1004.asp?p=0|print version] might be more convenient to read.

09:16:55 <dc_rdfig> Added comment A1.

09:17:13 <danbri_dna> A:" I’m three photographs from Frank Sinatra because I’m photographed with Bill Clinton who’s been photographed with one of the Kennedys who’s been photographed with Frank Sinatra. That’s a silly application, but it really shows the power of the reuse of information."

09:17:13 <dc_rdfig> Added comment A2.

09:28:59 <libby> it's not silly!

09:29:03 <libby> well, maybe a little

09:41:20 <danbri_dna> it is so! :)

09:41:50 * libby prefers to think of it as 'fun' ;)

09:46:54 <leobard> wazaa, who is alive?

09:47:32 <leobard> I want to implement an annotea server on top of jena, any ideas?

09:56:07 <leobard> maybe the annotea goal is too high :-)

10:00:38 <libby> it should be easy enough but I don;t know enough about either to help, sorry

10:20:46 <danbri_dna> leobard, one thing you could try is to do a 'read only' implementation first

10:21:16 <danbri_dna> ie. populate the database through some other mechanism, and then just engineer a server that can send annotations to the client in the appropriate form

10:21:37 <danbri_dna> I have some code that does that (in Ruby, with Squish queries). Rewriting in Jena should be trivial.

10:22:55 <danbri_dna>http://rdfweb.org/2002/foaf/scutter/bin/annotea_server

10:22:56 <dc_rdfig> B: http://rdfweb.org/2002/foaf/scutter/bin/annotea_server from danbri_dna

10:23:08 <danbri_dna> B:|A basic read-only annotea server in Ruby

10:23:08 <dc_rdfig> Titled item B.

10:23:25 <danbri_dna> B:Probably code-rotted, and the same files are offline (sorry). But should show a basic approach.

10:23:25 <dc_rdfig> Added comment B1.

10:24:32 <danbri_dna> B:Uses Squish notation for RDF queries (an RDQL precursor), but should map very directly into RDQL, if you want to do this in Jena.

10:24:33 <dc_rdfig> Added comment B2.

10:25:14 <danbri_dna> B:Of course you'll need some other source of annotations. My idea was to populate the database with a generic RDF crawler that traverses rdfs:seeAlso links, and to encourage people to use the Annotea vocab in RDF files they publish elsewhere in the Web.

10:25:15 <dc_rdfig> Added comment B3.

10:25:46 <danbri_dna> B1:Probably code-rotted, and the sample files are offline (sorry). But should show a basic approach.

10:25:46 <dc_rdfig> Replaced comment B1.

10:26:39 <danbri_dna> B:Another day, another half-finished project. [http://esw.w3.org/topic/DontWorryBeCrappy|DontWorryBeCrappy]?

10:26:39 <dc_rdfig> Added comment B4.

10:27:09 <peepo> anyone going to see cat stevens in NY?

10:28:19 <danbri_dna> not me. I thought he'd been ejected anyway, http://news.bbc.co.uk/2/hi/americas/3678694.stm

10:30:13 <libby> yeah, he has

10:30:34 <libby> they actually turned the plane around cos he was on a blacklist (or, his name, anyway)

10:31:27 <danbri_dna> he's made a few dumb public statements in his time, I think...

10:31:32 <danbri_dna> but i like 'peace train'

10:47:42 <Emmy> Hi all :)

10:52:27 <danbri_dna> hi emmy

10:53:01 <Emmy> hi danbri! :)

11:07:57 <crschmidt> morning all

11:57:34 * dajobe watches ht present http://www.cogsci.ed.ac.uk/~ht/webpropernames/

12:10:54 * danbri_dna wonders if ht's following the www-rdf-interest permathread it restarted

12:20:33 <leobard> dc_rdfig:view

12:20:33 <dc_rdfig> A: TimBL interview in MIT Technology Review (http://www.technologyreview.com/articles/04/10/frauenfelder1004.asp)

12:20:34 <dc_rdfig> B: A basic read-only annotea server in Ruby (http://rdfweb.org/2002/foaf/scutter/bin/annotea_server)

12:21:19 <leobard> dc_rdfig:morehelp

12:21:19 <dc_rdfig> Put emphasis in a comment by using *asterisks*

12:21:20 <dc_rdfig> To create an inline link in a comment, say:

12:21:21 <dc_rdfig> A:Look at [this thing here|http://pants.heddley.com]

12:21:22 <dc_rdfig> You can also link to inline images in a comment:

12:21:23 <dc_rdfig> A:Chump logo +[alt-text|http://pants.heddley.com/chump.png]

12:21:24 <dc_rdfig> To see the last n links, say dc_rdfig:view n (where n is a number)

12:21:25 <dc_rdfig> To see the details of a link labelled A, say A: on a line on its own

12:21:26 <dc_rdfig> To view a particular comment, say An:, where n is the number of the comment

12:21:27 <dc_rdfig> To replace, say, the second comment on a link labelled A, say A2:replacement_text

12:21:28 <dc_rdfig> To delete the second comment on a link labelled A, say A2:""

12:21:30 <dc_rdfig> To set keywords for a link labelled A, say A:->keyword1 keyword2 etc.

12:21:32 <dc_rdfig> Send any comments or questions to chump@heddley.com

12:27:54 <crschmid1> crschmid1 is now known as crschmidt

12:58:37 <adolflow> hola

12:58:49 <balbinus> hi / hola

12:58:56 <inkel> hola adolflow

12:59:21 <chaalsNCE> hola

13:00:06 <adolflow> hi chaals, are you going to oviedo?

13:01:42 <chaalsNCE> No idea at this stage.

13:01:44 <chaalsNCE> :-(

13:02:49 <dajobe> danbri_dna: I mentioned the permathread to henry's co-author

13:02:57 <dajobe> or danbri

13:14:25 <adolflow> it's about "tutorial on basic semantic web technlogies" http://www.euitio.uniovi.es/actividades/cursos.php3?key=60

13:15:27 * em gets a "Unable to connect to PostgresSQL server" error on this URL

13:35:10 <chaalsNCE> adolflow, si. Ho visto pero no tengo nada a ver entonces si voy es solo por curiosidad

13:58:07 * ndw wonders if DanC is around

13:59:59 <DanC> if it's about issue tracking, I can spare some cycles, ndw. I'm just tidying up DAWF ftf minutes.

14:00:21 <ndw> It's indirectly about issue tracking.

14:00:26 <DanC> do tell

14:00:50 <ndw> I want to get rid of spam, so I've arranged to tag all the non-spam messages as ?msg audit:class "real"

14:01:14 <DanC> hmm... I usually just respond to spam messages with a [closed] msg.

14:01:16 <ndw> Now I want to copy the RDF file, discarding all messages that aren't "real" and it's a bit of cwm hackery just outside my grasp

14:01:24 <DanC> you could get fancy and use the new annospam DB, maybe.

14:01:49 <ndw> There are just enough messages to be annoying and just few enough to make it worth my while to build a little spamdb of my own.

14:01:50 <DanC> marking all the real messages as "real" is just the sort of centralized audit step that I'm striving to avoid.

14:02:35 <DanC> the annospam thingy is somewhat operational, fyi. Dom has cleaned some 6000 messages out of our archives.

14:02:35 <ndw> Well, maybe [spam] instead of [closed] and move them to the bottom of the summary or something would be better, but that's well beyond my current groking of the system

14:02:47 <ndw> pointer to annospam?

14:03:10 <DanC> umm... put ,spam at the end of an archive cover page for some month...

14:03:17 <ndw> 'k

14:03:29 <DanC> I think only the team has write access to say "That's spam" though.

14:03:53 <DanC> support for a [spam] tag is straightforward... let's see... have you made a copy of the issue tracking stuff somewhere?

14:04:19 <DanC> you're working on xinclude?

14:04:28 <ndw> yes, /XML/2004/xml-id/wd-status

14:04:35 <ndw> I'm working on xml-id

14:04:59 * bparsia had an interesting discussion with J. Marsh and D. Orchard about xinclude

14:05:05 <bparsia> I hope it takes off :)

14:05:06 * DanC chacls...

14:05:07 <ndw> I just committed a snapshot

14:05:10 <bparsia> bparsia is now known as bijan

14:05:24 <DanC> er... did you just chacl too?

14:05:38 <ndw> Uhm, it's public by virtue of being under /XML, I think

14:05:51 <DanC> ah... /XML ... "[cannot be changed with this tool]"

14:05:54 <ndw> right

14:06:05 <DanC> well, the directory isn't public. I suppose that's OK...

14:06:55 <DanC> I can see http://www.w3.org/XML/2004/xml-id/wd-status/Makefile but I'm having trouble seeing status-report.html

14:07:43 <DanC> anyway, in comment-rules.n3, do you see str:notMatches "^\\[(close|Close|move|cosed|clodes)",

14:07:43 <DanC> ?

14:07:47 <ndw> yes

14:07:51 <DanC> add spam there.

14:07:54 <DanC> 2 places

14:07:59 <ndw> Ok.

14:08:12 * DanC apologizes for not factoring out the regex

14:08:49 <ndw> Right. So that will put spam threads in the closed category. But ideally, I'd like a new category down at the bottom

14:09:05 <DanC> hmm... 404 @ http://www.w3.org/XML/2004/xml-id/wd-status/spam.n3

14:09:12 <ndw> oh.

14:09:32 <DanC> oh. well, for a new category, you might have to think a bit.

14:09:52 <DanC> most of the work would be in status-report-rules.n3

14:09:52 <ndw> right. so I thought I'd cheat and just filter them out :-)

14:10:04 <ndw> Ok. I'll take a look and see what I can come up with.

14:10:23 <DanC> well, I strongly recommend against keeping state outside the mail archive.

14:10:31 <ndw> fair enough

14:10:43 <DanC> the main feature of this system, IMO, is that it's just a few of the mail archive.

14:10:58 <ndw> yes. that's true. ok, I'll try the [spam] trick.

14:11:10 <ndw> Hmm. I don't have copies of those messages to reply to, but I suppose I can fake the references headers...

14:11:22 <DanC> you can just click [reply] in the archive.

14:11:31 <ndw> ah, yes.

14:11:33 <DanC> it does The Right Thing with references

14:12:59 <DanC> speaking of which... my TODO list includes cleaning up public-webarch-comments to make up for mailers that do The Wrong Thing with references. sigh.

14:13:23 <ndw> It's "Subject: [spam] old subject" right?

14:13:29 <DanC> yes

14:14:19 <ndw> Hmm. Some of those messages "have been deleted from the archive"

14:14:43 * DanC honestly didn't think anybody else would want to use this system; wonders if ndw shares some deviant personality trait with me... ;-)

14:15:15 <ndw> Heh. Perhaps. Email is just *exactly* how I want to track issues. Experience demonstrates that doing it in a separate document is a nightmare.

14:15:17 <DanC> ah... annospam in action... hm... I wonder if you can get the data on which ones in RDF somehow...

14:16:09 <DanC> ah... so the "it's just a few of the email" feature appeals to you too. Yes, it's depressing that email is the only thing that is sufficiently robust, but there you are.

14:16:37 * DanC hunts for annospam XSLT implementation details...

14:17:54 * DanC wonders if http://lists.w3.org/Archives/Public/public-xml-id/2004Sep/,spam is visible to ndw or anybody else

14:18:14 <DanC> it's powered by http://www.w3.org/2003/08/kill-spam

14:18:19 <bijan> I can see it

14:18:39 * crschmidt can't

14:18:56 <ndw> I can see it

14:19:02 <crschmidt> Not that this is a huge surprise, since it asks for a login that i don't have ;)

14:19:12 <bijan> But the mark as spam button is W3C-Team only

14:19:13 <DanC> can you see 2003/08/kill-spam , ndw?

14:19:24 <DanC> incl this line: <xsl:variable name="isAnnotated" select="document(concat('http://annotest.w3.org/annospam/annotations?w3c_forceText=*&amp;w3c_algaeQuery=%28ask+%27%28%28http%3A%2F%2Fwww.w3.org%2F1999%2F02%2F22-rdf-syntax-ns%23type+%3Fannotation+http%3A%2F%2Fwww.w3.org%2F2003%2F08%2Fmlspam%23ArchivedSpam%29+%28http%3A%2F%2Fwww.w3.org%2F2003%2F08%2Fmlspam%23listedBy+%3Fannotation+',$baseURI,'%29%28http%3A%2F%2Fwww.w3.org%2F2000%2F10%2Fann

14:19:24 <DanC> otation-ns%23annotates+%3Fannotation+%3Fmessage%29%29+%3Acollect+%27%28%3Fannotation+%3Fmessage%29%29&amp;w3c_submit=query+RDF+DB'))/rdf:RDF"/>

14:19:53 <ndw> Yes, I can see kill-spam

14:20:28 <DanC> kill-spam seems to have the details about how to ask the annotest.w3.org/annospam DB "which msgs are spam?"

14:20:36 <ndw> Uhm.

14:20:51 <ndw> I don't actually see the xslt, I just see the web page it produces, I thik.

14:20:56 <DanC> view source

14:21:21 * ndw chuckles. yeah, why didn't I think of that

14:22:02 <ndw> Ok. I'll study this for a bit.

14:22:10 <DanC> enjoy

14:24:36 <jeen> dajobe, you around?

14:25:02 <dajobe> a bit

14:25:05 <dajobe> tea becogns

14:25:42 <jeen> right, i'll ask you when you get back then :)

14:25:51 <dajobe> oh go on!

14:25:57 <jeen> heh

14:26:18 <jeen> I was wondering, we're struggling a bit with how to properly handle \0 unicode chars in Sesame.

14:26:39 <jeen> since they're legal unicode (so should be allowed in rdf literals), but are not allowed in XML.

14:26:40 <dajobe> ouch

14:26:48 <jeen> how does redland handle those?

14:26:53 <dajobe> they are also handled poorly by Java

14:27:09 <dajobe> I import cstrings

14:27:16 <dajobe> but internally use counted strings IIRC

14:27:34 <dajobe> never really considered it much

14:27:40 <dajobe> ask an xml expert,.... ndw!

14:28:08 <DanC> \0 is a legal unicode character?

14:28:13 <jeen> yes

14:28:22 <DanC> source, pls?

14:28:26 <jeen> at least that's what arjohn tells me, and he's been reading the spec quite a bit lately.

14:28:31 <jeen> hang on

14:28:54 <jeen> see http://www.openrdf.org/forum/mvnforum/viewthread?thread=241&offset=10#953

14:29:00 <DanC> in any case, I'm not aware of any claims that RDF literals can express all unicode characters. They can only express those allowed in XML.

14:29:04 <golbeck_> golbeck_ is now known as golbeck

14:29:34 <DanC> BLURB: issue tracking for webarch, xml-id; annospam

14:29:36 <dc_rdfig> C: issue tracking for webarch, xml-id; annospam from DanC

14:29:37 <dajobe> (actually redland uses utf8 strings)

14:29:41 <DanC> logger, chump C:

14:29:41 <DanC> C:See [http://www.ilrt.bris.ac.uk/discovery/chatlogs/rdfig/2004-09-22#T14-29-41|discussion]

14:29:42 <ndw> danc, can you send me the source for http://www.w3.org/2002/08/xslt4html

14:29:42 <dc_rdfig> Added comment C1.

14:30:23 <DanC> ndw, when I go there I get redirected to http://www.w3.org/2001/05/xslt

14:30:53 <dajobe> tea

14:30:58 <ndw> yeah. so do i. odd. but that's the action for the form...hmmm.

14:31:00 <DanC> which says "It is a simple wrapper (source) around James Clark's XT and ..."

14:31:15 <DanC> with _source_ linked to http://dev.w3.org/cvsweb/java/classes/org/w3c/app/xsl/

14:31:25 * ear1grey perks up at the mention of tea

14:31:27 <ndw> Hmmm.

14:31:46 * ndw tries to unwind this damn thing to figure out how the xmlfile parameter winds up starting with /cgi-bin

14:33:14 <jeen_> DanC, re the rdf spec, the way I read it any unicode string in NFC is allowed.

14:33:48 * DanC checks the spec... thought we covered this...

14:34:15 <jeen_> oh i see arjohn has done a posting about this to the rdfig list. he probably explains it better than me.

14:35:24 * DanC wonders if Unicode Normal Form C allows 0

14:35:54 <CaptSolo> hi all

14:36:58 <DanC> my advice, jeen_, is that sesame should handle \0 by throwing an exception.

14:37:18 <jeen_> I was afraid of that

14:37:36 <jeen_> it means we have to do a parse of the string when we create a literal object. Expensive :(

14:37:42 <DanC> meanwhile, I don't think the spec (http://www.w3.org/TR/rdf-concepts/#section-Graph-Literal) is all that clear. I think a message to www-rdf-comments is in order.

14:39:28 <DanC> does expat check for char zero? yup. <b>&#0;</b> gives: ,zero.xml:1:3: reference to invalid character number

14:41:47 <DanC> re that openrdf forum... the RDF spec doesn't actually come right out and say "every sequence of Unicode characters is an RDF literal"

14:41:59 <DanC> it only says "every RDF literal has a lexical form which is a sequence of unicode chars"

14:42:56 <jeen_> right, so it's a matter of interpretation :)

14:43:32 * bijan plumps for only haveing one RDF literal

14:45:39 <DanC> yes, and www-rdf-comments is obliged to answer questions of interpretation

14:45:50 <jeen_> we'll send a comment there, thanks

14:46:19 <ndw> Hmm. DanC, the rdf is full of mid: URI and the spam script needs http: URI. Is there an easy way to map them?

14:46:55 <DanC> umm... http://www.w3.org/mid/$mid will redirect.

14:47:07 <DanC> I dunno if that qualifies as "easy"

14:47:43 <ndw> I don't think so. I need to get the text of the URI.

14:48:02 * DanC fought a small war to get the archive URIs to just _be_ /mid/$mid and eventually conceded that it's insufficiently usable. too hard to read such URIs during a telcon

14:48:30 <ndw> Well, mid: would be fine, if that's what was going into the annotation server, but it isn't :-(

14:49:35 <DanC> umm... you could write a little python program to read xml-id-comments.rdf , GET /mid/$mid , trap the redirect, and write out the results.

14:49:56 <DanC> kinda messy, but it's the only thing I can think of just now

14:50:52 * DanC noodles on putting that info in the annotea DB somehow...

14:52:02 <jeen_> jeen_ is now known as jeen

14:53:10 <ndw> yeah, that's what I'll hvae to do

14:55:30 <DanC> logger, pointer?

14:55:30 <DanC> See http://www.ilrt.bris.ac.uk/discovery/chatlogs/rdfig/2004-09-22#T14-55-30

15:17:25 <DanC> C:see [http://www.w3.org/XML/2004/xml-id/wd-status/Makefile|ndw's xml-id status Makefile], based on [http://www.w3.org/2001/tag/2004lc/Makefile| DanC's webarch makefile]

15:17:25 <dc_rdfig> Added comment C2.

15:18:08 <DanC> C:and other bits and bobs, such as [http://www.w3.org/2003/08/annospam2spamlist.xsl|annospam2spamlist.xsl] and [http://www.w3.org/2003/09/uri2mid|uri2mid]

15:18:09 <dc_rdfig> Added comment C3.

15:18:43 <DanC> C:and [http://www.w3.org/2003/08/kill-spam|kill-spam]

15:18:43 <dc_rdfig> Added comment C4.

15:20:38 <dom_> I see that http://www.w3.org/2002/08/xslt4html was mentionned before

15:21:28 <dom_> it's a simple PHP script that does directly "| tidy | XSLT servlet"

15:21:57 <dom_> (since you could not make an HTML form that would do the proper url encoding stuff that would be needed)

15:54:01 <danbri> ping #rdfig

15:54:18 <danbri> anyone mind if the semi-scheduled chat planned for #foaf moves into this channel?

15:54:31 <danbri> for 1hr, starting in 30 mins from now, ie. 17.30 BST

15:55:05 <CaptSolo> silence :)

15:55:08 <crschmidt> heh

15:55:13 <crschmidt> Sounds like it wouldn't be abig deal!

15:55:56 <danbri> speak up in the next 60 seconds or forever hold your silence!

15:57:49 <dajobe> hmm, topic is rather bare

15:58:12 <dajobe> dajobe has changed the topic to: Semantic Web / RDF chat n hack. blog http://rdfig.xmlhack.com/ logs http://www.ilrt.bris.ac.uk/discovery/chatlogs/rdfig/

15:58:29 <dajobe> now add your topic to the end

15:58:43 <JibberJim> hmm, not really forever danbri, surely just until after the chat

15:59:38 <danbri> FOREVER, jim ;)

16:00:45 <danbri> danbri has changed the topic to: #foaf invaders borrowing channel for a semi-scheduled chat, 17.30BST for 1hr /// Semantic Web / RDF chat n hack. blog http://rdfig.xmlhack.com/ logs http://www.ilrt.bris.ac.uk/discovery/chatlogs/rdfig/

16:00:46 <JibberJim> ooh, that's ominous a shouted threat and he's op'd... I'll shut up...

16:00:57 <dajobe> what's the topic of the chat, danbri

16:01:03 <danbri> foafy stuff

16:01:10 <crschmidt> FOAF ontology discussion

16:01:18 <danbri> I'll hack together the agenda in chump, hence the channel switch

16:12:29 <crschmidt> Bah, leaving in 5 minutes for lunch with boss.

16:13:36 <crschmidt> so, won't be leading a dob discussion

16:13:39 <libby> bummer chris - maybe you'll catch the end?

16:13:43 <crschmidt> maybe, maybe

16:13:49 <libby> when will you be back you think?

16:13:50 <crschmidt> reviewing for a raise htough, so i guess it's wroth it :)

16:13:57 <libby> heh, in that case, yeah :)

16:14:46 <crschmidt> ^q ?nick where (?p foaf:dateOfBirth ?bday) (?p foaf:nick ?nick) AND ?bday =~ /09-22/

16:14:47 <JibberJim> ooh it's a DOB discussion?

16:14:57 <JibberJim> I'll be here 'til 6ish now

16:15:03 <crschmidt> hm.

16:15:05 <JibberJim> assuming coffee shop remains open...

16:15:12 <crschmidt> apparently she's thinking...

16:15:16 <julie> elisesara, jessical, jenett

16:15:31 <crschmidt> there you go

16:16:01 * crschmidt rel:partner [a foaf:Person; foaf:nick "jessical"] .

16:16:23 <danbri> OK so agenda items I've heard: action item review, benjamin on stability vocab, crschmidt on dateOfBirth vocab; anything else?

16:16:37 <danbri> oh, website update too

16:18:45 <JibberJim> did you see my feedback on the website new look?

16:21:40 <danbri> Jim, no... but I may've just read past it in hurry; if so, sorry!

16:22:18 <libby> I think jim's point was that the font was too small to read and that it wasn;t valid (though nearly)

16:24:15 <JibberJim> yep, the fly-shit was the main problem.

16:24:23 <JibberJim> 70% font-size is no good.

16:24:45 <esigler> Question: St. Andre mentioned the "lack" of a community standards process for FOAF (like JEPs for Jabber), is there anything on an agenda anywhere to perhaps capture the current "process" for FOAF? Is anyone planning on sitting down and writing something that says "here's how to get something into / out of the FOAF vocab"?

16:25:06 <danbri> ack'd

16:25:31 <danbri> agenda+ FoafCommunityProcess, FOAF and standards, and relationship to W3C

16:25:34 <danbri> that cover it?

16:25:41 <esigler> yeah

16:25:56 <danbri> could you find a url for peter's msg, to provide context for others?

16:26:00 <esigler> sure

16:27:40 <esigler> here: http://www.saint-andre.com/blog/2004-09.html#2004-09-12T19:56 (relevant: "although I still think the FOAF community needs something like the JSF's JEP process, they do have an informal FOAFCommunityProcess")

16:28:23 <libby> ok, shall we use the chump for those items? we can alsways add more later

16:28:52 <libby> .t utc

16:28:54 <phenny> 2004-09-22T16:28:53Z

16:29:07 <libby> BLURB: FOAF meeting 16.30-17.30 UTC

16:29:07 <dc_rdfig> D: FOAF meeting 16.30-17.30 UTC from libby

16:29:55 <libby> BLURB: FOAF meeting agenda item 1 (E) Stability Vocab

16:29:55 <dc_rdfig> E: FOAF meeting agenda item 1 (E) Stability Vocab from libby

16:30:31 <libby> E:Benjamin/bengee (can someone add url of Benjamin's mail?)

16:30:31 <dc_rdfig> Added comment E1.

16:30:50 <bengee> hi

16:31:13 <libby> BLURB: FOAF meeting agenda item 2 (F) Web site update

16:31:13 <dc_rdfig> F: FOAF meeting agenda item 2 (F) Web site update from libby

16:31:26 <libby> F:including jim's concerns re 'flyshit'

16:31:27 <dc_rdfig> Added comment F1.

16:31:59 <libby> BLURB: FOAF meteing agenda item 3 (G) Foaf community process

16:31:59 <dc_rdfig> G: FOAF meteing agenda item 3 (G) Foaf community process from libby

16:32:11 <danbri> thanks libby

16:32:35 <libby> G:see [http://www.saint-andre.com/blog/2004-09.html#2004-09-12T19:56|Peter St. Andre's post]

16:32:35 <dc_rdfig> Added comment G1.

16:33:07 * danbri digs out prev logs

16:33:08 <JibberJim> fly-shit normally has a more polite (in english) german translation, but it's well known in CSS circles...

16:33:09 <libby> BLURB: FOAF meeting agenda item 4 (H): date of birth

16:33:09 <dc_rdfig> H: FOAF meeting agenda item 4 (H): date of birth from libby

16:33:40 <libby> H: crchmidt wanted to talk about this: he might be back if we put it last

16:33:40 <dc_rdfig> Added comment H1.

16:33:47 <libby> do add any relevant urls

16:34:06 <danbri> H:[http://www.ilrt.bris.ac.uk/discovery/chatlogs/foaf/2004-09-15.html#T16-00-00|last week's log]

16:34:07 <dc_rdfig> Added comment H2.

16:34:21 <bengee> E1:See [http://rdfweb.org/pipermail/rdfweb-dev/2004-September/013699.html] mail to rdfweb-dev

16:34:21 <dc_rdfig> Replaced comment E1.

16:34:24 <libby> D:attending: [libby miller|http://ilrt.org/people/libby]

16:34:25 <dc_rdfig> Added comment D1.

16:34:41 <libby> please add yourselves to D as i just did if you're here for the meet

16:34:53 <danbri> BLURB previous-action: danbri update FOAF spec to add a main section on naming, to provide a unified treatment of first/last/given/family

16:35:13 <danbri> BLURB: Previous actions

16:35:13 <dc_rdfig> I: Previous actions from danbri

16:35:14 <libby> [apologies if you're not here for the foaf meet: we've moved from #foaf here for an hour because there are better tools)

16:35:32 <JibberJim> D:attending until I get kicked out the cafe [Jim Ley|

16:35:32 <dc_rdfig> Added comment D2.

16:35:45 <JibberJim> D2:attending until I get kicked out the cafe [Jim Ley|http://jibbering.com/]

16:35:46 <dc_rdfig> Replaced comment D2.

16:36:04 <bengee> E: [http://rdfweb.org/pipermail/rdfweb-dev/2004-September/013703.html|Dan's reply]

16:36:05 <dc_rdfig> Added comment E2.

16:36:22 <danbri> I: [http://www.ilrt.bris.ac.uk/discovery/chatlogs/foaf/2004-09-15.html#T16-48-26|danbri update FOAF spec to add a main section on naming], to provide a unified treatment of first/last/given/family addressing Misha's [http://rdfweb.org/pipermail/rdfweb-dev/2003-August/011756.html|concern] re first/last

16:36:22 <dc_rdfig> Added comment I1.

16:36:32 <esigler> D3:attending: [eric sigler|http://esigler.2nw.net]

16:36:32 <dc_rdfig> Comment D3 not found.

16:36:36 <Talliesin> Hmm. DOB vs. Birthday.

16:36:45 <danbri> yes, pls do intros while I add some agenda bits

16:37:04 <esigler> :D:attending: [eric sigler|http://esigler.2nw.net]

16:37:09 <danbri> I:[http://www.ilrt.bris.ac.uk/discovery/chatlogs/foaf/2004-09-15.html#T16-55-28|action: crschmidt to circulate crawler stats on foaf:nick foaf:name foaf:firstName foaf:givenname foaf:surname foaf:family_name]

16:37:09 <dc_rdfig> Added comment I2.

16:37:12 <balbinus> esigler: just use D:blah

16:37:15 <Talliesin> If a date was a resource we could have an anniversary property. Then you could have a birthday expressed as the anniversary of the DOB without necessarily giving away the year.

16:37:18 <libby> so we should be able to see the agenda ok on rdfig.xmlhack.com

16:37:33 <esigler> D:attending: [eric sigler|http://esigler.2nw.net]

16:37:34 <dc_rdfig> Added comment D3.

16:37:59 <danbri> D:attending: [Dan Brickley|http://danbri.org/], W3C.

16:38:00 <dc_rdfig> Added comment D4.

16:38:16 <teefal> D:attending: [timothy falconer|http://bigfractaltangle.com]

16:38:16 <dc_rdfig> Added comment D5.

16:38:24 <danbri> So the chump c/o http://rdfig.xmlhack.com/ has the agenda, in reverse order...

16:38:27 <JibberJim> My view is I'm willing to give my age, but I'm not keen on giving the date, so I'd like that to be possible too

16:38:35 <bengee> D:attending: [benjamin nowack|http://www.appmosphere.com/]

16:38:35 <dc_rdfig> Added comment D6.

16:38:59 <danbri> <danbri> agenda+ FoafCommunityProcess, FOAF and standards, and relationship to W3C

16:39:03 <danbri> libby, could you add that?

16:39:18 <danbri> rdfig:view?

16:39:26 <danbri> dc_rdfig:view?

16:39:26 <dc_rdfig> E: FOAF meeting agenda item 1 (E) Stability Vocab (blurb)

16:39:27 <dc_rdfig> F: FOAF meeting agenda item 2 (F) Web site update (blurb)

16:39:28 <dc_rdfig> G: FOAF meteing agenda item 3 (G) Foaf community process (blurb)

16:39:29 <dc_rdfig> H: FOAF meeting agenda item 4 (H): date of birth (blurb)

16:39:30 <dc_rdfig> I: Previous actions (blurb)

16:39:32 <libby> could it not go with 3 (G)?

16:39:36 <Talliesin> Oh, and I'm not attending BTW. I have to go soon :(

16:39:40 <libby> boo

16:39:48 <danbri> ok yeah, let's fold it in there, sorry.

16:39:57 <libby> G: including FoafCommunityProcess, FOAF and standards, and relationship to W3C

16:39:58 <dc_rdfig> Added comment G2.

16:40:34 <libby> G:|FOAF meeting agenda item 3 (G) Foaf community process

16:40:34 <dc_rdfig> Titled item G.

16:40:45 <danbri> OK hi everyone, let's do brief action reviews before beginning w/ the items in the chump...

16:41:13 <danbri> see http://rdfig.xmlhack.com/2004/09/22/2004-09-22.html#1095870914.190001 for actions

16:41:32 <danbri> 1st: danbri update FOAF spec to add a main section on naming, to provide a unified treatment of first/last/given/family addressing Misha's concern re first/last

16:41:43 <balbinus> should i indicate me as attending?

16:41:47 <danbri> That needs continuing...

16:41:59 <danbri> yes, please, if you'd like to be so listed! (you're very welcome, of course...)

16:42:06 <balbinus> D:attending: [Vincent Tabard|http://www.bonjourlesmouettes.org]

16:42:06 <dc_rdfig> Added comment D7.

16:42:07 <danbri> I:

16:42:07 <dc_rdfig> blurb

16:42:08 <dc_rdfig> Previous actions

16:42:08 <balbinus> ok :)

16:42:09 <dc_rdfig> (1:danbri) [http://www.ilrt.bris.ac.uk/discovery/chatlogs/foaf/2004-09-15.html#T16-48-26|danbri update FOAF spec to add a main section on naming], to provide a unified treatment of first/last/given/family addressing Misha's [http://rdfweb.org/pipermail/rdfweb-dev/2003-August/011756.html|concern] re first/last

16:42:10 <dc_rdfig> (2:danbri) [http://www.ilrt.bris.ac.uk/discovery/chatlogs/foaf/2004-09-15.html#T16-55-28|action: crschmidt to circulate crawler stats on foaf:nick foaf:name foaf:firstName foaf:givenname foaf:surname foaf:family_name]

16:42:17 <danbri> I:danbri's action is CONTINUED

16:42:17 <dc_rdfig> Added comment I3.

16:42:28 <danbri> Chris's action, on the other hand, was completed before the meeting was even finished!

16:42:47 <danbri> chris, do you want to add anything to what you shared last week?

16:42:52 <danbri> re crawler stats?

16:43:14 <danbri> ah there was also some stuff from swoogle via bengee

16:43:23 <balbinus> hi edd... I'm gonna send some translations from DOAP-fr to the list, btw :)

16:43:45 <libby> yeah the swoogle stuff is cool

16:43:48 <danbri> I:Do the figures from Chris and Bengee [http://www.ilrt.bris.ac.uk/discovery/chatlogs/foaf/2004-09-15.html#T17-02-07|provided last week] close the crawler-stats action?

16:43:48 <dc_rdfig> Added comment I4.

16:43:57 <danbri> hi edd

16:44:05 <edd> hi. just here for the lurking.

16:44:39 * danbri suspects chris sidetracked by having dayjob

16:44:46 <danbri> did anyone else look at these? should we continue it?

16:45:07 <libby> chris is at a lunch meeting

16:45:59 <libby> my stats are probably not v useful at present

16:46:04 <danbri> I think we could probably get a little more on this, so will continue unless chris pops up later.

16:46:17 <danbri> I:Chris's action CONTINUED also, though noting data shared in IRC last week.

16:46:17 <dc_rdfig> Added comment I5.

16:46:37 <danbri> -dc_rdfig/#rdfig- E: FOAF meeting agenda item 1 (E) Stability Vocab (blurb)

16:46:39 <bengee> what was the stat's intention? to see if we can "discourage" some of them?

16:47:02 <libby> to see how bad the problem was?

16:47:12 <bengee> ah, ok

16:47:18 <danbri> bengee, we used (in early days) to be in habit of using stats to guide our understanding of where foaf was. This would be particularly welcome now w.r.t. naming vocab.

16:47:33 <danbri> re Stability Vocab, would you lik to introduce the issue, bengee?

16:47:45 <bengee> can do.

16:48:06 <danbri> context from bengee: http://rdfweb.org/pipermail/rdfweb-dev/2004-September/013699.html

16:48:26 <CaptSolo> D:attending: [CaptSolo - Uldis Bojars|http://purl.org/captsolo/blog]

16:48:26 <dc_rdfig> Added comment D8.

16:48:32 <danbri> I replied, though didn't go into detail, http://rdfweb.org/pipermail/rdfweb-dev/2004-September/013703.html

16:48:42 <bengee> I was just thinking if it would make sense to get a term's liufe cycle state directly by looking at its term_status value

16:48:54 <bengee> s/liufe/life

16:49:30 <bengee> to see if a term is "in active development" or just "kept" (i.e. not modified for ages)

16:49:33 <danbri> do you think all that can be squeezed into term_status?

16:50:03 <danbri> As i mentioned in mail, I talked a bit w/ Tom Baker from Dublin Core last week (Libby too)

16:50:14 <danbri> ...they have a usage board, and a bit more structure w.r.t. decision records.

16:50:16 <bengee> at the end of writing I realized that we could use a modified annotation for it

16:50:36 <bengee> od maybe prose would suffice, too

16:50:43 <danbri> ...we're hoping we might be able to converge on some common conventions for linking descriptions of terms in a vocab to the decision record / process state that surrounds it.

16:50:45 <bengee> "or" even

16:51:05 <danbri> modified annotation? can you give an example?

16:51:36 <danbri>http://www.jiscmail.ac.uk/cgi-bin/webadmin?A2=ind0409&L=dc-general&T=0&F=&S=&P=895

16:51:37 <dc_rdfig> J: http://www.jiscmail.ac.uk/cgi-bin/webadmin?A2=ind0409&L=dc-general&T=0&F=&S=&P=895 from danbri

16:51:38 <bengee> isn't there something kind of a dct:modified

16:51:56 <danbri> J:|Dublin Core Usage Board announces new term, "dc:provenance"

16:51:56 <dc_rdfig> Titled item J.

16:52:15 <libby> lastmodified makes sense as an interesting piece of information, but I'm not sure if it captues the semantics you want bengee

16:52:18 <danbri> J:Of intrinsic interest, but also re compare/contrast DC and FOAF process (and lack thereof in latter case).

16:52:19 <dc_rdfig> Added comment J1.

16:52:26 <bengee> so you could see "term_status=testing, modified=1993-01-01"

16:52:34 <danbri> Yep, I think 'last modified' is generally used in context of a document

16:53:08 <bengee> as a tool developer I'd know that it's probably not going to change too soon

16:53:10 <libby> you want: "deprecated but not removed" or "deprecated but not deployed" or similar

16:53:11 <danbri> With things like foaf:name, the _section_ of the foaf spec might be considered a document, though diff'ing, sha1'ing etc could be tricky; but there's also sections of the RDF schema too.

16:53:36 <danbri> Ah, I see, you want some element of 'crystal ball' to look into future and know roughly what's on the roadmap?

16:53:44 <bengee> more a "not deprecated, but not remover either"

16:54:00 <libby> "discouraged"?

16:54:02 <bengee> yeah, more like that

16:54:09 <Zenethian> by definition, isn't that what deprecated means? That it still exists, but not recommended, and may be removed later?

16:54:12 <danbri> (which is, btw, one of the reasons I've heard for why Organizations support W3C, to be able to know what's coming, and --- roughly! --- when)

16:54:30 <danbri> I think there are two notions of depractation around the foaf naming stuff

16:54:33 <libby> can you give an example bengee? I'd think that's deprecated too really

16:54:47 <libby> maybe we need to clarify what 'deprecated' shoudl mean in this context

16:54:50 <danbri> (i) things like 'last name' are pretty cheesy, and we encourage people to have nicer data, which uses 'family name'

16:55:05 <danbri> ...so we might discourage / depracate that sort of data

16:55:23 <danbri> (ii) we think we have a property that is in-itself somehow wrong/broken/bad, and may change or be retracted

16:55:33 <Zenethian>http://www.sebistar.net/dict.py/lookup?word=deprecated

16:55:34 <dc_rdfig> K: http://www.sebistar.net/dict.py/lookup?word=deprecated from Zenethian

16:55:38 <danbri> ...so we might discourage / depracate use of that specific property

16:56:01 <danbri> In case of (i) the property might stay there forever. But be mostly useful for as-is representation of crap/legacy data.

16:56:20 <danbri> Anyone v familar w/ the OWL vocabs for this stuff?

16:56:28 <bengee> maybe I'm mixing two things here: 1)see what a term's previous life cycle stage was 2)indicate discouraged (I think that's the term I meant) terms that are still going to be kept in the spec

16:56:42 * danbri glances at clock; hmm 30 mins left...

16:56:53 <danbri> can we make a specific proposal out of those two, bengee?

16:57:18 <teefal> hmm... a "magic eight ball" property .... signs point to yes

16:57:19 <bengee> 1) I guess a "modified" would be fine 2)prose ?

16:57:26 <danbri> eg. "proposed: that FOAF namespace should provide machine-readable annotation indicating place in vocab lifecycle"

16:58:31 <danbri> ..."proposed: that FOAF namespace should distinguish terms which may be removed from spec, from terms whose usage is discouraged because there are better idioms to use, and which will stay in namespace indefinitely"

16:58:49 <danbri> bengee, do either of those capture your thinking, or am i putting words in your mouth too much?

16:58:54 <libby> that makes sense to me

16:59:15 <libby> needs a bit of work seeing what's out there, how DC uses deprecated etc

16:59:25 <teefal> "chances of eventual adoption today for "foaf:collegeBeerDrinkingNickName ... 38%" :)

16:59:36 <libby> heh

16:59:44 <bengee> re 2) seems to be between "not modified for a long time, time will tell" (i.e. not really expressible for machines) and "deprecated"

16:59:50 <danbri> I agree (if I understand you right, bengee) that 'modified' in some broader-than-document-changes sense is relevant. Not sure dct:modified is exactly it.

16:59:51 <bengee> danbri, makes sense to mee, too

17:00:06 <bengee> let's move on..

17:00:08 <teefal> is there a way to point to better usage .... lastName -> givenName

17:00:13 <teefal> (oops... familyName)

17:00:28 <danbri> bengee, wanna make one or both those proposals? (by typing them without the quotes :) and we can call for objections and maybe resolve that those-present agreed-or-stayed-silent?

17:00:40 * danbri reads up on next agena item meanwhile

17:00:42 <libby> (on mailing list?)

17:01:01 <danbri> here's fine for now. can take it to list too. it's not a binding-in-stone thing just a record...

17:01:38 <danbri> dc_rdfig:view?

17:01:38 <dc_rdfig> G: FOAF meeting agenda item 3 (G) Foaf community process (blurb)

17:01:38 <balbinus> K:|Definitions of "deprecated"

17:01:39 <dc_rdfig> H: FOAF meeting agenda item 4 (H): date of birth (blurb)

17:01:40 <dc_rdfig> I: Previous actions (blurb)

17:01:41 <dc_rdfig> J: Dublin Core Usage Board announces new term, "dc:provenance" (http://www.jiscmail.ac.uk/cgi-bin/webadmin?A2=ind0409&L=dc-general&T=0&F=&S=&P=895)

17:01:42 <dc_rdfig> K: http://www.sebistar.net/dict.py/lookup?word=deprecated

17:01:44 <dc_rdfig> Titled item K.

17:02:30 <danbri> F:

17:02:30 <dc_rdfig> blurb

17:02:31 <dc_rdfig> FOAF meeting agenda item 2 (F) Web site update

17:02:32 <dc_rdfig> (1:libby) including jim's concerns re 'flyshit'

17:02:41 <libby> jim had to go :/

17:02:48 <danbri> did we lose bengee too?

17:02:54 * bengee back from phone

17:02:56 <bengee> ah

17:03:02 <bengee> sorry

17:03:18 <bengee> shall I send a proposal to the list?

17:03:28 <golbeck> E:

17:03:28 <dc_rdfig> blurb

17:03:29 <dc_rdfig> FOAF meeting agenda item 1 (E) Stability Vocab

17:03:30 <dc_rdfig> (1:bengee) See [http://rdfweb.org/pipermail/rdfweb-dev/2004-September/013699.html] mail to rdfweb-dev

17:03:31 <dc_rdfig> (2:bengee) [http://rdfweb.org/pipermail/rdfweb-dev/2004-September/013703.html|Dan's reply]

17:03:32 <libby> that's be great bengee

17:03:45 <libby> here's what jim initially said: http://www.ilrt.bris.ac.uk/discovery/chatlogs/foaf/2004-09-19.html#T18-54-10

17:03:50 <danbri> F:See also [http://rdfweb.org/pipermail/foafsite-webmasters/2004-September/000021.html|nicole's todo list], and [http://beta.foaf-project.org/|draft-in-progress] for a new site that danbri and nicole have been working on. Mostly nicole really.

17:03:50 <dc_rdfig> Added comment F2.

17:04:08 <libby> [[

17:04:08 <danbri> F:...and nicole's not here, so let's do this one quite briefly

17:04:08 <libby> 18:54:10 <JibberJim> hmm the foaf-project site is full of flyshit... 0.78em is too small!!!

17:04:09 <dc_rdfig> Added comment F3.

17:04:09 <libby> 18:54:43 <JibberJim> oh no it's 0.7em that really is too small.

17:04:11 <libby> ]

17:04:26 <danbri> OK, thanks Bengee that'd be great

17:04:29 <golbeck> could i suggest a sectino of the new site that contains stable extensions to FOAF?

17:04:57 <danbri> action: bengee propose some clarifications re lifecycle/stability vocab to list, goal of having better machine-readable status for FOAF namespace

17:05:01 <danbri> F:

17:05:01 <dc_rdfig> blurb

17:05:02 <dc_rdfig> FOAF meeting agenda item 2 (F) Web site update

17:05:03 <dc_rdfig> (1:libby) including jim's concerns re 'flyshit'

17:05:04 <dc_rdfig> (2:danbri) See also [http://rdfweb.org/pipermail/foafsite-webmasters/2004-September/000021.html|nicole's todo list], and [http://beta.foaf-project.org/|draft-in-progress] for a new site that danbri and nicole have been working on. Mostly nicole really.

17:05:05 <danbri> moving on...

17:05:05 <dc_rdfig> (3:danbri) ...and nicole's not here, so let's do this one quite briefly

17:05:14 * danbri feels bad, didn't ping nicole to come here

17:05:18 <libby> D:ACTION bengee propose some clarifications re lifecycle/stability vocab to list, goal of having better machine-readable status for FOAF namespace

17:05:18 <dc_rdfig> Added comment D9.

17:05:29 <danbri> OK just a few words on this I think.

17:05:32 <libby> move to next time maybe?

17:05:44 * libby thinks looks very nice

17:05:48 <danbri> Nicole (Sullivan) has been working on a new site. We're hacking on it as beta.foaf-project.org

17:06:19 <danbri> Do please make sure any feedback goes to nicole@postdiluvian.org (cc:'d to foafsite-webmasters@vapours.rdfweb.org) as she doesn't track the zillion other fora we might all hang out in.

17:06:32 <danbri> Jim, could you drop her a (friendly:) note re the fontsize concern?

17:06:45 <libby> jim's gone

17:06:48 <danbri> ...and let's revisit this at next meeting, when I'll invite nicole along

17:06:50 <danbri> OK. I'll do it

17:06:51 <libby> think we can action him anyway?

17:07:12 <danbri> F:action: danbri let nicole know that Jim had problems with fontsize on new site

17:07:12 <dc_rdfig> Added comment F4.

17:07:19 <libby> he wasn;t being rude; apparantly 'flyshit' is common parlance in css

17:07:41 <danbri> (re process, such that it is, I don't like actioning absentees, as its not a record of what people have agreed to do)

17:07:57 * danbri nods, but notes that we needn't stoop to the level of those guttermouths in CSS community ;)

17:08:03 <libby> D9:=""

17:08:03 <dc_rdfig> Replaced comment D9.

17:08:04 <danbri> OK next?

17:08:06 <libby> D9:""

17:08:07 <dc_rdfig> Deleted comment D9.

17:08:18 <danbri> what was wrong w/ D9?

17:08:22 <danbri> G:

17:08:23 <dc_rdfig> blurb

17:08:24 <dc_rdfig> FOAF meeting agenda item 3 (G) Foaf community process

17:08:25 <dc_rdfig> (1:libby) see [http://www.saint-andre.com/blog/2004-09.html#2004-09-12T19:56|Peter St. Andre's post]

17:08:25 <libby> E:ACTION bengee propose some clarifications re lifecycle/stability vocab to list, goal of having better machine-readable status for FOAF namespace

17:08:26 <dc_rdfig> (2:libby) including FoafCommunityProcess, FOAF and standards, and relationship to W3C

17:08:28 <dc_rdfig> Added comment E3.

17:08:31 <libby> shoulda been in E

17:08:49 <libby> peter not here either :(

17:08:53 <danbri> OK, a few things here. Esigler wanted to remind us of Peter's point.

17:09:37 <danbri> Peter is from the Jabber world; Jabber is something that started out as a standalone project, acquired steering committees etc and industrial backing in various ways, and ended up spawning a standards-track working group (XMPP?) within the IETF.

17:10:00 <danbri> He's right to ask about process. Various conversations have been happening on that topic since FOAFCamp and FOAF Galway

17:10:01 <deltab> yep, XMPP

17:10:34 <danbri> The wiki entry on http://rdfweb.org/topic/FOAFCommunityProcess is the most written-down thing we have.

17:10:58 <danbri> These IRC meetings are a start in direction of more structure too, though I'm rusty and decent pre-announcements have been lacking.

17:11:26 <danbri> Since we're squeeze for time, I'd just like to note a point re W3C, and then invite discussion for max 5mins.

17:11:32 <danbri> re W3C...

17:12:00 * pjz mumbles about rough consensus and working code.

17:12:09 <danbri> FOAF's bizarre and quirky rise hasn't gone un-noticed within W3C, and there have been a few discussions we've had within the W3C team re how it relates to W3C's formal standards-track efforts.

17:13:00 <danbri> We discussed some possibilities in a team meeting on monday. I sketched a couple options. (i) that FOAF itself, the vocab/namespace could go REC-track within W3C, with a public-access Working Group (similar to that we offered the Atom community).

17:13:07 * danbri nods re code/rough/etc

17:13:17 <danbri> ...I mentioned that nobody I knew was currently enthusing about (i)

17:14:04 <danbri> (ii) a looser thing, where we take better steps to hook FOAF experience into W3C, perhaps via a taskforce of the SW Best Practices WG (which Libby, I and othes are on) andthe SW Interest GRoup (which I chair, and whose IRC channel we're using right now).

17:14:45 <danbri> ...latter model would also perhaps include creation of a stark, stable and relatively dull 'core' namespace, maybe at w3.org URL, that covered the core concepts of email, name, homepage etc.

17:14:59 * teefal wants a #swig channnel ... pronouncable name, with pun potential

17:15:15 <danbri> ...and also spawn collaborations on important namespaces that are beyond FOAF's core competency (eg. gettiing a personal names vocab, and a postal addresses one).

17:15:25 <dajobe> [I had one once and got questions about www.swig.org]

17:15:53 <danbri> One point was made: if we don't take steps to make FOAF a little bit more standards-friendly, it'll end up being re-invented somewhere (maybe the P3P WG, who already have similar concepts in non-RDF XML).

17:16:06 <danbri> OK I'm getting RSI. Comments from others? :)

17:16:11 * golbeck is much in favor of core + extensions

17:16:21 <golbeck> assuming extensions get reasonable treatment in the system

17:16:29 <pjz> I agree re: it getting reinvented elsewehre if w3c/SW folk don't do something with it

17:17:05 <esigler> My tentative suggestion might be a document (maybe hung off that wiki entry) that says "here are the steps to go about suggesting an item for inclusion in the vocab".

17:17:10 <pjz> the guys at Technorati were talking about an XML schema 'Attention.xml' that was essentially the MeNow schema.

17:17:12 <esigler> And perhaps the vetting / structure for a proposal (like adding vocabulary stability data for example)) would probably help out a bit (at least short-term / psa's case).

17:17:13 * danbri notes also some discussion of FOAF informational RFC possibilities for IETF groups to be able to draw on FOAF; also Tom Baker suggested Dublin Core initiative might be a possible umbrella for lightweight standards work that's open to public/individual participation

17:17:17 <esigler> Not really adding any "process", but a capture of what's usually been done so far.

17:17:36 <danbri> pjz, is there a pointer to Attention.xml doc?

17:17:56 <pjz> will dig up via google, gimme a sec

17:18:01 <golbeck> one of the problems with vocabularies is that if someone doesn't like a small bit of yours, they go write their own. this is good in some sense, but if we want to keep coherent extensions to foaf in a community of vocabularies, i fear bickering will lead to factions

17:18:21 <danbri> esigler, yep. At the moment we're on the genial-dictator model, but with ability to do extension namespaces as a safety valve. But that's not even really explained in the spec. Which might cause frustration/confusion.

17:18:22 <golbeck> i.e. not the collaboration we see on the core

17:18:29 <pjz> #foaf invaders borrowing channel for a semi-scheduled chat, 17.30BST for 1hr /// Semantic Web / RDF chat n hack. blog http://rdfig.xmlha

17:18:32 <pjz> status for FOAF namespace

17:18:35 <pjz> 12:08 -dc_rdfig:#rdfig- (2:libby) including FoafCommunityProcess, FOAF and standards, and relationship to W3C

17:18:38 <pjz> 12:08 -dc_rdfig:#rdfig- Added comment E3.

17:18:39 <danbri> yep, I agree golbeck

17:18:40 <pjz> 12:08 < libby> shoulda been in E

17:18:43 <pjz> 12:08 < libby> peter not here either :(

17:18:46 <pjz> 12:08 <@danbri> OK, a few things here. Esigler wanted to remind us of Peter's point.

17:18:49 <pjz> 12:09 <@danbri> Peter is from the Jabber world; Jabber is something that started out as a standalone project, acquired steering

17:18:52 <pjz> committees etc and industrial backing in various ways, and ended up spawning a standards-track working group (XMPP?)

17:18:55 <pjz> within the IETF.

17:18:57 <pjz> 12:10 <@danbri> He's right to ask about process. Various conversations have been happening on that topic since FOAFCamp and FOAF Galway

17:19:00 <pjz> 12:10 < deltab> yep, XMPP

17:19:02 <danbri> pjz, copy/paste explosion?

17:19:03 <pjz> 12:10 <@danbri> The wiki entry on http://rdfweb.org/topic/FOAFCommunityProcess is the most written-down thing we have.

17:19:06 <pjz> 12:10 <@danbri> These IRC meetings are a start in direction of more structure too, though I'm rusty and decent pre-announcements have

17:19:07 <golbeck> ha

17:19:09 <pjz> been lacking.

17:19:11 <pjz> 12:11 <@danbri> Since we're squeeze for time, I'd just like to note a point re W3C, and then invite discussion for max 5mins.

17:19:15 <pjz> 12:11 <@danbri> re W3C...

17:19:17 <pjz> arrgh

17:19:17 <pjz> on a 'puter with no middle button. hate hate hate.

17:19:20 <pjz> (sorry all)

17:19:23 <pjz>http://developers.technorati.com/wiki/attentionxml

17:19:24 <dc_rdfig> L: http://developers.technorati.com/wiki/attentionxml from pjz

17:19:25 <pjz> yup

17:19:30 <danbri> s'ok, done it myself enough times!

17:20:02 <danbri> any more thoughts on standards-track vs standalone project?

17:20:34 <esigler> As a suggestion for next week's meeting, should someone drag (er, invite) Peter St. Andre here?

17:21:04 <pjz> I think putting something out there that we have rough consensus on is better than not vetting anything

17:21:07 <danbri> Here's a proposal... that it should be a goal to ultimately get the FOAF spec in a shape (whatever shape that is) such that IETF, Dublin Core and W3C Working Groups are comfortable citing it when they need to draw upon core concepts like Person, homepage, mbox, mbox_sha1sum etc.

17:21:13 <libby> sounds like a job for danbri (re dragging :)

17:21:17 <teefal> from industry side, w3c connection would be helpful

17:21:26 <libby> +1 danbri

17:21:29 <teefal> current conversations go .... danbri, who works at w3c

17:21:38 <danbri> Yes, I'd like to get him here, to share Jabber experience at least

17:22:06 * teefal will implement new jabber/foaf stuff very quickly (we're using both)

17:22:09 <danbri> E:action danbri: invite Peter St Andre to a future IRC FOAF meeting.

17:22:09 <dc_rdfig> Added comment E4.

17:22:15 * bengee is working on that foaf dictionary thingy which could help us offer "subsets", so that we could offer "profiles" that a W3C could be more happy to cite (e.g. without geekCode etc.)

17:22:17 <danbri> teefal, that'd be great...

17:22:26 <libby> E4:""

17:22:26 <dc_rdfig> Deleted comment E4.

17:22:35 <danbri> OK we're running out of time fast. Would anyone have a problem with extending for 1/2 hour?

17:22:37 <bengee> similar to what foafnet is trying to establish

17:22:46 <danbri> we've taken over the channel now, anyways ;)

17:22:52 <libby> G:ACTION danbri: invite Peter St Andre to a future IRC FOAF meeting.

17:22:52 <dc_rdfig> Added comment G3.

17:23:04 <danbri> ta

17:23:20 <danbri> OK let's just keep going till we get bored. another 1/2 hour say.

17:23:34 <danbri> proposed: that it should be a goal to ultimately get the FOAF spec in a shape (whatever shape that is) such that IETF, Dublin Core and W3C Working Groups are comfortable citing it when they need to draw upon core concepts like Person, homepage, mbox, mbox_sha1sum etc.

17:23:41 <pjz> seconded

17:23:54 <danbri> I'll call for objections here (anyone object? :) and if not, will propose it to the list as well...

17:24:01 * danbri counts to 20

17:24:19 <esigler> but...what's the shape going to be?

17:24:22 * danbri wonders when chris schmidt gets back...

17:24:26 <danbri> shape?

17:24:41 <danbri> ah, to be determined thru dialog w/ the relevant groups.

17:24:43 <esigler> "FOAF spec in a shape (whatever shape that is)" - what form will that take?

17:24:44 <libby> G:ACTION danbri propose to the list "that it should be a goal to ultimately get the FOAF spec in a shape (whatever shape that is) such that IETF, Dublin Core and W3C Working Groups are comfortable citing it when they need to draw upon core concepts like Person, homepage, mbox, mbox_sha1sum etc."

17:24:44 <dc_rdfig> Added comment G4.

17:24:53 <danbri> note that W3C might need to hange its view of externally run namespaces

17:25:11 <libby> chris might not make it back for awhile, luch with his boss

17:25:15 <danbri> ...or it might decide to create its own, eg a pimcore namespace for PIM-related stuff, but subclass the stable terms from FOAF.

17:25:29 <pjz> would it make them happier if we marked bits of the core namespace 'optional'?

17:25:45 <danbri> them=us, to a certain extent, also...

17:25:56 <danbri> the W3C semweb community substantially hangs out here in #rdfig

17:26:05 <pjz> ie. anything non-stable = 'MAY implement'

17:26:08 <CaptSolo> pjz - that'd allow to put more into a core vocab?

17:26:10 <danbri> ...I think optionality is something that comes as standard in RDF vocabularies

17:26:21 <danbri> ...no RDF namespace can mandate what you say in a given RDF instance

17:26:23 <pjz> danbri: good point

17:26:34 <libby> more fruitful might be to publish application profile things for specific apps

17:26:35 <danbri> but yes, I think there would be a concern to distinguish 'fluff' from 'core', even if in same namespace

17:26:52 * danbri would love to see foaf:dnaChecksum in a W3C RECommendation of course ;)

17:27:00 <libby> i.e. you don;t want to just use foaf but also geo stuff, image stuff, calendar stuff etc etc

17:27:00 <CaptSolo> :)

17:27:27 <libby> ...but for an app you do need to say what sort of stuff you can handle

17:27:33 <danbri> Yes, one point I made in the in-house W3C discussion was that FOAF is rich and fun cos all sorts of people have made up their own extensions

17:27:42 <danbri> ...and that we want to be able to trust those too

17:27:42 * libby favouite hobbyhorse

17:27:45 <CaptSolo> libby: support that

17:27:52 <CaptSolo> usage profiles...

17:27:58 <libby> yeah

17:28:03 <danbri> ...and we can't bless them all formally thru W3C, but we can work on improving process/documentation that surrounds all these

17:28:10 <danbri> see also talk(s) from Galway conf

17:28:16 <pjz> yeah, my toy app has a list of schemas that it knows about

17:28:18 <danbri> aw crap, I didn't get my galway materials online yet

17:28:29 <pjz> may end up with a separate 'setup' page for each one

17:28:39 <libby>http://www.w3.org/2001/sw/Europe/events/foaf-galway/papers/pp/validating_rdf/

17:28:39 <dc_rdfig> M: http://www.w3.org/2001/sw/Europe/events/foaf-galway/papers/pp/validating_rdf/ from libby

17:29:00 <danbri> G:action danbri: get his FOAF talks from FOAFCamp and Galway online (and do the workshop report!)

17:29:00 <dc_rdfig> Added comment G5.

17:29:06 <libby> M:|possible 'validation' approach for RDF application profiles/ usage profiles

17:29:07 <dc_rdfig> Titled item M.

17:29:32 <teefal> we've been talking about "views" (same as profiles) ... subsets of foaf properties that have ACLs applied to them (came from talk with smarr/plaxo & danbri at king's head pub)

17:29:41 <danbri> W3C folk also well aware that FOAF is a success 'cos lots of grassroots participation from people who wouldn't normally be in a position to join a W3C Working Group, and there's willingness to work around that.

17:30:15 <danbri> ...after all, W3C spent a huge amount of effort making an opensource-friendly, Royalty Free patent policy, so obviously welcomes opensource-style input into stds track.

17:30:19 <libby> L:|Attention.XML wiki

17:30:20 * danbri nods re views

17:30:24 <dc_rdfig> Titled item L.

17:30:38 * bengee remembers his relatd(?) "documentation types" (e.g use cases, ..) question on rdfweb-dev, taht could allow generating profiles

17:30:39 <danbri> anything else on this, or shall we talk about date-of-birth etc? (noting Chris ain't here)

17:30:44 <pjz> teefal: I'm interested in any info you have on 'view's or 'profiles'

17:30:51 <DanC> oh yeah... foaf ScheduledTopicChat... I'm surprised to see it show up in http://rdfig.xmlhack.com/ ; I was expecting it in #foaf.

17:30:53 <pjz> teefal: I looked at PEDAL and it seems a bit heavyweight

17:31:04 <libby> yeah, sorry danc; we wanted to use the chump for the agenda

17:31:09 <danbri> DanC, it was nearly left too late, so we stole this channel to use chump

17:31:11 <DanC> no need to apologize

17:31:23 * danbri really meant to do agenda in advance this time :/

17:31:53 <danbri> So we're over the original time. Let's maybe have a few words on date of birth, then adjourn and chitchat can resume (incl. on the above topics).

17:32:02 <danbri> Anybody want to introduce the data of birth situation?

17:32:06 <danbri> As I recall it...:

17:32:22 <danbri> - a few old foaf files, like mine, have a plain literal field called dateOfBirth

17:32:27 <CaptSolo> pjz: views or profiles could probably be made another action/topic for the regular irc meetings?

17:32:28 <danbri> that takes things like 1972-01-09

17:32:39 <CaptSolo> i would like to get more info on that [as well]

17:32:54 <danbri> ...but it never got in spec, cos we also wanted to be able to say 'my birthday is jan 9th', without giving away age

17:32:57 <libby> - DOB'd be cool because then we could do birthdays

17:32:59 <libby> yeah

17:33:11 <danbri> useful for orkut-style birthday reminders, and a bit less revealing (some security issues re identity theft potential)

17:33:12 <libby> a few proposals were made. and it's in bio (via event)

17:33:26 * DanC attempts to sync with this meeting...

17:33:44 <danbri> to my mind, the core problem is that both fields have similar name, so its an aesthetic question, what would a good pair of names be, and then there's usual issues around date syntax.

17:33:46 * DanC had it starting an hour ago in his calendar; zat right?

17:33:56 <danbri> yes, right

17:33:56 <libby> yep danc

17:34:06 <libby> we've extended by 30mins

17:34:10 * teefal seconds profiles/views as future topic

17:34:11 <danbri> libby/danc, can you comment from an rdf/ical perspective?

17:34:15 <danbri> +1 re profiles

17:34:26 <danbri> can someone chump that appropriately?

17:34:33 <pjz> danbri: I disagree and think there should be just one field with different values depending on your view

17:34:39 <libby> D:future topics, website, profiles/views

17:34:39 <dc_rdfig> Added comment D9.

17:34:41 * DanC is in pronoun overload; isn't sync'd up sufficiently to comment on anything

17:35:08 <danbri> pjz, so you're happy mixing "1972-01-09" with "01-09" ?

17:35:22 <chaalsNCE> pjz if you mean that you have one birthday thing, that seems fine to me.

17:35:23 <libby> pretty icky

17:35:34 <chaalsNCE> you can include the year you were born, or not.

17:35:38 <pjz> danbri: sure, it's all text anyway.

17:35:45 <libby> giving it tons of internal structure not that useful either

17:35:50 <teefal> re .. views... it's simliiar to the issue of not wanting everyone to see my homePhone

17:35:50 <DanC> ew... better to have separate properties for "1972-01-09" and "01-09"

17:35:53 * danbri would like to be able to put the former in databases that can be told things like 'foaf:dateOfBirth takes iso-8601 dates in plain string form', and answer qs like 'is this person over 18?'

17:36:07 <teefal> i might want some to see my birth *day of year*, but not my birth year

17:36:08 <kpreid> foaf:dateOfBirth & foaf:birthday ?

17:36:12 <danbri> +1 on separate properties; 'cos they're easier to treat as data, then.

17:36:18 * bengee suggested separate dayOfBirth, monthOfBirth, yearOfBirth

17:36:19 <chaalsNCE> I would suggest not aking it all text. either make it some kind of date ( 01/09 is notoriously useless)

17:36:21 <libby> that could work kpreid

17:36:25 <teefal> it's the same data, just different access to it

17:36:27 <danbri> <kpreid> foaf:dateOfBirth & foaf:birthday ?

17:36:29 <DanC> +1 foaf:dateOfBirth & foaf:birthday

17:36:31 <danbri> best proposal i've seen so far

17:36:32 <teefal> this is probably implementation specific

17:36:41 <libby> yeah, +1 foaf:dateOfBirth & foaf:birthday

17:36:51 <libby> and there shoudl ba a warning on DOB

17:36:54 <danbri> pjz, could you live with foaf:dateOfBirth & foaf:birthday ?

17:36:56 <pjz> sure

17:36:57 <bengee> so you could easily find birthdays this month etc

17:36:59 <danbri> thanks

17:37:10 <danbri> warning, re 'security alert! privacy concern!' etc?

17:37:12 <chaalsNCE> hang on, are you going to say someting about what it should look like?

17:37:15 <libby> yeah danbri

17:37:18 * DanC mulls over 3 props for month, day, year... doesn't feel good... not sure why...

17:37:27 <chaalsNCE> ie 03-04-05 means what?

17:37:28 <libby> W3C's datetime profile?

17:37:30 <danbri> I fear, that if we do that, we'll create an expectation that we've done that analysis for all of them?

17:37:39 <danbri> dateOfBirth can be W3C's datetime profile.

17:37:50 <chaalsNCE> danbri, that == ??

17:37:52 <libby> I think birthday might work like that too

17:37:52 <danbri> Is there a defined syntax for day/month elsewhere?

17:37:59 <danbri> .g w3c datetime note charles misha

17:38:01 <DanC> s/W3C's datetime profile/XML Schema date lexical form/

17:38:02 <phenny> w3c datetime note charles misha: http://www.w3.org/TR/NOTE-datetime

17:38:10 <pjz> if you want it to be an iso-8601 date though, can't you say something like <rdf:type>iso-8601</rdf:type> inside <foaf:dateOfBirth> ?

17:38:15 <danbri> Ah OK that was pre-xmlschema

17:38:21 <libby> YYYY-MM-DD chaals

17:38:25 <chaalsNCE> .g gMonth

17:38:28 <phenny> gMonth: http://www.techquila.com/topicmaps/xmlschema/x1h64smfal-1df.html

17:38:29 * DanC checks to see if http://www.w3.org/TR/NOTE-datetime has a "cite XML schema instead" marker

17:38:42 * crschmidt is here now

17:38:45 <bengee> is DARQ going to make it possible to find "today's birthdays" from a "mm-dd" literal?

17:38:49 <chaalsNCE> .g gMonth XML Schema datatype

17:38:51 <teefal> number of days from start of year :)

17:38:52 <phenny> gMonth XML Schema datatype: http://www.w3.org/TR/xmlschema-2/

17:38:59 <danbri> pjz, the idiom is similar, but more verbose. RDF has a datatyping syntax, uses an rdf:datatype="...long URI here" markup.

17:39:10 <crschmidt> ^q ?nick, ?bday where (?p foaf:dateOfBirth ?bday) (?p foaf:nick ?nick) AND ?bday =~ /09-22/

17:39:13 <julie> 1972-09-22 elisesara, 1981-09-22 jessical, 1975-09-22 jenett

17:39:19 <danbri> OK let's capture what we've agreed so far?

17:39:21 <crschmidt> That's why I like LJ's way of doing things, for the record.

17:39:42 <chaalsNCE> you could use the RDF datatype http://www.w3.org/2001/XMLSchema-datatypes#gMonthDay ...

17:39:43 <bengee> yeah, crschmidt, but is that going to be in the standard QL?

17:40:02 * crschmidt shrugs. I just walked in, trying to catch up on backscroll

17:40:20 <libby> ah chris, we waited awhile for you :)

17:40:20 * bengee thought using separate props would facilitate querying/reasoning

17:40:21 <teefal> we just had this whole conversation for bingo (when tags)

17:40:29 <teefal> finally gave up and did MM-DD

17:40:30 * DanC makes no guarantees that DAWG will standardize =~ as use above

17:40:38 <libby> yeah, it might bengee

17:40:46 <libby> think it's a valid proposal

17:40:49 <crschmidt> Just a summary about dateOfBirth: LJ uses it, I pulled it from Dan's file in a rush towards the end, it was mostly a mistake, but it's now used in all LJ foaf files

17:40:51 <chaalsNCE> Then you can refer to the existin warnings about calendars at http://www.w3.org/TR/xmlschema-2/#gMonthDay

17:40:54 <danbri> proposed: FOAF namespace should offer both foaf:dateOfBirth and foaf:birthday (ns editors to resolve CamelCase vs _ issues), the former being a specific date, the latter being some mention of day/month within any year. Syntax/value to be discussed.

17:41:01 <danbri> Is that where we're up to? anyone 2nd?

17:41:04 <DanC> anybody who requires =~ should say so to public-rdf-dawg-comments

17:41:10 <crschmidt> It uses either yyyy-mm-dd or mm-dd

17:41:10 <libby> I'd 2nd danbri

17:41:21 <libby> but we shoudl give bengee's proposal some talk too

17:41:35 <chaalsNCE> I prefer bengee's proposal

17:41:39 <libby> in some ways more elegant perhaps

17:41:40 <danbri> Sorry I talked over your proposal, bengee. Or was busy typing. Can you restate?

17:41:40 * crschmidt probably can not change LJ format at all at this point, so it's probably stuck

17:41:52 * DanC advises against delegating name changes to the editor; seems worth deciding the actual URIs

17:41:58 <danbri> Chris, what property name does LJ use?

17:42:02 <CaptSolo> re foaf:dateOfBirth & foaf:birthday - how would one indicate _only_ year of birth OR year & a month

17:42:04 <crschmidt> foaf:dateOfBirth

17:42:06 <DanC> ... since it's critical to interoperability

17:42:26 <DanC> one wouldn't, CaptSolo, as I understand the proposal

17:42:27 <bengee> I was thinking that using sepeate props for day/month/year would leave any freedom and still keep querying easy

17:42:28 <CaptSolo> knowing only a year can be sufficient in most cases to answer, i.e., if a person is over 18

17:42:39 <danbri> Chris, do you think they'd be persuadable to do another cycle of edits to the FOAF emission script? Or would that risk it being switched off through annoying them?

17:42:41 <libby> I would suggest that all proposals get sent to the list anyway to be discussed further, at least at this stage

17:42:52 <libby> true, capt

17:43:02 <danbri> yup, I was thinking to clarify things here then have them bubble up to the list...

17:43:07 * libby proposes danbri send around both proposals

17:43:12 <libby> in same mail

17:43:14 <crschmidt> danbri: i don't think it's risking a switch off at all at this point, I just know that I've been asked not to do any more development for them

17:43:34 <danbri> OK, thanks. (bummer btw, their loss though!)

17:43:35 <crschmidt> And no one was very open to FOAF changes back in the day

17:43:50 <danbri> I realise also that we didn't discuss the Bio namespace from Ian Davis.

17:43:53 <chaalsNCE> It's not hard to convert from somethig with a name an an iso8601 (I bet they don't add timezones, so that simplifies) to three properties...

17:44:01 <crschmidt> (someone else can offer changes to FOAF, but it was a long fight for me in the first place, so prepare for that)

17:44:01 <danbri> Libby, I took a bunch of actions already, fancy owning this one?

17:44:13 <libby> yeah, the bio takes a difefrent approach: an event (e.g. birth) and a date attached to it. interesting

17:44:17 <libby> ok

17:44:27 <libby> shall I include the bio one too reckon?

17:44:35 <danbri> Yes, please do...

17:45:08 <libby> ok

17:45:39 <danbri> It's so hard to remember everything relevant that's going on. Embarrassed I nearly forgot bio: ...

17:45:43 <CaptSolo> imo having 'yyyy' and 'yyyy-mm' for dates are valid uses

17:45:58 <CaptSolo> meaning, usages which might be necessary to express sometimes

17:46:05 <DanC> dates attached to events smells bad to me. I much prefer cyc's model, where you can say { ev1 startsBefore ev2 } directly without indirecting thru dates (ala SUMO.)

17:46:10 <CaptSolo> 'mm-dd' is valid for birthdays

17:46:30 <CaptSolo> only 'yyyy-dd' would not make sense of all the combinations

17:46:32 <pjz> events are nice b/c they offer a way to tie other info (ie. place) to them

17:47:09 <danbri> Blessing bio: has some merits

17:47:09 <danbri> ...also other properties sorta call for date-oriented extensions, and we don't have a design for doing that consistently

17:47:09 <libby> H:ACTION libby: send mail to rdfweb-dev outlining 3 proposals for date or birth / birthday

17:47:09 <dc_rdfig> Added comment H3.

17:47:09 <danbri> for example, foaf:schoolHomepage is sorta nifty...

17:47:09 <danbri> but there are 1000s of kids who've been to westergate school since I left there in 1988, and I dunno I care so much to arrange re-unions with _them_

17:47:09 <libby> c'mon, chump!

17:47:10 <danbri> for school, you can just about get away with using dateOfBirth as an additional filter

17:47:14 <chaalsNCE> right. And I don't see that events that have dates attached and events that have information about where in a temporal sequence thye are have to be incompatible

17:47:33 <danbri> for college,university etc., you can't assume ppl all the same age.

17:48:03 <danbri> ...and so that shades into wanting to do something like date-qualify states of affairs, which soon takes us off into the rich, if rathole-filled, world of contexts, rules etc.

17:48:21 <pjz> is ther a way in rdf to define foaf:birthday as a shortcut for <bio:event><bio:birth> ?

17:48:42 <teefal> ????-mm-dd :)

17:48:43 <bengee> H:[http://rdfweb.org/pipermail/rdfweb-dev/2004-August/013585.html|bengee's proposal from august]

17:48:44 <danbri> pjz, not in RDF as such. You could hack it in Cwm's N3, perhaps....

17:48:44 <dc_rdfig> Added comment H4.

17:48:45 <DanC> D: sorta tuned in, but mostly regrets: [http://www.w3.org/People/Connolly|Dan Connolly]

17:48:45 <dc_rdfig> Added comment D10.

17:48:55 <danbri> thanks bengee, appreciated

17:49:19 <danbri> OK anyone still burning to talk about birthday etc?

17:49:23 <DanC> F1:see [http://rdfweb.org/pipermail/rdfweb-dev/2004-September/013699.html|agenda request]

17:49:24 <dc_rdfig> Replaced comment F1.

17:49:33 <DanC> oops

17:49:35 <chaalsNCE> D: likewise [Charles McCathieNevile|http://www.w3.org/People/Charles]

17:49:43 <DanC> E1:see [http://rdfweb.org/pipermail/rdfweb-dev/2004-September/013699.html|agenda request]

17:49:44 <dc_rdfig> Replaced comment E1.

17:49:45 <teefal> i'm more interested in the privacy side of it, and how to do it

17:49:50 <danbri> If not, I'm interested to hear about any demos/hacks/etc ppl have made this week...

17:49:52 <danbri> or lately...

17:49:58 <chaalsNCE> I assume that foaf doesn't like datatypes much?

17:50:24 <DanC> F1:including jim's concerns

17:50:25 <dc_rdfig> Replaced comment F1.

17:50:39 <pjz> RDF doesn't like datatypes much

17:50:40 * DanC can't bring himself to repeat the expletive

17:50:48 <teefal> sent an email to leigh dodds about FOAF wrapper API ... came out of talks in galway

17:50:48 <danbri> I don't like datatypes in namespaces I use in RDF that non-SW-geeks might see... the syntax is too heavy. I'd be wary of putting datatype-based properties in FOAF.

17:51:09 <chaalsNCE> (as in people aren't rushing to use them. I ask because I have some friends who record their birthday in jewish or muslim calendars as well as gregorian ones. (neat trick - two birthdays a year) :-)

17:51:12 <teefal> we've talked a lot about how to make our foaf api public

17:51:20 <danbri> FOAF wrapper, cool I saw that. Leigh's got some stuff looming I think, might be interesting to have a compare/contrast IRC meeting re APIs

17:51:22 <pjz> I don't see FOAF as beeing seein by non-SW geeks

17:51:41 <teefal> and how developers who don't grok RDF could be lured into using foaf (and then RDF) by making it simpler

17:51:43 * chaalsNCE doesn't see code being seen by non-geeks

17:51:43 * pjz sees it as a data schema that happens to be human readable.

17:51:47 <crschmidt> D: Very late [Christopher Schmidt|http://crschmidt.net/]

17:51:47 <dc_rdfig> Added comment D11.

17:51:57 <danbri> I'm very interested in the question of when people will be happy(ish) using a 'raw' RDF API, vs wanting vocab-specific extensions (FOAF, RSS, DC etc).

17:51:59 * pjz is newish to using foaf and RDF.

17:52:05 * bengee is rewriting OWLchestra. it has an english UI now and allows attaching differetn types of documentation to terms. free version to be announced in the not too distant future..

17:52:27 * danbri notes that lots of people are encountering foaf, via livejournal, typepad etc., and its their first encounter with RDF (sometimes even with XML)

17:52:28 <chaalsNCE> I'll be happy when I have a query langauge that I can use and expect any system to understand the query.

17:52:31 <crschmidt> At my meeting with my boss today, discussed using both RDF and RSS as a way to show that company is on "new" edge

17:52:54 <danbri> Oh, before we wrap up, Next Meeting?

17:52:57 * pjz is working on a foafster-like thing where everyone can run their own little CGI node that gives them a good UI to look at their FOAF info (and maybe is a scutter too, though that's not written yet)

17:53:06 <bengee> Am now working on a term browser to see if things work..

17:53:14 <danbri> I'm going on a wholesome bikeride next week, 5pm or so, so can't make same date/time...

17:53:28 <teefal> i talked with the ecadamy guy (perl/php?) and ... one other person ... a c# guy ... about presenting a consistent API across languages (leigh & us are on the java side)

17:53:31 * danbri wonders if libby wants to run a meeting anyway?

17:53:55 <crschmidt> (Perhaps add a BLURB for projects re: FOAF?)

17:54:11 <ear1grey> D: actively lurking [Rich Boakes|http://dsg.port.ac.uk/~rjb]

17:54:12 <dc_rdfig> Added comment D12.

17:54:12 <teefal> goal is similiar to C++, in that it lets non-RDF-minded people use RDF before they realize it, then grow into other benefits

17:54:27 <pjz> I suggest putting date/time of next mtg in the topic of #foaf

17:54:38 <crschmidt> If anyone has a foaf file that julie doesn't know about, you can add with ^add http://example.org/foaf.rdf

17:54:39 * bengee won't be able to make it next week.

17:54:41 <libby> yep can do danbri if there's suffient interest

17:55:00 <danbri> there are actions to review, so if you don't mind, let's stick with the pattern...

17:55:07 <libby> pjz did you add yourself to D: btw?

17:55:09 <danbri> ...though would be nice to do a Japan-friendly meeting sometime

17:55:18 * pjz doesn't know what D: is

17:55:23 <chaalsNCE> I figured out why the mapping thing at http://www.w3.org/2001/sw/Europe/200306/geo only picked up Jim Ley (it's because it only looks for one namespace pattern, and it isn't the one I have seen most often)

17:55:32 <crschmidt> pjz: type D: here [name|url]

17:55:36 <chaalsNCE> pjz, it's attendees

17:55:46 <pjz> D: here Paul Jimenez

17:55:46 <dc_rdfig> Added comment D13.

17:55:52 <pjz> ah, got it

17:55:55 <crschmidt> for chomp bot, goes to http://rdfig.xmlhack.com/2004/09/22/2004-09-22.html#1095870548.004705

17:56:03 <crschmidt> you can type D13: to replace it

17:56:11 <pjz> I see

17:56:18 <crschmidt> Anyway

17:56:18 <chaalsNCE> crschmidt, any way to find out if julie already knows a particular file?

17:56:36 <crschmidt> chaalsNCE: not really, i'm still working on supporting contexts / list of added urls

17:56:50 <crschmidt> chaalsNCE: was having lots of problems wrt smushing when i was using contexts

17:56:59 <crschmidt> if there's a foaf:Person, you can do ^available to find available predicates

17:57:01 <crschmidt> ^available chaals

17:57:02 <julie> rdf:type, foaf:mbox_sha1sum, foaf:nick, foaf:name

17:57:10 <crschmidt> looks like you haven't added a file with that nick yet :)

17:57:33 <chaalsNCE> ^available mccathie-neville

17:57:33 <julie> No results.

17:57:44 <teefal> ^available teefal

17:57:44 <julie> No results.

17:57:44 <crschmidt> BLURB: Current projects underway for FOAF

17:57:45 <dc_rdfig> N: Current projects underway for FOAF from crschmidt

17:57:51 <chaalsNCE> ^available McCathie-Neville

17:57:51 <julie> No results.

17:57:54 <chaalsNCE> ^available McCathie-Nevile

17:57:55 <julie> No results.

17:57:59 <chaalsNCE> ^available McCathieNevile

17:57:59 <julie> No results.

17:58:06 <crschmidt> N: I'm currently working on redlandbot/julie, an IRC interface to a Redland storage

17:58:07 <dc_rdfig> Added comment N1.

17:58:45 <libby> D:next meeting [http://www.timeanddate.com/worldclock/fixedtime.html?day=29&month=9&year=2004&hour=16&min=30&sec=0&p1=0|2004-09-29, 1630 UTC] for 60-90 mins

17:58:45 <crschmidt> N: All code is stored in SVN, committed on a release early/often schedule in [public svn|http://crschmidt.net/svn/redlandbot/]

17:58:45 <dc_rdfig> Added comment D14.

17:58:46 <dc_rdfig> Added comment N2.

17:58:58 <chaalsNCE> ^add http://www.w3.org/People/Charles/foaf.rdf

17:58:58 <julie> Adding that to my database...

17:59:01 <danbri> OK so next meeting: same time, place (#rdfig tmp takeover), Libby's chumping the details...

17:59:08 <julie> Added 317 statements from http://www.w3.org/People/Charles/foaf.rdf. Model size is 382603.

17:59:09 <danbri> ...next week, I mean.

17:59:12 <danbri> regrets: danbri

17:59:15 <chaalsNCE> ^available McCathieNevile

17:59:15 <julie> No results.

17:59:21 <danbri> julie++ #nifty

17:59:27 <chaalsNCE> ^available Charles

17:59:27 <julie> No results.

17:59:32 <chaalsNCE> ^available charles

17:59:32 <julie> rdf:type, rdfs:seeAlso, foaf:nick, foaf:weblog

17:59:33 <crschmidt> it uses a nick

17:59:43 <chaalsNCE> ^available chaals

17:59:43 <julie> rdf:type, foaf:mbox_sha1sum, foaf:nick, foaf:name

17:59:59 <chaalsNCE> ^available Charles McCathieNevile

17:59:59 <julie> No results.

18:00:26 <crschmidt> N: also #julie

18:00:27 <dc_rdfig> Added comment N3.

18:00:34 <danbri> Oh, random fact time: there are 250 people (er, addresses) on the FOAF list, ie. rdfweb-dev.

18:00:53 <danbri> even after mailman's efforts to discourage subscriptions :/

18:01:09 <crschmidt> ^query select ?p where (?s foaf:givenname "Charles") (?s ?p ?o) using foaf for <http://xmlns.com/foaf/0.1/>

18:01:09 <julie> foaf:projectHomepage, srw1.1:speaks, srw1.1:masters, rdf:type, foaf:mbox, foaf:interest, foaf:givenname, srw1.1:reads, foaf:family_name, foaf:name, dc:description, foaf:workPlaceHomepage, foaf:knows

18:02:04 <CaptSolo> ^query select ?p where (?s foaf:givenname "Uldis") (?s ?p ?o) using foaf for <http://xmlns.com/foaf/0.1/>

18:02:04 <julie> Query returned no results

18:02:23 <danbri> OK I think we're...

18:02:24 <crschmidt> ^available CaptSolo

18:02:25 <julie> No results.

18:02:26 <danbri> ...ADJOURNED

18:02:35 <danbri> thanks everyone!

18:02:41 <danbri> but don't go!

18:02:51 <balbinus> :)

18:02:56 * danbri notes that #rdfig is now at 93 members, highest I've seen it

18:03:00 <CaptSolo> ^query select ?p where (?s foaf:firstname "Uldis") (?s ?p ?o) using foaf for <http://xmlns.com/foaf/0.1/>

18:03:00 <julie> Query returned no results

18:03:04 <danbri> ...#foaf at 62

18:03:57 <CaptSolo> danbri: will #rdfig inhabitants tolerate the randeom #foaf discussions? :)

18:04:05 <CaptSolo> random

18:04:08 <danbri> sure, we're mostly same people anyway ;)

18:04:11 * pjz waits impatiently for debian testing to go stable.

18:04:17 <CaptSolo> :)

18:04:22 <danbri> #foaf tends to be a bit more random

18:04:31 <CaptSolo> ok - gotta leave now - bye

18:04:33 <pjz> anyone know a way to get rdflib to output something... nice ?

18:04:49 <balbinus> bye CaptSolo

18:04:53 <balbinus> hi ndw

18:04:57 <pjz> I'm currently using xmltramp to manipulate my foaf data and rdflib to interpret/reason about it.

18:05:00 * chaalsNCE waits impatiently for the argentine peso to go stable :-)

18:05:17 <chaalsNCE> (really. But I think pjz will get there first)

18:05:26 * pjz chuckles.

18:05:27 <balbinus> :)

18:05:37 * pjz wants python-crypto in stable, and it's not.

18:05:45 <ndw> hi balbinus

18:05:50 <crschmidt> pjz: switch to gentoo!

18:06:03 <pjz> crschmidt: not on my server, thanks.

18:08:00 * ndw throws caution to the wind and runs unstable

18:08:10 <danbri> oh, chris... we weren't sure if you would be able to do more by way of crawler stats...

18:08:21 <danbri> ...you reported some in irc directly after last week's meeting

18:08:22 <crschmidt> danbri: i can, i haven't yet though

18:08:28 <crschmidt> it's on my ACTION list

18:08:37 <danbri> ...so I tenatively continued your action; ok that sounds appropriate then.

18:08:39 <danbri> cheers :)

18:08:42 <DanC> I run unstable too; I used to feel OK about it because I had redundancy... 2 machines. hmm... now that I went with a mac laptop, that no longer applies.

18:09:08 <danbri> I have mac laptop and unstable debian desktop, mostly ok

18:09:16 <danbri> when things go bad, i retreat to laptop

18:09:36 <danbri> I don't code much, so its mainly email/web/irc/ssh/cvs on both anyway, not rocket science

18:09:43 <DanC> right... but I can no longer use one to experiment

18:09:46 <danbri> things don't go bad much

18:09:54 <ndw> I've been bitten badly by unstable twice in the last, uhm five years.

18:10:08 <ndw> That's close enough to stable for me :-)

18:10:23 <danbri> worst thing currently, is my I18N'ized font setup is screwed

18:10:36 <danbri> oh hey, I reinstalled recently. Nearly forgot that. Now why was that? hmm

18:10:45 <danbri> Oh yeah, HD failed

18:10:59 <crschmidt> heh

18:11:14 <danbri> It isn't finding my arabic fonts and I have a very sketchy mental model for how font finding happens now...

18:11:25 <danbri> or japanese :(

18:11:31 <danbri> the Web looks distinctly less cool lately

18:12:38 <ndw> Speaking of stability, DanC, how stable is LVM?

18:13:17 <DanC> umm... very, I think.

18:13:28 <ndw> LVM1 or LVM2?

18:13:33 <DanC> There was a LVM1 -> LVM2 transition; I was spooked by it for a while, but it turned out to be nothing.

18:13:48 <DanC> LVM2, if you're using kernel 2.6

18:14:07 <ndw> So you're running LVM2 w/ReiserFS. I'll investigate. Reseirfs4 sounds cool but a wee bit too bleeding edge even for me :)

18:14:08 <DanC> but there are no horribly unescapable combinations

18:14:38 <DanC> I concur... reiserfs4 is something to watch, not to trust implicitly, yet

18:14:53 <DanC> reiserfs is the default for gentoo, suse, and such though. that did it for me.

18:15:10 * DanC wanders off to a telcon...

18:15:16 * ndw wonders how reiserfs differs from ext3...

18:17:00 <bengee> bye, folks

18:17:34 <crschmidt> reiser4 is totally different, i think? has all the metadata stuff?

18:17:43 <crschmidt> so you can cd filename, and it has metadata inside?

18:17:44 <ndw> yes, reiser4 is a totally different beast

18:17:50 <ndw> apparenty.

18:18:05 * ndw feels the need to setup an experimental reiser4 partition

18:18:19 <ndw> Must...work...today...resist...temptation...to...play...

18:18:21 <crschmidt> i use reiser, but not reiser4. reiser doesn't behave differently, although it does take a lot longer to do an ls of my folder with 100 gig of movies in it

18:18:25 <crschmidt> heh

18:18:39 * ndw wanders back to work. Work. You heard me: work.

18:20:10 <crschmidt> hm

18:20:10 <crschmidt> <rdf:RDF xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#" xmlns="http://www.w3.org/2003/01/geo/wgs84_pos#">

18:20:14 <crschmidt> <Point lat="-44" long="120" alt="12" /> </rdf:RDF>

18:20:14 <crschmidt> valid or not?

18:20:21 <crschmidt> rdf validator says deprecated, rapper says "not allowed"

18:20:37 <libby> needs ns on attributes

18:20:49 <crschmidt> ah

18:20:56 <libby> then ok I think

18:21:13 <crschmidt> well, the question was " can i do it without using ns on attributes" :)

18:21:25 <libby> heh. deprecated I think

18:21:35 <crschmidt> yeah

18:22:42 <danbri> chris, you can, if you use subelemnts instead

18:23:12 <crschmidt> subelements beoing <Point><lat></lat></Point> ?

22:05:25 <me_> me_ is now known as golbeck