Semantic Web Interest Group IRC Chat Logs for 2005-04-20

00:30:42 <daydream> can somebodu help me? http://www.openrdf.org/forum/mvnforum/viewthread?thread=555

00:33:58 <GabeW> DanC: you around?

00:36:39 * GabeW steps away for a minute

00:44:23 * GabeW back looking for DanC

00:50:16 * GabeW goes home

00:51:13 <crschmidt> later GabeW

00:55:21 * sbp reads the XRI specification, wonders why he's bothering

00:55:27 <sbp> pain: "the most significant difference between URN and XRI architecture is that URNs deal with only with persistent identifiers, whereas XRIs deal with all types of abstract identifiers--both persistent and reassignable"

00:56:49 <sbp> "But most of these are specific to one means of interaction. None of them is persistent across the many different ways that people, applications, and devices can communicate, and so they don’t function well as identifiers in the long run."

00:57:25 <sbp> rubbish. you can use HTTP URIs as XML namespaces, email addresses in TAG URIs and all sorts of sign up locations, as IFP identifiers in FOAF, etc.

00:59:20 <sbp> 3.1 is just "Cool URIs..." reiterated from the negative standpoint. "people will author bad URIs, so we need to change the scheme... it'll help in the following way: [handwave]"

01:01:05 <sbp> "descriptive and memorable for the side of a bus, for example, or dense and opaque for persistence" - dense != persistent. it's a social contract, not a function of the entropy of the URI!

01:02:31 * sbp just skims the RDF-done-badly bit

01:02:43 <GabeW> hey sbp

01:03:33 <sbp> hi Gabe

01:03:45 <sbp> hmm, "+" appears to be their "x-" space

01:04:13 * crschmidt wonders how people would do unit tests for something like a Redland wrapper

01:04:17 <crschmidt> or whether it's even worth it

01:04:21 <sbp> okay we...

01:04:22 <sbp> oh dear

01:04:29 <GabeW> please stop reading the intro ;-)

01:04:30 * sbp sees Gabe's name in the references

01:04:42 * sbp whistles, walks away

01:05:33 <sbp> as long as the specifications are the exact opposites of the technology specified in the introduction, it sounds very well designed with a good future

01:05:39 <GabeW> heh

01:05:45 <GabeW> you might be surprised

01:06:20 <sbp> it'd have to defy a good ten years' worth of the same stuff

01:06:23 <sbp> over and over and over again

01:06:32 <GabeW> i'd say that most of the stuff in the intro are what *some* folks might use XRIs for, but those examples aren't really what XRI is

01:06:35 <GabeW> if that makes sense

01:06:53 <sbp> I'd very much welcome links to material that sets it straight

01:07:01 * GabeW feels like he's being pulled into a vortex of confusion

01:07:22 <GabeW> I kinda wish the intro wasn't even there, because then people wouldn't be distracted and would have just read the specs...

01:07:58 * GabeW bangs head on keyboard

01:08:02 <chillywilly> go vortexes of confusion!

01:08:29 <sbp> well it looks like it's attempting to justify the rationale behind yet another new URI scheme, which is indeed the first question one must answer

01:08:48 <GabeW> well, it is the first question one has to answer *here*

01:09:20 <sbp> it's the first question you have to answer if you want to standardise a new URI scheme

01:10:35 * sbp goes looking for the XRI specifications, finds that Gabe is the TC *chair*... oh man

01:10:49 <GabeW> oh MAN, have I lost ALL cred or WHAT!

01:11:20 <GabeW> I'm trying to find a document that will make sense to you - and I'll be honest - that might be hard since most of these documents aren't targeted at folks like you, sbp (or me)

01:11:21 <sbp> it can't be as bad as HumanML

01:11:26 <GabeW> heh

01:12:04 <GabeW> <smile>

01:12:16 <sbp> the oh man was mainly because I didn't mean to be dissin' yo stuff too blatantly in front of you

01:12:21 <GabeW> thats ok

01:12:37 <GabeW> I'm torn here - I need to run, but I REALLY want to talk about this

01:12:52 <sbp> well that is, I'd like to provide a more in-depth critique but I've only just been exposed

01:13:00 <sbp> that's no problem. I'm around often :-)

01:13:10 <GabeW> hold on a sec

01:13:31 <GabeW> btw, there's been an #xri for a LONG time ...

01:13:42 <GabeW> not that anybody is ever there....

01:13:54 <GabeW> ok

01:13:56 <GabeW> so I can start at least

01:14:07 <sbp> yeah, I heard about it (the scheme) a while ago but there's been a *lot* of new URI schemes proposed over the past few years and they've been slipping further and further from my attention the wilder they get

01:14:16 <GabeW> sure

01:14:31 <GabeW> so, the first question is "why not HTTP URIs"...

01:14:54 <sbp> (heard: at least, I think I did... they're all TLAs!)

01:15:00 <sbp> right

01:15:21 <sbp> that's bound to the be DanC/RoyF-ian first question, at least

01:15:27 <GabeW> sure

01:15:30 <sbp> s/the be/be the/

01:15:34 <sbp> odd thetamesis

01:15:52 <GabeW> actually DanC's first issue seems to be with IPR and I hope to clear that up with him asap

01:15:57 <GabeW> royalty free

01:16:07 <sbp> ah, didn't even notice that

01:16:21 <sbp> generally these things have to go IETFwards so it's no issue

01:16:40 <GabeW> yah, but DanC was saying on the tag list that we wasn't even going to read the specs

01:16:44 <GabeW> foo

01:16:46 <GabeW> anyway

01:17:24 <sbp> a standard impartial concern for anything that you don't want to be sued for reincorporating (accidentally) elsewhere

01:17:34 <sbp> so, why not HTTP URI?

01:17:36 <GabeW> so, we've got some requirements that made using DNS for delegated naming very difficult and/or convoluted

01:19:00 <GabeW> (i'm thinking of the best way to describe this)

01:19:22 <sbp> DNS isn't optimal, agreed, but most approaches to solve that have been in one of two categories: a) assign a whole new space exactly like DNS but with a different set of guardians b) use an identifier along with some other component such as time, email, IP, etc. that forms an unambiguous identifer independent of the failings of DNS

01:19:45 <sbp> er, combination of components. tag uses time + email or domain, for example

01:19:51 <GabeW> well, delegation is very important to us

01:19:59 <sbp> I wrote a draft for one that's a combination of time + IP

01:20:03 <GabeW> well

01:20:13 <GabeW> we want to delegate between organizations

01:20:17 <GabeW> not unlike DNS does

01:20:25 <sbp> between organizations? how so?

01:20:45 <GabeW> ie not a flat, centrally controlled namespace

01:21:19 <GabeW> in fact, you can have "private roots" if you want using cross-references as the root (left-most) authority subsegment

01:21:46 <sbp> for non-Web use?

01:21:55 <sbp> or using a random ID?

01:22:01 <GabeW> yes, XRIs can be used on web or off web

01:22:17 <sbp> I mean the private root ones specifically. can you use those on the Web?

01:22:18 * GabeW realizes that was a statement likely to draw some reaction

01:22:22 <GabeW> ok

01:22:36 <sbp> you can use HTTP URIs off of the Web

01:22:38 <GabeW> think of it similar to a private "dot"

01:22:46 <GabeW> where the private dot is identified by a URI

01:22:52 <GabeW> (in terms of DNS speak)

01:22:53 <sbp> I've actually identified my paper documents with tags from time to time, sad person that I am

01:23:22 <GabeW> so xri://(mailto:gwachob@wachob.com)*home*door

01:23:40 <GabeW> if you don't know about (mailto:gwachob@wachob.com) a priori, you certainly can't *resolve* that XRI

01:23:44 <GabeW> but you can still use it

01:23:51 <GabeW> note that this is not in URI form

01:24:01 <sbp> okay. why's that different to, say, tag:gwachob@wachob.com,2005-04:home:door?

01:24:01 <GabeW> you have to apply escaping

01:24:16 <GabeW> because I can't delegate further to your tag URI

01:24:19 <GabeW> oh I see

01:24:21 <GabeW> ok

01:24:33 <GabeW> hmm

01:24:43 <GabeW> thats not really what I mean by a private root

01:24:50 <GabeW> I want the private root to be unique

01:25:03 <GabeW> (ie a URI)

01:25:13 <sbp> which part of the URI xri://(mailto:gwachob@wachob.com)*home*door is the private root--all of it?

01:25:23 <GabeW> the cross refernece (stuff in the parentheses)

01:25:29 <sbp> okay

01:25:48 <GabeW> so cross references can appear as a subsegment anywhere

01:25:49 <sbp> again, I don't see why xri://(mailto:gwachob@wachob.com) is any different to tag:gwachob@wachob.com,2005-04

01:26:07 <sbp> ah, now that's interesting

01:26:12 <GabeW> yah

01:26:18 <sbp> so one can start to extend someone else's scheme you mean?

01:26:25 <GabeW> so xri://=GabeWachob*(+email)

01:26:26 <GabeW> yah

01:26:39 <GabeW> cross references can have URIs or XRI values

01:26:47 <sbp> okay, we're getting somewhere. are there many scenarios you've identified where people would want to do such a thing?

01:26:55 <GabeW> yah

01:26:56 <sbp> can you possibly give me one?

01:26:57 <GabeW> ok

01:27:10 <GabeW> i'm trying to think of one thats illustrative in a useful way

01:27:13 <GabeW> without lots of setup

01:27:21 <sbp> yeah, I know it's hard to come up with quick examples

01:27:50 <GabeW> xri://=GabeWachob*(http://rdf.com/Parent)

01:27:56 <sbp> once you think of one though, try to remember it so that you can foist it on anybody else who asks the same question :-)

01:27:57 <GabeW> in XRI land, we might use +

01:28:04 <GabeW> xri://=GabeWachob*(+Parent)

01:28:07 <sbp> okay, what's going on in there?

01:28:27 <GabeW> the stuff in () is JUST a opaque string in the form of a URI or XRI value (ie the stuff after xri://)

01:28:30 <sbp> I only see one set of parens, for a start

01:28:51 <GabeW> huh?

01:29:15 <GabeW> "one set"?

01:29:34 <sbp> well, I thought that the interesting part was this: "<sbp> so one can start to extend someone else's scheme you mean?" which I naturally figured would take the form of xri://(personOne)foo(personTwo)bar

01:29:40 <GabeW> ooh

01:29:42 <GabeW> so

01:29:42 <sbp> one set: an oparen and a cparen

01:30:14 <GabeW> xri://(mailto:gwachob@wachob.com)*foo*(mailto:sbp@infomesh.net)*bar

01:30:30 <GabeW> like that?

01:30:37 <sbp> right, that's what I was thinking

01:30:47 <GabeW> there are a ton of applications of cross references

01:30:51 <sbp> but I was mainly asking for a scenario for that

01:30:57 * chillywilly 's head spins

01:30:57 <GabeW> yah

01:31:05 <sbp> heh, heh. hey chillywilly

01:31:11 <chillywilly> hallo

01:31:26 <GabeW> so the other thing to remember is that we (mostly me) wrote a resolution mechanism for XRIs (if you want to use it)

01:31:35 <GabeW> that is based on HTTP + XML

01:31:40 <GabeW> its pretty simple

01:31:45 <sbp> (chillywilly: incidentally, every time I see you around on Freenode I'm reminded of Over The Hills And Far Away. you said you could play it once, back in 2001 or so, and I tend not to forget a Zeppelin reference)

01:32:07 <sbp> GabeW: that'd enable XRIs to be embedded in browsers to resolve to HTTP?

01:32:12 <GabeW> yah

01:32:29 <GabeW> it should be possible to do the client side of XRI resolution even in Javascript (using AJAX stuff)

01:32:30 <sbp> which is a good first step if you want to go for acceptance...

01:32:31 <chillywilly> I still rock it out here and there

01:32:39 <sbp> heh

01:32:44 <GabeW> yah, i know folks are talking about a firefox plugin

01:32:47 <GabeW> ok

01:32:48 <chillywilly> I play every sunday in the band at my church

01:32:51 <chillywilly> ;P

01:32:53 <sbp> there's also protozilla

01:33:06 <GabeW> so the thing is -- we've been talking about the "authority resolution" part - basically the stuff before the slsh

01:33:07 <sbp> chillywilly: it'd make an interesting church song...

01:33:09 <GabeW> slash

01:33:14 <chillywilly> sbp: haha

01:33:24 <GabeW> in HTTP, authority resolution is basically just DNS -> IP

01:33:48 <GabeW> in XRI, its authority -> XML (actually a series of XML elements called XRI Descriptors)

01:34:16 <sbp> hmm, this is very close to URCs indeed

01:34:23 <sbp> have you heard about URCs?

01:34:24 <GabeW> and you can stuff whatever you want in those descriptors - though we define some basic stuff to make XRI authority resolution work

01:34:35 <GabeW> its been a while since I"ve heard URC

01:35:08 <GabeW> so, anyway, when you are done resolving the authority, you end up with an XML document that contains metadata

01:35:28 * sbp tries to find a good reference for it

01:35:36 <sbp> it was essentially exactly this, but pre-XML

01:35:43 <GabeW> And we have a concept of "local access"

01:36:05 <sbp> "Uniform resource characteristic (URC) is a method for encoding information about a given information object. This information is called meta-information as it is information that is not found in the resource itself." - http://www.cdt.luth.se/~peppar/master_thesis/node14.html

01:36:09 <GabeW> local access URIs which talk certain protocols -

01:36:11 <GabeW> e.g. HTTP

01:36:19 <sbp> what's their purpose?

01:36:22 <GabeW> so at the end you can interact with a resource identified by an XRI through HTTP

01:36:28 <GabeW> or any other protocol

01:37:15 <sbp> oh, I see now

01:37:27 <sbp> yes, I think that's just what the real world calls "URL"

01:37:36 <sbp> as the XRI introduction correctly states

01:38:00 <sbp> I define a URL scheme as one that is bound closely to a particular access protocol, usually network based

01:38:13 <GabeW> sure

01:38:15 <GabeW> thats it

01:38:57 <sbp> so how does that link to the XML used in the authority look-up in XRIs?

01:39:04 <GabeW> cool, glad you asked

01:39:09 <sbp> :-)

01:39:49 <GabeW> the result of an authority lookup is the XML document that contains (among other things) one or more "Service" (local access service) elements

01:39:59 <GabeW> each of which contains one or more URIs and metadata about the service

01:40:15 <GabeW> btw, you could stick in RDF in the resulting XML documents if you wanted to

01:40:36 <sbp> so an XRI can map onto multiple URLs, you mean, each of which can be marked up with additional metadata that may or may not include RDF?

01:40:40 <GabeW> yes

01:40:51 <sbp> so... you might resolve an XRI and get a choice of several different languages say

01:40:59 <GabeW> "languages"?

01:41:08 <sbp> each of which is represented by a different HTTP URI (ignore conneg for now)

01:41:10 <sbp> natural languages

01:41:11 <GabeW> yes

01:41:18 <GabeW> thats right

01:41:19 <sbp> somedocument.es.html, somedocument.en.html

01:41:47 <sbp> so, next question: how does one get an XML document from the XRI?

01:41:49 <GabeW> or you could get LDAP URIs or SS7 URIs (or tel: URIs)

01:42:06 <GabeW> one performs authority resolution

01:42:13 <GabeW> and gets an XRI Descriptor

01:42:26 <sbp> which protocol is used in that resolution process?

01:42:42 <GabeW> the XRI Descriptor is an XML document describing the authority

01:42:43 <GabeW> ok

01:42:51 <GabeW> the resolution process is a series of HTTP requests

01:42:58 <GabeW> defined in the XRI Resolution spec

01:43:36 <sbp> okay, I think I understand. so you need to convert the names in the XRI into IP addresses from which you can get the XML document, yes? and you're using HTTP to do that

01:44:40 <GabeW> well - the name-parts (subsegments) get converted into HTTP URIs

01:44:46 <GabeW> and then we do a HTTP GET to get descriptors

01:45:29 <GabeW> remember, resolution is only for the authority segment (everything before the URI path part of the XRI)

01:45:37 <GabeW> that is, XRI-defined resolution

01:46:45 * GabeW hopes the pause indicates illumination above sbp's head

01:47:13 <sbp> sorry, got momentarily distracted. reading

01:47:30 <sbp> right, that's pretty much what I expected

01:47:37 <sbp> so then that raises a further question:

01:48:09 <sbp> if you have something along the lines of the xri://(authOne)...(authTwo)... URI that I'm so keen on, I presume you'll first resolve authOne and then authTwo, yes?

01:48:17 <GabeW> well

01:48:28 <GabeW> XRI resolution treats those as single units

01:48:32 <sbp> if so, that would mean authTwo is still dependent on authOne

01:48:35 <GabeW> XRI resolution goes from left to right

01:49:03 <GabeW> and therefore, (authTwo) gets resolved, as an opaque string, in the context of the prefix of subsegments to the left of it

01:49:06 <GabeW> ie

01:49:18 <GabeW> xri://(authOne)*foo*(authTwo)

01:49:36 <GabeW> in that case, (authTwo) is resolved at the authority identified by xri://(authOne)*foo

01:49:49 <sbp> okay, so authOne *explicitly* has to say that anything to the right can be extensible by using yet another authority component?

01:50:01 <GabeW> well

01:50:05 <GabeW> it delegates

01:50:12 <GabeW> so foo is the one who decides

01:50:18 <GabeW> that is xri://(authOne)*foo

01:50:22 <sbp> oh

01:50:35 <sbp> I thought foo was just an identifer that authOne controlled

01:50:36 <GabeW> its just left-to-right delegation

01:50:39 <GabeW> it is

01:50:43 <GabeW> well

01:50:49 <GabeW> or it can be an independent authority

01:51:04 <GabeW> *'s are the delimiters between authorities

01:51:21 <sbp> but authOne gets to choose that? so how is xri://(authOne)*foo*(authTwo) different from xri://(authOne)*foo*authTwo?

01:52:12 <GabeW> its not different except that the string (authTwo) is not a legal cross reference - it would be (=AuthTwo) ... Xrefs are just a syntactic way of including a URI - not unlike quotes in the english language

01:52:29 <GabeW> and no, authOne doesn't really have control over foo in our case

01:52:44 <sbp> this is quite mind-bending!

01:52:53 <GabeW> its just left to right

01:53:14 <GabeW> not unlike DNS (though the other way - DNS is right to left delegation)

01:53:29 <GabeW> com doesn't control what I put under wachob.com

01:53:29 <sbp> well only because they spelled domain names backwards for some reason

01:53:41 <GabeW> sbp++

01:53:46 <sbp> I think it's TimBL's fault technically: he could've specified it be reversed

01:54:05 <GabeW> i'm not going to jump down his throat for that ;-)

01:54:25 <sbp> well, I believe he's expressed his own regret at not doing so

01:55:07 <sbp> "I have to say that now I regret that the syntax is so clumsy. I would like http://www.example.com/foo/bar/baz to be just written http:com/example/foo/bar/baz" - http://www.w3.org/People/Berners-Lee/FAQ.html#etc

01:55:20 <GabeW> yow

01:55:50 <sbp> yeah. I'd be happy with just http://com.example/foo/bar/baz

01:55:55 <GabeW> ok

01:56:00 <GabeW> well thats very interesting

01:56:02 <sbp> no subdomains, but // is kinda useful sometimes

01:56:34 <GabeW> ok sbp

01:56:39 <sbp> (it may be the case that the domain reversal actually helped uptake though--blargh.com seems to stick in people's minds more)

01:56:40 <GabeW> did I give you a flavor?

01:56:54 <GabeW> or rather, a *flavour*?

01:56:56 <sbp> you did indeed--many thanks for staying and helping me to understand it more! it's much appreciated

01:57:03 <sbp> either way. flavo[u]r

01:57:15 <GabeW> i'm glad #swig is archived

01:57:24 <sbp> yep. I'll chump it all in a moment

01:57:26 <GabeW> yay

01:58:04 <GabeW> i'm glad I got to do this too

01:58:19 <GabeW> I haven't had a conversation like this about XRI with anyone at this level of detail outside the TC

01:58:20 <sbp> my opinion is currently that I think it's on shaky ground, but there are glimpses of some interesting features that certainly are lacking in other schemes that I know of, and that *might* be useful in the wider world, though I'd still need to be shown lots and lots of real-world examples first

01:58:24 <sbp> really? wow

01:58:37 <sbp> well I hope it's been useful for you too, and not just a sweeping dismissal

01:58:39 <GabeW> thats actually great to hear

01:59:01 <GabeW> I think getting beyond the "what the hell is it" is that hardest thing

01:59:06 <GabeW> s/that/the/

01:59:24 <sbp> and also getting hardened URI folk over the fact that there's just so much of this. this does seem probably the most interesting URI proposal since tag: though

02:00:09 <GabeW> cool

02:00:35 <sbp> "Sensational new way to identify your resources." - The New York Times; "probably the most interesting URI proposal since tag:" - Sean B. Palmer; "Yowee! Frozen child found encased in XRI!" - Weekly World News

02:00:38 <GabeW> well, for what its worth, we've certainly paid attention to the world of URIs

02:01:03 <GabeW> we've worked really hard to comply with the new URI/IRI RFCs, etc

02:01:07 <GabeW> we do unicode!

02:01:10 <sbp> good. do you have any other URI big-names on the group?

02:01:12 <GabeW> which is a bit scary

02:01:16 <sbp> heh, heh. that's very good to hear

02:01:23 <GabeW> *other*?

02:01:23 <sbp> though IRI opinion still seems a bit split

02:01:26 <GabeW> yah

02:01:55 <sbp> er, I dunno why I said other, but you can take it as me counting you in :-)

02:02:02 <GabeW> oh gee thanks

02:02:07 <sbp> heh, heh

02:02:15 <GabeW> We have a good number of folks, but I'm not sure they are very active W3C or IETF folks

02:02:24 <sbp> I meant Masinter, Durst, etc.

02:02:27 <GabeW> ah

02:02:30 <GabeW> no

02:02:33 <sbp> ah

02:02:41 <sbp> it'll be nice to see what their responses are

02:03:12 <sbp> very divided opinions in that top set, usually, so it'll be like Variety Hour at the URI Hippodrome

02:03:52 <GabeW> yah

02:04:13 * GabeW has thick skin

02:04:17 <sbp>http://ilrt.org/discovery/chatlogs/swig/2005-04-20.html#T00-33-58

02:04:18 <dc_swig> A: http://ilrt.org/discovery/chatlogs/swig/2005-04-20.html#T00-33-58 from sbp

02:04:24 <sbp> A:XRI Discussion on #swig

02:04:24 <dc_swig> Added comment A1.

02:04:26 <sbp> argh

02:04:30 <sbp> A:|XRI Discussion on #swig

02:04:30 <dc_swig> Titled item A.

02:05:14 <sbp> A:Which [http://ilrt.org/discovery/chatlogs/swig/2005-04-19.html#T22-48-35-4|DanC] actually started about an hour earlier, and GabeW and I discussed for an hour more.

02:05:15 <dc_swig> Added comment A2.

02:06:15 <sbp> A:The [http://www.oasis-open.org/apps/group_public/download.php/11857/xri-intro-V2.0-wd-04.pdf|XRI Intro (V2.0, WD)] got us started on this, but GabeW didn't think it sufficient as an explanation for #swig types so he graciously stayed around to answer questions.

02:06:15 <dc_swig> Added comment A3.

02:06:33 <sbp> A1:""

02:06:34 <dc_swig> Deleted comment A1.

02:06:48 <GabeW> :-)

02:10:15 <GabeW> and I don't know if this is officially launched, but http://sourceforge.net/projects/openxri

02:10:53 <sbp> ooh, code I can play around with?

02:11:00 <sbp> ah no, just Java. good though

02:11:08 <GabeW> oh

02:11:16 * sbp downloads it

02:11:26 <GabeW> http://www.wachob.com/xriescape - not everything, but something

02:11:26 <dc_swig> B: http://www.wachob.com/xriescape from GabeW

02:11:33 <GabeW> woops

02:11:39 <GabeW>http://sourceforge.net/projects/openxri

02:11:40 <dc_swig> C: http://sourceforge.net/projects/openxri from GabeW

02:11:57 <GabeW> B:Some XRI conversion code in python

02:11:57 <dc_swig> Added comment B1.

02:12:09 <sbp> B:|XRI to URI Conversion Service

02:12:10 <dc_swig> Titled item B.

02:12:13 <GabeW> C:|OpenXRI Sourceforge Project (Java)

02:12:14 <dc_swig> Titled item C.

02:12:18 <sbp> B:Written in Python (yay)

02:12:18 <dc_swig> Added comment B2.

02:12:23 <sbp> B1:""

02:12:24 <dc_swig> Deleted comment B1.

02:12:28 <GabeW> sbp: would I do it in ANYTHING else?

02:12:31 <GabeW> ;-)

02:12:59 <GabeW> B: No guarantee its 100% compliant (feedback welcomed)

02:13:00 <dc_swig> Added comment B2.

02:13:12 <sbp> aw, POST not GET?

02:13:32 <GabeW> uh

02:13:43 <GabeW> the source is there dude

02:13:49 <sbp> hmm?

02:13:56 <sbp> GET is bookmarkable and pointable

02:13:58 <sbp> POST isn't

02:14:00 <GabeW> er

02:14:34 <GabeW> ok hold on

02:15:07 <sbp> all you have to do is s/POST/GET/ in the CGI

02:15:19 <sbp> still works (e.g. http://www.wachob.com/xriescape/escape.cgi?xri=xri://(http://www.news.com)/%30test*here/(+value/(+sub/thing))/blah/(+test)/foo/(+another) already Does The Right Thing)

02:15:43 <GabeW> actually, just in the HTML

02:16:02 <GabeW> which I just did ;-)

02:16:05 <GabeW> python is great that way

02:16:38 <sbp> cool

02:16:42 <GabeW> ok, my kids are going to stop calling me dad if I don't get home now

02:16:48 <sbp> chuckle

02:16:52 <GabeW> ttyl

02:16:56 <sbp> c'ya, and thanks again!

02:17:07 <GabeW> sure

02:17:09 <GabeW> bye

02:18:14 * crschmidt wonders if anyone knowas anything about PHP memory management

03:21:34 * crschmidt heads to bed.

07:58:34 <jeen__> jeen__ is now known as jeen

08:30:24 <AndyS>http://lists.w3.org/Archives/Public/semantic-web/2005Apr/0139.html

08:30:24 <dc_swig> D: http://lists.w3.org/Archives/Public/semantic-web/2005Apr/0139.html from AndyS

08:30:33 <AndyS> D:|Jena 2.2

08:30:33 <dc_swig> Titled item D.

08:31:12 <AndyS> D:[http://jena.sourceforge.net/|Jena web site] and [http://prdownloads.sourceforge.net/jena/Jena-2.2.zip?download|download]

08:31:13 <dc_swig> Added comment D1.

08:44:45 <libby> nice one AndyS :)

08:46:31 <AndyS> And the rest of the team. Anything OWL related has been the most time consuming.

08:46:46 <AndyS> BTW Which version of Joseki are you using? SPARQL version?

08:49:23 <jeen__> congrats andy (+ rest of jena team) :)

08:49:28 <jeen__> jeen__ is now known as jeen

08:50:46 <libby> yep andys

08:51:29 <libby> it doesn't do write what I want - I want to pass it a url of some rdf and have it load that into the database. Guess I'll have to get hacking :)

08:52:04 <AndyS> Yep!

08:52:22 <AndyS> There is a command line app (jena.dbload) that will do it for you.

08:52:40 <AndyS> So a CGI that kicks that would be one way.

08:53:04 <AndyS> Unless the data is really, really big.

08:53:37 <AndyS>http://www.w3.org/TR/rdf-sparql-query/

08:53:39 <dc_swig> E: http://www.w3.org/TR/rdf-sparql-query/ from AndyS

08:53:45 <AndyS> :|SPARQL Query Language for RDF

08:53:52 <AndyS> E:|SPARQL Query Language for RDF

08:53:52 <dc_swig> Titled item E.

08:54:13 <AndyS> E::New syntax with Turtle-style triple patterns

08:54:14 <dc_swig> Added comment E1.

08:54:30 <AndyS> E:Comments to public-rdf-dawg-comments@w3.org please

08:54:31 <dc_swig> Added comment E2.

08:54:44 * AndyS gets chump overdose

08:55:02 <AndyS> E1:New syntax with Turtle-style triple patterns

08:55:03 <dc_swig> Replaced comment E1.

08:55:34 <libby> oh cool andys. the commendline app might be enough

08:56:09 <libby> so does this version of jena include a new version of joseki or are these separate?

08:56:46 * libby wonders if arc is tracking E alreday?

08:57:25 <AndyS> Joseki is still at http://www.joseki.org/

08:57:34 <AndyS> ARQ is stil separate from Jena as a download

08:58:03 <AndyS>http://jena.hpl.hp.com/~afs/ARQ/

08:58:04 <dc_swig> F: http://jena.hpl.hp.com/~afs/ARQ/ from AndyS

08:58:21 <AndyS> F:|ARQ - SPARQL for Jena

08:58:21 <dc_swig> Titled item F.

08:58:42 <AndyS> (chump syntax is growing on me!)

08:59:04 <libby> right

09:00:07 <AndyS> F:[http://jena.hpl.hp.com/~afs/ARQ/download.html|Download ARQ] from SourceForge

09:00:08 <dc_swig> Added comment F1.

09:00:48 <AndyS> ARQ is update mostly. I need to check and improve value testing against the latest text

09:01:21 <AndyS> Eric has clarified the corner cases I need to go back and check.

09:09:09 <libby> cool :)

09:18:19 <danbri> anyone working on new SPARQL syntax parser for PHP?

09:26:31 <libby> did you see alberto's stuff danbri? I guess that was the old syntax tho

09:26:39 <libby> atm anyway

09:32:28 <danbri> old syntax, yep

09:32:34 * danbri mailed chris bizer to ask his plans too

09:35:01 <dajobe>http://www.w3.org/TR/2005/WD-rdf-sparql-query-20050419/

09:35:01 <dc_swig> G: http://www.w3.org/TR/2005/WD-rdf-sparql-query-20050419/ from dajobe

09:35:22 <dajobe> G:|SPARQL Query Language for RDF - W3C Working Draft 19 April 2005

09:35:22 <dc_swig> Titled item G.

09:35:41 <dajobe> G:congrats eric, andy

09:35:41 <dc_swig> Added comment G1.

09:41:18 <darobin> wheee, kudos

09:45:06 * dajobe blogues

11:20:49 <chaals> anyone know where to get a copy of the java version of RDF Author?

11:21:14 <chaals> (Damian's ~pld.. isn't there :-(

11:23:42 * crschmidt wakes up

11:35:32 <libby> chaals I'll bug him about it - he's done an update to it too

12:06:22 <chaals> libby, thanks

12:06:37 * chaals used wordpad instead (<BIGSIGH/>

12:06:53 <libby> :(

12:08:31 <chaals> No worries... just trying to show how to build the human rights stuff modelled in RDF.

12:08:38 <chaals> Not as pretty as validator :-(

12:08:45 <chaals> (Does the querying work in RDFAuthor?)

12:08:51 <chaals> (java version?)

12:09:09 * chaals going to have his mac back in 24 hours and be ableto se the real version anyway...

12:09:10 <libby> it does now! that's what shellac was working on at sparqling days

12:09:16 <chaals> Cool!

12:09:20 <chaals> shelac++

12:09:21 <libby> ...though only in the java version

12:09:24 <chaals> shellac++

12:09:25 <libby> yeah!

12:09:35 <libby> it used to use soap, now just restful

12:10:38 * chaals gotta run...

12:10:44 <libby> cu!

12:10:48 <chaals> cu!

12:10:54 <libby> some back soon :)

12:26:14 <libby>http://rdfweb.org/pipermail/rdfweb-dev/2005-April/014043.html

12:26:14 <dc_swig> H: http://rdfweb.org/pipermail/rdfweb-dev/2005-April/014043.html from libby

12:26:31 <libby> H:|Foaf meeting later, here on #swig, if that's ok

12:26:32 <dc_swig> Titled item H.

12:27:26 <libby> H:[http://www.timeanddate.com/worldclock/fixedtime.html?month=4&day=20&year=2005&hour=17&min=0&sec=0&p1=0|1800 UTC]

12:27:27 <dc_swig> Added comment H1.

12:55:19 <Pike_> libby: how many reviewers does a single paper for sfsw get?

12:56:40 * libby checks

12:57:33 <libby> why, you too busy pike_?

12:58:18 <Pike_> libby: no, I'm just full of doubt on one of them, and wonder what's going to happen to it

12:58:46 <libby> it has 3

12:58:55 <libby> all have 3, rather

12:59:10 <Pike_> 'k

12:59:25 <libby> just note your comments and we'll have a look. many thanks for doing it pike_

12:59:28 <libby> :)

12:59:40 <libby> hey, you work on mozilla rdf stuff, is that right?

13:00:02 <Pike_> I'm module owner. And sometimes, I even work on it ;-)

13:00:06 <libby> :)

13:00:13 * libby been playing with it a bit

13:00:19 <libby> pretty neat

13:01:04 <Pike_> we gonna whack the api quite a bit, though, to make it more performant and closer to the current spec

13:01:14 * libby wonders how scalable it is

13:02:07 <libby> I don;t know much about it - I've just been trying to parse rather large rdf files with jim ley's (neato) javascript rdf api but it's struggling with very large files

13:03:43 <Pike_> yes, likely. under the hood, it has two hashtables for forward and backward arcs, which may be suboptimal in terms of memory consumption (but good in terms of speed)

13:03:59 <Pike_> and we suck at large containers, too

13:04:21 <Pike_> so if you had a big Bag or something, you went for the sweet spot ;-)

13:04:31 <libby> heh

13:04:49 * libby not really done much testing

13:05:02 <libby> is the moz implementation in .js or something else?

13:05:04 <crschmidt> Is there a tutorial on how to use it anywhere?

13:05:16 <Pike_> c++

13:05:19 <crschmidt> I only found out about it by reading through the source somewhere

13:05:30 <libby> ah, right, I was wondering pike_

13:05:47 <libby> liekly faster than the javascript one I was using then

13:05:59 <Pike_> well, the docs are old, danbri did most of those a few years back. Not that a whole lot changed since then yet

13:06:12 <crschmidt> got a URL? couldn't find it when I was looking

13:06:18 <Pike_> libby: maybe you just used a js wrapper to the c++ stuff

13:06:58 <crschmidt> Pike_: no, she's using a complete javascript parser, written by Jim Ley: he uses it for IE stuff

13:07:07 <Pike_> http://www.mozilla.org/rdf/doc/, but everything in there is deprecated without explicit notice

13:07:08 <dc_swig> I: http://www.mozilla.org/rdf/doc/, from Pike_

13:07:13 <crschmidt> heh, i like that I'm the first hi for "mozrdf"

13:07:18 <libby> sorry pike_, I'm confusing you, I was using the mozilla rdf stuff to access the bookmarks but *another* jaavscript rdf parser (by jibberjim) to parse incoming RDF

13:08:09 <libby> my guess is that jim's is slower than the mozilla one but has the advantage that I can bug jim for updates when I need them :)

13:10:01 <crschmidt> I:| Mozilla RDF Documentation

13:10:01 <dc_swig> Titled item I.

13:11:42 <Pike_> you can bug the mozilla impl, too, or rather, me. It may be that jim is quicker to respond, sadly

13:12:08 <crschmidt> certainly the release process is quicker: just release a new javascript :)

13:12:46 <JibberJim> of course libby's last requests I was allready w

13:12:58 <JibberJim> supporting if she'd only used an up-to-date version :-)

13:13:02 <crschmidt> eep: The most common cause for RDF/XML not to load from a web server is incorrect MIME type. Make sure your server is delivering the file as text/xml (recommended) or text/rdf (bogus).

13:13:06 <crschmidt> Note that the W3C RDF Core WG are registering application/rdf+xml, though this isn't understood by any Mozilla code yet. (do we have a bug registered to track this? -- danbri)

13:13:58 <libby> well, I was sat near jim at the time so I could hassle him directly!

13:14:54 <crschmidt> so, can the mozilla code still not read application/rdf+xml RDF?

13:18:01 <Pike_> crschmidt: looking at the source, it does now

13:18:55 <Pike_> https://bugzilla.mozilla.org/show_bug.cgi?id=61839

13:18:56 <dc_swig> J: https://bugzilla.mozilla.org/show_bug.cgi?id=61839 from Pike_

13:21:47 <crschmidt> J:| Add Support for application/rdf+xml to Mozilla RDF parser

13:21:47 <dc_swig> Titled item J.

13:26:03 <crschmidt> hm, can't use mozRDF from in-page javascript?

13:26:46 * crschmidt just tried it at http://crschmidt.net/tmp/mozrdf.html , getting permission denied on getting property UnnamedClass.classes

13:26:47 <Pike_> no, this is one of the things that got me sucked into hacking on the engine

13:27:30 <crschmidt> gotta be from XUL or something, then?

13:28:12 <Pike_> from chrome

13:28:16 * crschmidt nods

13:28:44 <Pike_> basically, the rdfservice loads anything you feed into it right now, without any security checks. We can't expose that

13:28:56 <crschmidt> Yeah, kinda figured

13:29:15 <Pike_> and you can't create arbitrary xpcom components from web content, either, for exactly the same reason

13:30:40 * crschmidt does not like the idea of going back into mozilla extension hacking

13:33:26 <crschmidt> I did it once, and that was far too much :)

13:35:09 <Pike_> he

13:35:25 * Pike_ notices that foaflog is full of comment spam

13:35:52 <libby> foaflog?

13:36:18 <crschmidt> the old MT blog, probably

13:36:33 <libby> rdfweb.org blog?

13:36:34 <Pike_> yes, for example, http://rdfweb.org/mt/foaflog/archives/000051.html

13:36:42 <libby> ah crap

13:36:50 <libby> that's an old installation of the rdfweb.org blog

13:36:53 <crschmidt> Pike_: yeah, we're switching to a new weblog system Real Soon Now with the new beta site

13:37:01 <libby> also, we alreday did

13:37:05 <crschmidt> did?

13:37:07 <crschmidt> missed that

13:37:16 <libby> i.e. that's old, immuatble pages form an old version :/

13:37:37 <Pike_> libby: that one is still linked on foaf-project.org

13:38:14 <libby> http://rdfweb.org/mt/foaflog/archives/2004/07/21/13.40.56/index.html

13:38:18 <libby> are the right ones

13:38:20 <libby> hm, I fix

13:39:12 <crschmidt> Pike_: are you going to be at the foaf meeting in a couple hours? :)

13:40:13 <Pike_> I'd be suprised :-)

13:40:20 <libby> pike_ this is the right site: http://rdfweb.org/ it's just that there are old dead archives lurking around

13:40:37 <libby> where did you find that spammed one linked form?

13:40:45 <crschmidt> getting started

13:40:49 <crschmidt> on foaf-project.org, first link

13:41:08 <libby> aha! thanks chris

13:41:20 <Pike_> libby: yeah, and there are more, on http://rdfweb.org/topic/FAQ

13:41:33 <libby> cheers

13:42:26 <dajobe>http://www.ilrt.bris.ac.uk/people/cmdjb/2005/04-sparql/

13:42:26 <dc_swig> K: http://www.ilrt.bris.ac.uk/people/cmdjb/2005/04-sparql/ from dajobe

13:42:49 <dajobe> K:|SPARQL RDF Query Language Reference

13:42:50 <dc_swig> Titled item K.

13:42:56 <dajobe> K:2 page summary of sparql

13:42:56 <dc_swig> Added comment K1.

13:44:58 <crschmidt> dajobe: should SPARQL queries support things like <#foo> a:foo ?x; a:bar ?y. (in other words, the ; abbreviation)?

13:45:12 <crschmidt> oh, nevermind, found an example

13:45:16 <crschmidt> (the answer is yes)

13:50:03 <dajobe> K:managed to sneak a turtle reference into about 3/4 of a page too

13:50:04 <dc_swig> Added comment K2.

13:51:07 <crschmidt> K: Very nice summary

13:51:08 <dc_swig> Added comment K3.

13:52:11 <libby> H:agenda+ getting people to link galleries, calendars etc from foaffiles

13:52:11 <dc_swig> Added comment H2.

13:54:51 <CaptSolo> is chaals around?

13:56:26 <uche> Hi all

13:56:41 <uche> <uche> I'm needing to convince a new group at Sun to use RDF for a project

13:56:41 <uche> <uche> They're asking about examples of high scalability RDF implementations out there

13:56:41 <uche> <uche> I mentioned a few, including MusicBrainz

13:56:41 <uche> <uche> Anyone know of some?

13:56:41 <uche> <uche> I also mentioned some big & visible companies that are involved in RDF:

13:56:42 <uche> <uche> Oracle, IBM, HP,

13:56:43 <libby> heya uche :)

13:56:44 <uche> <uche> Adobe, Nokia, Philips Electronics, Fujitsu, Daimler Chrysler, Samsung,

13:56:46 <uche> <uche> Unisys, Motorola

13:56:48 <uche> <uche> ANy I missed in that list?

13:57:00 <libby> are all those companies really using rdf?

13:57:03 <uche> you're everywhere

13:57:13 <uche> Well, I told them "involved"

13:57:18 <uche> some I got fro the WG list

13:57:28 <libby> right, subtle difference :)

13:57:35 <uche> 'xcatly

13:57:59 <uche> Subtelty is at the heart of persuasion, I've learned

13:58:09 <uche> As long as it doesn't bleed into prevarication

13:58:19 <crschmidt> Adobe, Nokia, Samsung, Motorola all are in some way at least: the uaprof stuff has all those companies in it

13:58:19 <dajobe> there was a meeting in .CA recently with industrial semweb

13:58:29 <crschmidt> other than adobe, which has all the XMP stuff

13:58:55 * libby having trouble thinking of projects using massive amounts of rdf data

13:59:03 <crschmidt> Nokia uses RDF to back all their phone technical spec displays

13:59:11 <crschmidt> there was slides from ... patricks? I think

13:59:24 <libby> oh good point chris

13:59:30 <dajobe> their life blogging thingy

13:59:32 <libby> from the tech plenary

13:59:37 <crschmidt> http://lists.w3.org/Archives/Public/www-archive/2005Mar/0002.html

13:59:39 <crschmidt> slides

13:59:59 <crschmidt> seealso links from http://swig.xmlhack.com/2005/03/01/2005-03-01.html#1109715158.355800

14:00:14 <dajobe> hmm, "w3c owl, rdf, swrl, dawg" from www.networkinference.com

14:00:50 <dajobe> OK, I was thinking of http://www.semantic-conference.com/

14:02:06 <libby> might be some useful stuff here uche: http://esw.w3.org/topic/SwigAtTp2005

14:02:10 <crschmidt> isn't Jena an HP backed project? oh, hpl is HP labs.

14:02:14 <uche> Wow.

14:02:17 <uche> Thanks, guys

14:02:22 <uche> Trying to keep up :-)

14:02:35 <crschmidt> http://www.hpl.hp.com/semweb/ is HP's stuff on SemWeb

14:02:42 <dajobe> there's several US three-letter-agencies with interests, not public usually

14:02:55 * libby heads out for a bit - l8rs

14:03:21 <uche> l8r

14:03:24 <crschmidt> "At IBM we use RDF in the "Grand Central" family of intelligent search engines, of which [jCentral] which searches for Java related resources is an example." -- http://www.w3.org/TandS/QL/QL98/pp/rdfquery.html

14:04:50 <dajobe> that's a while ago, that project ended.

14:05:34 <uche> Well, at least AlphaWorks is releaseing SemWeb software

14:06:22 <crschmidt> ooh, ARQ release

14:21:38 <CaptSolo> hi hannes

14:22:39 <hg> hi uldis.

14:23:15 <dajobe> so is this meeting in 37 minutes or 1:37 ?

14:23:26 <crschmidt> 2:37

14:23:39 <crschmidt> .t Z

14:23:41 <phenny> 2005-04-20T14:23:40Z

14:23:59 <crschmidt> hm

14:24:02 <crschmidt> or am i wrong

14:24:05 <crschmidt> man, time sucks

14:25:23 <crschmidt> H1: [http://www.timeanddate.com/worldclock/fixedtime.html?month=4&day=20&year=2005&hour=17&min=0&sec=0&p1=0|1700 UTC]

14:25:23 <dc_swig> Replaced comment H1.

14:25:57 <crschmidt> not sure where the 18:00 came from, since it's the right link for 17:00

14:30:02 <CaptSolo> it's 15:29Z now

14:30:13 <CaptSolo> 15:30

14:30:22 <crschmidt> it is?

14:30:48 <crschmidt> phenny is usually pretty good at Z time, are you sure you're not being messed up by daylight savings?

14:31:46 <LotR> CEST is +0200...

14:32:36 <CaptSolo> chris: I am in GMT, i should know

14:32:45 <crschmidt> CaptSolo: You're not in BST?

14:32:54 <CaptSolo> nope

14:33:06 <CaptSolo> GMT is easier for time calculations, hence I am in GMT ;)

14:33:26 <crschmidt> "Current UTC (or GMT/Zulu)-time used: Wednesday, April 20, 2005 at 14:33:06" -- http://www.timeanddate.com/worldclock/ I'm sorry, but I beg to differ

14:33:28 <LotR> GMT doesn't exist when it is summer :)

14:33:59 <crschmidt> the meeting is at @750. it's currently @647

14:34:01 <CaptSolo> ok - UTC

14:34:21 <LotR> martijn@fangorn:~/POE-Component-Client-HTTP$ date --utc

14:34:21 <LotR> Wed Apr 20 14:35:47 UTC 2005

14:34:25 <crschmidt> if you think it is currently 15:30, you're not in UTC or GMT or Zulu or anything else :p

14:34:49 <crschmidt> And we've just spent 8 minutes arguing about time, so I'm going to go back to work. FOAF meeting is here in ~2.5 hours.

14:34:52 <LotR> BST is 15:30 now, isn't it?

14:35:00 <JibberJim> er Morocco is on GMT in the summer

14:35:10 <CaptSolo> oh my god, now we are stuck in time discussions

14:36:17 <CaptSolo> ok i'm on BST then

14:36:56 <CaptSolo> did not expect UTC not to move when the daylight saving takes effect (now the local time is not equal to UTC any more :( )

14:37:51 <LotR> that's the nice thing of UTC. no DST :)

14:38:27 <dajobe> just click on H1

14:41:43 * Pike_ actually finished his reviews, *way* early

15:10:51 <jsled>http://research.microsoft.com/research/pubs/view.aspx?type=technical+report&id=898

15:10:52 <dc_swig> L: http://research.microsoft.com/research/pubs/view.aspx?type=technical+report&id=898 from jsled

15:11:02 <jsled> L:| Semantic Streams: a Framework for Declarative Queries and Automatic Data Interpretation

15:11:03 <dc_swig> Titled item L.

15:11:25 <jsled> L: "We present a framework called Semantic Streams that allows users to pose declarative queries over semantic interpretations of sensor data."

15:11:26 <dc_swig> Added comment L1.

15:12:00 <jsled> L: from Microsoft Research.

15:12:00 <dc_swig> Added comment L2.

15:12:03 <jsled> L: [(via)|http://lambda-the-ultimate.org/node/view/654]

15:12:03 <dc_swig> Added comment L3.

16:10:34 <crschmidt> http://www.w3.org/2001/sw/ could use an update pointing to the new SPARQL draft

16:46:08 <crschmidt> Going to run acorss the street and grab lunch

16:46:11 <crschmidt> should be back in less than 10

16:46:15 <crschmidt> feel free to start without me if not :)

16:53:35 <dajobe> -- foaf chat on the hour, in 5 mins --

16:54:31 * crschmidt returns

16:54:34 <crschmidt> (short line today)

16:58:18 <crschmidt>http://rdfweb.org/pipermail/rdfweb-dev/2005-April/014043.html

16:58:18 <dc_swig> M: http://rdfweb.org/pipermail/rdfweb-dev/2005-April/014043.html from crschmidt

16:58:26 <crschmidt> M:| FOAF Meeting Announcement / Agenda

16:58:26 <dc_swig> Titled item M.

16:59:13 <crschmidt>http://swig.xmlhack.com/2005/03/16/2005-03-16.html#1110991616.043090

16:59:14 <dc_swig> N: http://swig.xmlhack.com/2005/03/16/2005-03-16.html#1110991616.043090 from crschmidt

16:59:19 <crschmidt> N:| Action Items from last meeting

16:59:19 <dc_swig> Titled item N.

16:59:44 <crschmidt> BLURB: FOAF Specification Issues

16:59:44 <dc_swig> O: FOAF Specification Issues from crschmidt

16:59:46 <mortenf> logger, chump M:

16:59:46 <mortenf> M:See [http://www.ilrt.bris.ac.uk/discovery/chatlogs/swig/2005-04-20#T16-59-46|discussion]

16:59:47 <dc_swig> Added comment M1.

16:59:48 <phenny> mortenf: 02:26Z <crschmidt> tell mortenf that redland API documentation says " If the statement already exists in the model, it is not added. Duplicate statements can be added when used with Redland Contexts such as with librdf_model_context_add_statement"

16:59:58 <mortenf> ok, thanks

17:00:23 <crschmidt> .t Z

17:00:25 <phenny> 2005-04-20T17:00:24Z

17:00:43 <crschmidt> M: [http://crschmidt.net/ Christopher Schmidt], leading

17:00:43 <dc_swig> Added comment M2.

17:00:49 <mortenf> M:attending [http://www.wasab.dk/morten/ mortenf]

17:00:50 <dc_swig> Added comment M3.

17:00:54 <crschmidt> Anyone who's here for the FOAFmeet, feel free to add your name

17:01:00 <dajobe-lap> M:Dave, eating

17:01:00 <dc_swig> Added comment M4.

17:01:05 <crschmidt> mmm, food

17:01:10 <mortenf> M3:attending: [http://www.wasab.dk/morten/ mortenf]

17:01:10 <dc_swig> Replaced comment M3.

17:01:14 <benja_> M:Benja, lurking

17:01:14 <dc_swig> Added comment M5.

17:01:34 <CaptSolo> M:attending [http://captsolo.net/info/ Uldis Bojars]

17:01:34 <dc_swig> Added comment M6.

17:01:35 <libby> M:attending: libby

17:01:36 <dc_swig> Added comment M7.

17:01:39 <jesperll> M:attending: [http://cloud.jogu.dk/ jesperll]

17:01:40 <dc_swig> Added comment M8.

17:01:43 <crschmidt> So, first agenda item, pending any argument, is discussion of previous action items

17:01:44 <kasei> M: attending [http://kasei.us/ gregory williams]

17:01:44 <dc_swig> Added comment M9.

17:01:57 * libby waves at kasei

17:02:08 <crschmidt> (Also, hi to everyone, blah blah)

17:02:10 <kasei> hi libby

17:02:12 <crschmidt> thanks for coming :)

17:02:16 <mortenf> :)

17:02:27 <crschmidt> My action from last meeting was the vocab tools

17:02:38 <crschmidt> Thankfully, i actually took some time and got them done this time!

17:03:04 <crschmidt> Rewrote dan's specgen script into Python, and it's now hosted at http://crschmidt.net/semweb/redland/ along with some of my other Redland tools

17:03:13 <crschmidt> dan is now using it to generate the FOAF spec

17:03:28 <crschmidt> so, updates to the specification should be easier now, although we are still working on better ways to do documentation and so on

17:03:45 <crschmidt> N: Completed specgen rewrite, is now being used to generate FOAF spec

17:03:45 <dc_swig> Added comment N1.

17:04:14 <crschmidt> last month, we mentioned issues with the website that Mortenf might be looking to help with: I think we were also going to wait a bit for wordpress to work out its postrelease kinks

17:04:19 <mortenf> N:mortenf's action re weblog/planet continues...

17:04:19 <dc_swig> Added comment N2.

17:04:36 <crschmidt> danbri seems confident in his email that the Foaf website will go live Real Soon Now

17:04:39 <mortenf> also, sorry to report no progress on the issue issue...

17:05:03 <crschmidt> no problems, lots of stuff going on this month

17:05:11 <mortenf> indeed

17:05:13 <crschmidt> any other continued actions that people want to bring up?

17:05:22 <CaptSolo> chris: does it process term_status?

17:05:47 <crschmidt> CaptSolo: No, I don't think it does. Dan may have written the code to do that himself: i'm not sure

17:06:18 <crschmidt> I didn't do it though :)

17:06:26 <crschmidt> Specgen script can still use some work, but been doing more interesting things

17:06:26 <bengee> M:somewhat attending: [http://www.bnode.org/ bengee]

17:06:27 <dc_swig> Added comment M10.

17:06:39 * mortenf waves to danja

17:07:05 <CaptSolo> bengee :)

17:07:08 * danja waves

17:07:09 <crschmidt> Okay, so issue status is: work continuing on specgen, foaf website nearing completion, mortenf still going through issues and sending on to mailing list (and more people always welcome)

17:07:25 <libby> danbri did talk to nicole about the site at sparqlcamp, so I think they may be some progress there

17:07:40 <crschmidt> good deal

17:07:44 <mortenf> great

17:08:03 <crschmidt> Okay, with that, we'll move on to next agenda item, unless anyone else has something to add?

17:08:12 <mortenf> sound good

17:08:25 <crschmidt> N: issue status is: work continuing on specgen, foaf website nearing completion, mortenf still going through issues and sending on to mailing list (and more people always welcome)

17:08:26 <dc_swig> Added comment N3.

17:08:35 <crschmidt> Next is discussion of changes to FOAF specification

17:08:43 <bengee> heyup CaptSolo, crowd ;)

17:09:07 <crschmidt> O: [http://rdfweb.org/pipermail/rdfweb-dev/2005-April/014044.html danbri requests] "something like foaf:isPrimaryTopicOf"

17:09:07 <dc_swig> Added comment O1.

17:09:23 <mortenf> i have outstanding action re birthday decision...

17:09:27 <crschmidt> oh, do you?

17:09:42 * crschmidt is not so good at noticing everythign :)

17:09:43 <mortenf> well, we decided something last time, need to write it up for spec

17:09:52 <crschmidt> ah, yeah

17:09:55 <mortenf> half way done methinks

17:09:57 <crschmidt> same with iand+bio

17:10:00 <mortenf> right

17:10:19 <danja> M: hereabouts: danja, http://dannyayers.com

17:10:19 <dc_swig> Added comment M11.

17:10:26 <mortenf> re isPrimaryTopicOf is like a super property of homepage/weblog

17:10:38 <kasei> i haven't written an email about it yet, but I'd like to discuss symmetry between dated terms (like pastProject) and others (like workplaceHomepage)

17:10:38 <mortenf> which could be nice

17:10:47 <crschmidt> M: mortenf also to continue "mortenf: ACTION: mortenf to complete writeup, include main alternatives in docs" from BirthdayIssue

17:10:47 <dc_swig> Added comment M12.

17:10:51 <libby> libby has changed the topic to: foaf-takeover for 60-90 mins from 1700 UTC: agenda - http://rdfweb.org/pipermail/rdfweb-dev/2005-April/014043.html -- Semantic Web Interest Group hack-n-chat - UTF-8 charset please - Weblog: http://swig.xmlhack.com/ - Logs: http://ilrt.org/discovery/chatlogs/swig/ - please identify with NickServ - FOAF meeting 2005-04-20T17:00Z

17:10:53 <mortenf> also re atom, which only has a "uri"

17:10:57 <CaptSolo> M6:attending [Uldis Bojars|http://captsolo.net/info/]

17:10:57 <dc_swig> Replaced comment M6.

17:11:05 <libby> heh, sorry, didn;t realise you'd changed topic already chris

17:11:10 <crschmidt> heh, heh

17:11:14 <crschmidt> that was dajobe actually :)

17:11:42 <kasei> i think either the description of workplaceHomepage needs to change, or it needs a past equivalent.

17:11:53 <crschmidt> kasei: okay, let's start with primaryTopicOf discussion

17:11:59 <kasei> absolutely

17:12:07 <crschmidt> mortenf mentioned that it's something like a superproperty of homepage/weblog

17:12:15 <CaptSolo> fix how you write links here - otherwise we get broken links on the chump page

17:12:25 <libby> the past* stuff never got used much or stopped being unstable

17:12:36 <mortenf> M3:attending: [http://www.wasab.dk/morten/|mortenf]

17:12:37 <dc_swig> Replaced comment M3.

17:12:40 <crschmidt> oops, captsolo, thanks

17:12:47 * CaptSolo got not found on http://rdfweb.org/pipermail/rdfweb-dev/2005-April/014044.html%20danbri%20requests otherwise

17:13:06 <jesperll> M8:attending: [jesperll|http://cloud.jogu.dk/]

17:13:07 <dc_swig> Replaced comment M8.

17:13:11 <crschmidt> O: [http://rdfweb.org/pipermail/rdfweb-dev/2005-April/014044.html|danbri requests] "something like foaf:isPrimaryTopicOf"

17:13:11 <dc_swig> Added comment O2.

17:13:20 <danja> what's the domain/range for primaryTopicOf?

17:13:20 <crschmidt> O1:""

17:13:20 <dc_swig> Deleted comment O1.

17:13:37 <kasei> M9:attending: [http://kasei.us/|gregory williams]

17:13:37 <dc_swig> Replaced comment M9.

17:13:48 <mortenf> i'd say Resource/foaf:Document

17:14:06 <crschmidt> ^range foaf:topic

17:14:08 <bengee> M10:somewhat attending: [http://www.bnode.org/|bengee]

17:14:09 <dc_swig> Replaced comment M10.

17:14:10 <crschmidt> ^rangeOf foaf:topic

17:14:11 <julie> rdfs:Resource

17:14:17 <crschmidt> ^domainOf foaf:topic

17:14:18 <julie> foaf:Document, rdfs:Document

17:14:26 <crschmidt> So, I'd say that we're talking about the inverse of that

17:14:31 <mortenf> oops, where'd rdfs:Document come from

17:14:42 <crschmidt> who knows? damn lack of provenance

17:14:49 <crschmidt> i think there was a testing document that got in there somehow

17:14:53 <mortenf> ;)

17:15:06 <crschmidt> whoever wrote that bot should be shot, seriously

17:15:32 <danja> what? no RDF schema validator?!

17:15:47 <mortenf> anyway, i think it may be a good idea, however would encourage lax usage, instead of more specific homepage/weblog

17:16:20 <crschmidt> hm, possibly declare the Domain disjoint with foaf:Agent?

17:16:34 <crschmidt> It's basically a substitute of foaf:homepage for nonAgents, right?

17:16:53 <crschmidt> Although I suppose the wikipedia page for Benedict XVI isn't really his "homepage" per se, but he's still an Agent

17:16:58 <mortenf> hmm, possible, but i don't see why

17:17:01 <mortenf> right

17:17:20 <benja_> I think primary topic is much wider than homepage

17:17:28 <mortenf> agreed

17:17:31 <benja_> (wikipedia pages being a good example)

17:17:56 <danja> yep

17:18:17 <crschmidt> I suppose warnings in the documentation can serve to inform people as to best use

17:18:32 <crschmidt> "use homepage/weblog if it's appropriate: this is a superclass for those properties, they are more specific"

17:18:45 <mortenf> right

17:18:45 <benja_> is that fair for weblog?

17:19:01 <mortenf> so, let's ask danbri to write it up as an issue :)

17:19:43 <benja_> it's certainly fair for homepage

17:19:59 <crschmidt> O: General opinion is that primaryTopicOf is a good property, with ample warning that more specific properties like foaf:homepage are used where appropriate.

17:20:00 <dc_swig> Added comment O2.

17:20:34 <crschmidt> O: property would have Domain/Range of Resource/Document

17:20:34 <dc_swig> Added comment O3.

17:20:42 <crschmidt> O: ACTION danbri to write up as issue :)

17:20:42 <dc_swig> Added comment O4.

17:20:52 <crschmidt> benja_: can you explain why weblog wouldn't be?

17:20:54 <mortenf> O:primaryTopicOf could also be used for atom integration

17:20:55 <dc_swig> Added comment O5.

17:21:25 <danja> mortenf, how? (in brief)

17:21:25 <benja_> well, is the person writing the weblog really the topic of the weblog? -- I'm thinking in particular of blogs focussing on one topic

17:21:46 <crschmidt> benja_: not neccesarily, but it generally does have a specific topic

17:21:48 <mortenf> the author/uri is exactly that

17:21:56 <danja> ah

17:22:13 <mortenf> good point

17:22:40 <crschmidt> so, I could say that <http://w3.org/2001/sw/> foaf:isPrimaryTopicOf <http://planetrdf.org/>

17:23:13 <benja_> crschmidt, yes, sure

17:23:16 <crschmidt> whereas if planetrdf was a person, it might be wiser to do it the other way around: <http://crschmidt.net/foaf.rdf#crschmidt> foaf:weblog <http://crschmidt.livejournal.com/>

17:23:26 <crschmidt> it's not always going to be a simple reversal

17:23:40 <benja_> but if you had a weblog about semantic web stuff, it would be your foaf:weblog

17:23:46 <benja_> but its primary topic wouldn't be you

17:23:54 <crschmidt> hm.

17:23:57 <benja_> so I think foaf:weblog shouldn't be a sub-property

17:23:58 * bengee doesn't think that the primary topic of planet rdf is the rdf page..

17:24:05 <benja_> heh

17:24:08 * mortenf agrees

17:24:19 <mortenf> ... re foaf:weblog

17:24:23 <crschmidt> bengee: I was using it as a concept in that sense

17:24:23 <benja_> [ foaf:isPrimaryTopicOf <http://w3.org/2001/sw/>

17:24:31 <benja_> , <http://planetrdf.org/> ]

17:24:34 <bengee> shame on you

17:24:36 <bengee> ;)

17:24:55 <crschmidt> bengee: I shun your #!

17:24:57 <benja_> nice example of why you'd like isPrimaryTopicOf :)

17:25:00 <crschmidt> Anyway :)

17:25:21 <crschmidt> benja_: I agree, you're right that foaf:weblog isn't a subclass

17:25:25 <benja_> ok

17:25:28 <crschmidt> er, subproperty

17:25:30 <crschmidt> whatever

17:25:46 <crschmidt> but we all agree foaf:homepage is?

17:25:52 <benja_> I do

17:26:11 <crschmidt> any opposed?

17:26:17 <mortenf> i think i agree

17:26:26 <kasei> i agree

17:26:29 <crschmidt> heh, heh

17:26:39 * jesperll can't think of a reason not to agree

17:26:57 <crschmidt> close enough! I'll send a summary at some point later today to the list: any disagreements on it can be brought up then (and probably before danbri gets around to it ;))

17:27:04 <kasei> i think we call a lot of things "weblogs" that aren't approriate for use with foaf:weblog.

17:27:08 * danja strokes cat

17:27:16 <kasei> heh

17:27:44 <crschmidt> kasei: it depends what foaf:weblog actually means, I suppose

17:27:54 <mortenf> O:foaf:homepage could be rdfs:subPropertyOf foaf:isPrimaryTopicOf

17:27:55 <dc_swig> Added comment O6.

17:28:04 <crschmidt> kasei: if foaf:weblog means "This is someplace where I write", I think it's used properly most of the time

17:28:07 <CaptSolo> chris: you took words out of my mouth :)

17:28:12 <kasei> yeah, but since it's an ifp, it clearly doesn't count pages that I'd consider weblogs... things like that.

17:28:31 <crschmidt> kasei: if foaf:weblog means "the topic of the object is the subject", it's probably not

17:28:44 <benja_> kasei: I think it's most useful to apply it to any blog that belongs to one specific person/group...

17:28:46 <crschmidt> i'm certainly not the subject of crs.net/blog, but i am the topic of crs.lj.com

17:28:46 * mortenf fails to parse that

17:29:11 <crschmidt> hm, i'm mixing terminology, which doesn't help :)

17:29:21 <jesperll> mortenf: as long as you don't segfault

17:29:27 <mortenf> right :)

17:29:36 <kasei> right, but when you're talking about the inverse, and using examples like planetrdf, things get weird, because that isn't a fixed "group"

17:29:43 <CaptSolo> chris: but crs.net/blog is your weblog, no?

17:30:10 <crschmidt> CaptSolo: Right. So what does foaf:weblog actually mean?

17:30:20 <crschmidt> The blog is sompleace I write, or the blog is about me?

17:30:47 <CaptSolo> there is a subclass of weblogs (weblogs in general sense) which are online diaries (where people write about themselves, their life, ...)

17:30:52 <crschmidt> M2:[http://crschmidt.net/|Christopher Schmidt]

17:30:53 <dc_swig> Replaced comment M2.

17:30:53 <benja_> I think 'is about me' would limit it far too much

17:31:09 <CaptSolo> now primarySubjectOf online diaries is the person

17:31:14 <CaptSolo> with high probability

17:31:23 <CaptSolo> not of weblogs in general

17:31:30 <jesperll> how about a foaf:ramblings prop? :)

17:31:39 <CaptSolo> jesper :)

17:31:49 <crschmidt> heh, heh. I already call it my ramblings via the dc:title ;)

17:31:50 <CaptSolo> chris: weblog is somewhere you write

17:32:01 <kasei> somewhere you (and only you) write

17:32:02 <CaptSolo> [ with basic assumption that you are a sole author of this]

17:32:11 <CaptSolo> [ although some blogs have multiple authors ]

17:32:15 <crschmidt> So, I think that we're proving that foaf:weblog is definitely not a a subPropertyOf primaryTopicOf

17:32:22 <CaptSolo> [ mine do, but 99% of posts are by me anyway ]

17:32:23 <jesperll> so more than one person isn't allowed to use foaf:weblog to the same blog?

17:32:26 <benja_> [ CaptSolo's blog being a particularly thorny issue on that subject ]

17:32:31 <mortenf> jesperll: right

17:32:43 <crschmidt> jesperll: correct, you should make [a foaf:Group] then

17:32:48 <jesperll> right

17:33:07 <crschmidt> If you do have multiple people with the same foaf:weblog, the go SMUSH

17:33:10 <crschmidt> when i run mortenf's code

17:33:11 <CaptSolo> jesperll: but you may probably say that it is a foaf:weblog of foaf:Group containing all the authors

17:33:11 <crschmidt> :)

17:33:34 <crschmidt> That happened to the NetalleyNetworks folks back when i first started using julie

17:33:43 <CaptSolo> benja: quite a lot of weblog software allow multiple authors

17:34:08 <crschmidt> for people who want non IFP weblogs, see http://www.hackcraft.net/FOAFNonIFP/

17:34:16 <benja_> heh

17:35:04 * crschmidt thinks we've probably rambled long enough on this, perhaps continue later

17:35:13 <crschmidt> i'd like to move on to kasei's mention of the past* properties in FOAF

17:35:53 <mortenf> k.

17:35:55 <CaptSolo> on the theme of weblogs - now aggregators are quite widespread

17:35:59 <CaptSolo> online aggregators

17:36:12 <kasei> right. i'd like to be able to use things like workplaceHomepage to discuss past workplaces, but the spec doesn't help here.

17:36:21 <CaptSolo> doesn't it mean we need <Aggregator> in foaf?

17:36:28 <CaptSolo> listing all the members / feeds

17:36:36 <kasei> there is no pastWorkplaceHomepage, and the defn. of workpaceHomepage seem to suggest current workplace.

17:36:58 <crschmidt> kasei: agreed

17:37:00 <libby> teh past* stuff is pretty icky - I'd avoid adding to it

17:37:11 <crschmidt> is there a better solution?

17:37:24 <kasei> well, we need *some* way to discuss that sort of thing :)

17:37:26 * mortenf points to http://xml.mfd-consult.dk/foaf/explorer/?foaf=http://www.wasab.dk/morten/blog/feed/foaf re groups/multiweblogs

17:37:29 <crschmidt> I suppose you can always turn employer into a class

17:37:32 <benja_> libby, why? (except that it perhaps isn't used much)

17:37:44 * mortenf likes the the past properties - they never change...

17:37:53 <crschmidt> heh, heh

17:37:54 <libby> true enough

17:38:28 <jesperll> wouldn't it be nice to not only say that you used to work someplace, but also when?

17:38:31 <libby> dunno, it just seems horrible encoding the temporal stuff in teh property. mind you lots of other ways are eqwually horrible

17:38:40 <crschmidt> jesperll: that's what I was just about to say

17:38:44 <libby> in that case could use bio"event type contruscts

17:38:45 <jesperll> :)

17:38:51 <mortenf> the when part can be done by the whois vocab

17:39:17 <libby> yeah I think doing that is more useful, although more verbose

17:39:19 <libby> whois?

17:39:28 <kasei> if you don't encode it in the property, queries for both current and past workplaces become very complex.

17:39:32 <mortenf> see http://www.kanzaki.com/ns/whois

17:39:35 <jesperll> libby: knock knock?

17:39:40 <mortenf> lol

17:40:08 <crschmidt> kasei: well, can't you just say <job:employed><job:Employment><dc:date>past</></></>

17:40:38 <benja_> crschmidt, due to the flexibility of dc:date? ;-)

17:41:03 <crschmidt> [a foaf:Person; job:employed [ dc:date "past"; dc:description "I used to work here"; job:homepage <http://www.wedu.com>]].

17:41:12 <CaptSolo> well - recommended for dc:date is ISO 8601

17:41:20 <crschmidt> yeah, yeah, it was an example :p

17:41:21 <jesperll> crschmidt: is that begin or end date?

17:41:29 <kasei> sorry... missing something. is job: related in any way to whois?

17:41:30 <benja_> jesperll: both ;)

17:41:33 <benja_> sorry

17:41:37 <crschmidt> jesperll: okay, so not dc:date, but a job:start, job:end?

17:41:53 <crschmidt> Anyway, I think it's something that could receive enough attention outside of FOAF to possibly be pulled out

17:42:04 * CaptSolo needs to look at resume rdf vocab - how it was there

17:42:10 <crschmidt> deprecating the whole "workplaceHomepage" (which is kind of clunky anyway: it doesn't let you describe a job)

17:42:14 <CaptSolo> [in the long to-do list pile]

17:42:42 <crschmidt> I can't say "I was working at UIUC for 2 years when I was there" in FOAF, so I think it's something that'd be appropriate for an extension

17:42:45 <crschmidt> Thoughts?

17:42:47 <benja_> -- hm, instead of job:homepage, I would much prefer job:employer [ foaf:homepage ... ]

17:42:57 <benja_> but of course that's details then

17:42:57 <kasei> in that case, shouldn't pastProject be deprecated as well? my big issue here is the lack of consistency.

17:43:12 <libby> I think this stuff would be better in a CV vocab like the one captsolo did

17:43:18 <libby> yeah I think so kasei

17:43:23 <crschmidt> kasei: yep, I think so

17:43:25 <CaptSolo> ok - re. deprecation of workplaceHomepage

17:43:26 <mortenf> example of whois usage: http://xml.mfd-consult.dk/foaf/explorer/?foaf=http://xml.mfd-consult.dk/foaf/morten-history.rdf

17:43:29 <danja> +1

17:43:32 <libby> but workplacehomapge is very useful: I'd argue for kkeeping that

17:43:34 <CaptSolo> though it is a simple way to say where you are working now

17:43:41 <kasei> yes, i would as well

17:43:46 <CaptSolo> and lets you easily find all people who are working there now

17:43:48 <libby> no, no depreciation of workplacehomapage - it's too useful

17:43:50 <mortenf> includes previous jobs

17:43:54 <crschmidt> ah, okay

17:44:05 <libby> keep wphp, deprecate past*

17:44:10 <crschmidt> Okay, so workplaceHomepage and currentProject could stay, as "now" props?

17:44:16 <libby> look at putting some more work into CV vocab

17:44:24 <crschmidt> and then go into more extensive stuff for more description fromo CV vocab, etc.

17:44:26 <danja> sounds a good compromise

17:44:29 <mortenf> now sucks

17:44:33 <libby> currentProject shoudl probably be something like projectHomepage

17:44:37 <danja> ;-)

17:44:49 <libby> for consistency

17:45:18 * crschmidt nods

17:45:34 <crschmidt> O: Discussion of past* properties vs. current* properties, started by kasei

17:45:35 <dc_swig> Added comment O7.

17:45:51 <crschmidt> O: Can't say "I used to work here" like you can for projects, but it's hard to say much about it anyway - no when, no information, etc.

17:45:51 <benja_> if it has to change, couldn't it change to foaf:project [ foaf:homepage ... ]?

17:45:52 <dc_swig> Added comment O8.

17:45:57 <CaptSolo> libby: yep, CV needs more work added

17:46:30 <CaptSolo> plus see how it relates to whois, bio, ...

17:46:32 <crschmidt> O: discussion of using [http://www.kanzaki.com/ns/whois#|whois] namespace alongside CV namespace for describing when you used to something

17:46:32 <dc_swig> Added comment O9.

17:46:54 <crschmidt> O: And keeping the workplaceHomepage prop meaning "where I work now"

17:46:54 <dc_swig> Added comment O10.

17:47:17 <crschmidt> benja_: well, part of the point of having the "current" stuff is to make it easy to query against

17:47:29 <benja_> crschmidt, hum

17:47:31 <libby> workplacehompage is a shortcut property and projecthompage would be like that

17:47:52 <libby> in these days of tags, shortcuts are useful I think

17:48:01 <benja_> libby, shortcut for? (there is no foaf:workplace?)

17:48:11 <libby> it's implicit :)

17:48:15 <benja_> gnn =)

17:48:22 <crschmidt> benja_: well, you would need one to do the extension

17:48:36 <libby> foaf has never been terribly consistent ... :/

17:48:46 <crschmidt> too long of a development time :)

17:49:03 * crschmidt has an inconsistent Redland wrapper right now, and he wrote it in 3 days, damn changing my mind as I go

17:49:13 <jesperll> :)

17:49:21 <CaptSolo> there is inconsistency between having current and getting rid of past properties

17:49:25 <libby> I actually think spending a lot of time making it consistent isn;t really worth the effort. howvere pastproject is ugly, and shoudl be removed imo. I added it, so... :)

17:49:26 <mortenf> i like the wording in foaf:mbox better

17:49:32 <mortenf> CaptSolo: exactly

17:49:40 <CaptSolo> as thinkgs like LinkedIn have 'past workplaces'

17:49:52 <libby> yep, I think we have to assume that everythign is 'current'

17:50:01 <libby> without explicitly saying so

17:50:08 * benja_ would like to follow the workplace/project link and see summary information about the workplace/project, rather than the homepage URI =)

17:50:14 <mortenf> or, as for mbox, "at all points in time"

17:50:23 <libby> yeah

17:50:27 <libby> that's weirder

17:50:31 <libby> you're working there...forever!

17:50:38 <benja_> heh

17:50:40 <mortenf> yeah, but works better for queries

17:50:40 <kasei> haha

17:50:43 <mortenf> :)

17:50:44 <libby> foaf has no memory I guess

17:50:44 <CaptSolo> actually they have 'workplaces' which are current or past depending on their end date

17:50:44 <CaptSolo> i do not know if LinkedIn would be exporting their data

17:50:44 <CaptSolo> probably not

17:51:30 <libby> anyone care to summarise? propose something?

17:51:31 <kasei> anyway, it's not like we'd be getting rid of the possibility. only moving it out to other vocabs that can do a better job at it.

17:51:32 <crschmidt> I'm in favor of workplaceHomepage meaning "now", having a foaf:workplace allowing you to point to more info, and doing the same for project

17:51:39 * bengee would like to be able to easily link a person to projects (i.e. a person-centric property)

17:51:52 <CaptSolo> so... do we get rid of both 'past' and 'current'? and just have projects/workplaces for eternity?

17:51:54 <crschmidt> so, you have workplaceHomepage, workplace, projectHomepage, project

17:52:26 <libby> anyone volunteer to write a proposal?

17:52:32 <kasei> what would the range of workplace be?

17:52:33 <crschmidt> I guess the question is "does workplaceHomepage mean now, or always"?

17:52:37 <kasei> Organization?

17:52:51 <crschmidt> kasei: resource, or whatever the new job/CV thing came up with

17:52:57 <kasei> the way it's worded now, it's "now"

17:53:03 <kasei> heh

17:53:04 <kasei> ok

17:53:20 <crschmidt> basically, it becomes a container for all kinds of info about the job

17:53:45 <crschmidt> Personally, I'd like to have a real easy way to say 'who do I work with right now?'

17:53:54 <libby> perhaps captsolo would like to resurrect the CV vocab - send it around foaf mailing list perhaps

17:53:57 <libby> _1 chris

17:54:01 <libby> +1 even

17:54:09 <libby> that's what it's there for really

17:54:20 <crschmidt> so, I'd like for workplaceHomepage to mean "now", same with projectHomepage

17:54:31 <kasei> agreed

17:54:33 <jesperll> +1

17:54:34 <crschmidt> and if you want that other information, you can do all the hard stuff like determining dates, doing comparisons, all that other crap

17:54:34 <CaptSolo> libby: i'll look at the CV stuff in next couple of weeks and can tell more then

17:54:34 * mortenf doesn't like statements that will become untrue at some point

17:54:53 <libby> in a sense if we drop past* 'now' isn;t really a problem

17:54:58 <libby> thanks capt :)

17:55:03 <CaptSolo> libby - yes, will do

17:55:19 <CaptSolo> [ it's all the time in todo list, just more urget things pushed it back :( ]

17:55:23 <mortenf> yep, sounds good CaptSolo

17:55:38 <libby> mortenf, yeah, I know what you mean....but there are several like that in foaf

17:55:41 <crschmidt> mortenf: well, we're kind of maintaining the status quo in that respect :)

17:55:45 <libby> names changign etc...

17:55:49 <mortenf> true...

17:56:02 <kasei> and a lot of triples *might* become untrue at some point.

17:56:08 <crschmidt> mortenf: If people don't want to do that, you can always just use the workplace property

17:56:13 <crschmidt> and not use the shortcut

17:56:15 <kasei> (what libby said)

17:56:28 <CaptSolo> so what is the summary re. this issue?

17:57:05 <kasei> deprecate, and work on getting support for the CV stuff?

17:57:27 <crschmidt> O: Discussion leads to the idea of changing around currentProject, workplaceHomepage to be more consistent, changing to props like workplaceHomepage, meaning "now", and a seperate workplace property, pointing to a collection of information about it, possibly including dates when you worked there

17:57:27 <dc_swig> Added comment O11.

17:57:31 <libby> deprecate past* yeah, and what kasei said

17:57:35 <kasei> right

17:57:37 * mortenf hopes someone will write up a coherent proposal

17:57:44 <libby> yeah

17:57:46 <crschmidt> kasei: can you action to write that to the mailing list?

17:58:00 <kasei> sure

17:58:03 <crschmidt> (never bring up issues around me, it'll just bring work down upon your shoulders... mwhaha)

17:58:06 <crschmidt> ahem.

17:58:21 <crschmidt> O: ACTION kasei to write up proposal to mailing list

17:58:21 <dc_swig> Added comment O12.

17:58:29 <crschmidt> okay! now that we've dealt with that!

17:58:32 <libby> cheers kasei :)

17:58:38 <mortenf> go kasei!

17:58:40 <jesperll> weeee

17:58:45 <kasei> heh

17:59:10 <crschmidt> any more issues with the spec for today, or can we move over to the more fun stuff?

17:59:19 <libby> that's 60 mins...when you guys planning on windoing up?

17:59:25 * crschmidt chants fun stuff, fun stuff

17:59:37 <mortenf> after 89 minutes? :)

17:59:37 <crschmidt> another 30 is about all i can spare

17:59:45 <crschmidt> heh, heh

17:59:51 <libby> yeah, 89 sounds good

17:59:56 <libby> or 90

17:59:58 <crschmidt> heh, heh

18:00:01 <jesperll> West Wing is on in 35

18:00:03 <libby> so what's this fun stuff then?

18:00:15 <crschmidt> since we have so many people here, was anyone at the whole asemantics thing and wants to share?

18:00:16 * mortenf is now watching cup ball as well

18:00:19 <crschmidt> or are you all sworn to silence?

18:00:26 <libby> ch4 news for me

18:00:33 <mortenf> oh?

18:00:42 <mortenf> oh, read that s wh4 ;)

18:00:58 <crschmidt> [a foaf:Person; foaf:nick "libby"; menow:hasStatus [a menow:Status; menow:watching "Channel 4 News"]].

18:01:19 * libby has fun stuff, sorta... anyone fancy linking their latest photos as in http://libby.asemantics.com/2005/04/pigsty/ form foaf file?

18:01:25 <mortenf> when was/is that true crschmidt? ;)

18:01:29 * benja_ read that as meow:

18:01:45 <crschmidt> mortenf: typically, you'd attach a date to the Status, and assume it's true for that second

18:01:49 <libby> heeh heh

18:02:01 * crschmidt looks at pigsty

18:02:04 * mortenf can add stuff like that

18:02:11 <crschmidt> BLURB:Fun Stuff

18:02:11 <dc_swig> P: Fun Stuff from crschmidt

18:02:14 <libby> not sure re 'gallery-3'

18:02:15 <kasei> libby: looks interesting. i'll definitely check it out.

18:02:49 <libby> it's a bit flakey. but flakey pihgsty asside, I think linking picture galleries from foaf would be nice

18:03:06 <libby> thanks guys :)

18:03:06 <benja_> is there really a good reason to say that a region of an image is not an image?

18:03:10 <mortenf> agreed

18:03:13 <libby> I'll change the class if anyone has better ideas

18:03:21 <mortenf> benja_: see archives

18:03:35 <crschmidt> P: Libby asks that we all link our photo galleries from our FOAF files [http://libby.asemantics.com/2005/04/pigsty/|a la pigsty-style]

18:03:35 <dc_swig> Added comment P1.

18:03:38 <benja_> mortenf, is there discussion? I saw only a mention of it

18:03:53 <mortenf> it was on the imgreg list

18:04:05 <mortenf> might have been some discussion in irc

18:04:06 <benja_> hmm, that's not the list linked from the issue ;)

18:04:18 <libby> it's on www-archive womehwere benja_

18:04:25 <mortenf> which issue?

18:04:31 <libby> what else have people been up to then?

18:04:32 <crschmidt> While we're on fun stuff, I've been playing with SPARQL for julie:

18:04:45 <crschmidt> P: [http://crschmidt.net/julie/sparql|sparql for julie]

18:04:45 <dc_swig> Added comment P2.

18:04:51 <deltab> Channel 4 News: http://imdb.com/title/tt0239164/

18:04:51 * mortenf is still working on sparql and sparqlingnaut

18:05:12 <benja_> ImagePartsIssue

18:05:14 <CaptSolo> the Apprentice for me at 9pm

18:05:14 * crschmidt will be back in 10-15

18:05:25 <crschmidt> (quick run out for work)

18:05:39 <libby> nice mortenf

18:05:45 <mortenf> anyway, part of the reason was html/browser support

18:05:48 <libby> cool chris

18:06:32 <libby> no more fun stuff ? :(

18:06:34 <CaptSolo> pigsty looks interestying

18:06:37 * bengee working on a query builder

18:06:45 <mortenf> next meeting and possibly meetup?

18:06:52 <libby> what sort of thing bengee?

18:06:54 <libby> yeah

18:06:55 <CaptSolo> anything more about the rdf4food?

18:07:03 * libby sense fading and tv watching going on

18:07:14 <mortenf> bengee: your stuff is looking really nice

18:07:20 <libby> for those who haven;t seen, have alook at gargonza.org

18:07:35 <kasei> and did anyone have anything on possible foafmeet later in summer?

18:07:43 <bengee> libby, something that's more user-friendly than a "enter a full sparql query here"-form

18:07:53 <bengee> thanx, morten! :)

18:08:02 <libby> P:[http://gargonza.org|the gargonza experiment] - a community coming out of rdf4food, not quite set up yet though

18:08:03 <dc_swig> Added comment P3.

18:08:15 <libby> ah gotya bengee. interesting

18:08:27 <bengee> still stuck on a sparql parser atm..

18:08:41 <mortenf> a php parser should be coming out sooonish

18:09:01 * libby tried (and failed) to do something like that, giving you a set of starting points form a database and then filtering down the possible options as you clicked on one, building the query as you went

18:09:03 <bengee> ah, you're doing one, too?

18:09:14 <mortenf> no, just waiting for alberto et al

18:09:18 <libby> alberto's already done a php5 parser

18:09:27 <libby> but the sparql syntax is the old syntax

18:09:35 * bengee 's working on turtly one

18:09:36 <mortenf> right

18:09:39 <libby> chris bizer is looking to do something I think

18:09:42 <libby> cool :)

18:09:55 <mortenf> heh, that's three, in php

18:09:59 <libby> guys, it'd be great if you joined the gargonza thing

18:10:25 <mortenf> indeed, i think chris b is already on board

18:10:29 <libby> otherwise there's a fair bit of duplication! and we never intended to exclude people who were interested - just cos we ran out of space...

18:10:36 * bengee will do after he has delivered the deri portal beast...

18:10:45 <mortenf> great

18:11:06 * CaptSolo has been playing with SIOC (http://www.rdfs.org/sioc/) stuff

18:11:07 * mortenf will be starting work on sparl-rewriting-to-sql

18:11:20 * crschmidt returns

18:11:30 <libby> P:re gargonza thing: so mail us if you'd like to join - there's now a mailing list which will have an archive, and we can give people access to the site using svn. the wiki's opento all of course

18:11:31 <dc_swig> Added comment P4.

18:11:33 <bengee> your query builer idea sound nice, libby. mine will be limited to classes and props form the underlying ontology, which will make it easier to generate

18:11:40 <libby> right

18:11:40 <bengee> s/sound/sounds/

18:11:45 <mortenf> so, meet and meetup?

18:11:46 <libby> it's more sensible taht way bengee

18:11:53 <libby> +1 mortenf

18:12:05 <crschmidt> oh yeah, forgot about that one

18:12:42 <mortenf> what's this, third wednesday?

18:12:47 <crschmidt> yep

18:13:07 <crschmidt> I think we should just stick to same bat time, same bat channel

18:13:11 <crschmidt> it's worked for the past couple months

18:13:17 <mortenf> right

18:13:22 <mortenf> so may 18th@17Z?

18:13:26 <crschmidt> yep

18:13:43 <CaptSolo> +1

18:13:55 <CaptSolo> what about the face2face meet?

18:13:59 <mortenf> anyone counter?

18:14:02 <crschmidt> BLURB:Next FOAF Meeting May 18th@17Z

18:14:02 <dc_swig> Q: Next FOAF Meeting May 18th@17Z from crschmidt

18:14:12 <CaptSolo> many already met in the rdf4food

18:14:23 <CaptSolo> but it's not so big as all foaf crowd

18:14:30 <mortenf> right

18:14:33 <crschmidt> Q: Same bat time, same bat channel - third wednesday

18:14:33 <dc_swig> Added comment Q1.

18:14:41 <CaptSolo> i don't know if there are any activities re. foaf workshop

18:14:47 <CaptSolo> danbri may have some idea

18:14:59 <crschmidt>http://swig.xmlhack.com/2005/03/16/2005-03-16.html#1110991876.689249

18:15:00 <dc_swig> R: http://swig.xmlhack.com/2005/03/16/2005-03-16.html#1110991876.689249 from crschmidt

18:15:01 <libby> not much progress on workshop front I don;t think

18:15:04 <CaptSolo> but we can see what about foaf camp or similar stuff

18:15:15 <crschmidt> R:| Possible Workshop/Meetup Idesas from previous FOAF meeting

18:15:15 <dc_swig> Titled item R.

18:15:37 <crschmidt> I have no idea where I'm going to be in more than a months time

18:15:48 * crschmidt is moving soon, not sure about employment, etc.

18:16:07 <dajobe-lap> where you going?

18:16:07 <mortenf> Q:[http://www.timeanddate.com/worldclock/fixedtime.html?month=5&day=18&year=2005&hour=17&min=0&sec=0&p1=0|May 18th @ 17Z in your tz]

18:16:08 <crschmidt> Although I'm obviously in favor of a gathering in the US, even though all you silly FOAF people seem to be in the other side of the atlantic

18:16:08 <dc_swig> Added comment Q2.

18:16:18 <kasei> crschmidt: sounds like a perfect chance to bum around and meet up with some foaf people! :D

18:18:30 <CaptSolo> so not much progress on foaf-meet

18:18:38 <CaptSolo> who is gonna be in Crete for ESCW?

18:18:40 <mortenf> right

18:18:47 <libby> me

18:18:57 <CaptSolo> me

18:19:00 <libby> xtech/ams anyone?

18:19:07 <CaptSolo> next chance to meet up i guess

18:19:12 <CaptSolo> no xtech :(

18:19:19 <dajobe-lap> xtech yes

18:19:59 <ocelma> Hi! am a bit late (but just finish this demo right now!).

18:19:59 <ocelma> Re FUN STUFF, I've been playing a bit, doing music recommendations from FOAF profiles

18:19:59 <ocelma> [ http://mtg127/simac/user_profiles ] (you might find a few bugs there, then :-)).

18:19:59 <ocelma> Tests were done based on livejournal profiles.

18:20:01 <DanC> ^travel DanC

18:20:03 <julie> Boston 2005-03-01, Minneapolis 2005-03-07, Amsterdam 2005-05-24, Washington 2005-04-26, Cambridge 2005-06-14, Edinburgh 2005-09-20, Boston NaN-02-26, Boston NaN-NaN-01

18:20:13 <danja> re. fun stuff

18:20:20 <crschmidt> ^addturtle [a caltz:Vevent; caltz:dtstart "2005-05-18T13:00:00"^^<http://www.w3.org/2002/12/cal/tzd/America/New_York#tz>; caltz:dtend "2005-05-18T14:30:00"^^http://www.w3.org/2002/12/cal/tzd/America/New_York#tz>; caltz:summary "FOAF Meeting"].

18:20:21 <julie> Adding Turtle failed (The namespace prefix in "http:" was not declared.).

18:20:26 <crschmidt> hm

18:20:35 <crschmidt> apparently I don't know how to do datatypes in turtle

18:20:39 <DanC> 00"^^ht

18:20:41 <dajobe-lap> missing <

18:20:45 <DanC> should be: 00"^^<ht

18:20:53 <danja>http://pragmatron.org/docs/sparqlsphere.html

18:20:54 <dc_swig> S: http://pragmatron.org/docs/sparqlsphere.html from danja

18:20:54 <crschmidt> ^addturtle [a caltz:Vevent; caltz:dtstart "2005-05-18T13:00:00"^^<http://www.w3.org/2002/12/cal/tzd/America/New_York#tz>; caltz:dtend "2005-05-18T14:30:00"^^<http://www.w3.org/2002/12/cal/tzd/America/New_York#tz>; caltz:summary "FOAF Meeting"].

18:20:55 <julie> Model size increased by 4 to 2145724 via turtle statements.

18:20:59 <danja> bum

18:21:02 <danja> ah well

18:21:31 <danja> S: | SparqlSphere - formerly known as SparqlyRssThing

18:21:31 <dc_swig> Added comment S1.

18:22:05 <crschmidt> ^q select ?start ?end where (?e rdf:type caltz:Vevent) (?e caltz:summary "FOAF Meeting") (?e caltz:dtstart ?start) (?e caltz:dtend ?end)

18:22:08 <julie> 2005-05-18T13:00:00^^<http://www.w3.org/2002/12/cal/tzd/America/New_York#tz> 2005-05-18T14:30:00^^<http://www.w3.org/2002/12/cal/tzd/America/New_York#tz>

18:22:17 <libby> danja, write a doap for it?

18:22:21 <DanC> ^q select ?pg where (?who foaf:nick "DanC") (?e ?caltz:url ?pg) (?e cyc:socialParticipants ?who)

18:22:21 <julie> Bad query (RDQL syntax error at ':').

18:22:24 <danja> will do

18:22:35 <crschmidt> no ? before caltz in second triple

18:22:42 <DanC> ^q select ?pg where (?who foaf:nick "DanC") (?e caltz:url ?pg) (?e cyc:socialParticipants ?who)

18:22:42 <crschmidt> anyway, i'm distracting matters

18:22:42 <julie> http://www.w3.org/2004/12/allgroupoverview.html, http://www.w3.org/2001/tag/ (x2), http://www.w3.org/2004/12/rules-ws/

18:22:56 * DanC wonders why xtech hp isn't there

18:23:04 * kasei heads off for a bit. good meeting!

18:23:11 <mortenf> see ya

18:23:16 <jesperll> ta

18:23:21 <danja> bye

18:23:26 <DanC> aha... missing <b> on my homepage...

18:23:34 <crschmidt> okay, we've achieved a bunch, and chatted some, and seems like people are just going to talk and so on now

18:23:41 <crschmidt> so, any objections to me closing meeting and dispersing?

18:23:52 * danja already dispersed

18:24:05 <mortenf> fine

18:24:09 <crschmidt> With that, I declare this the end of the April FOAF Meeting. Go forth, and FOAF! Or something.

18:24:20 <crschmidt> logger: chump M

18:24:20 <crschmidt> M:See [http://www.ilrt.bris.ac.uk/discovery/chatlogs/swig/2005-04-20#T18-24-20|discussion]

18:24:21 <dc_swig> Added comment M13.

18:24:26 <mortenf> thanks chris

18:24:27 <CaptSolo> chris: thanks

18:24:29 <crschmidt> M13: Meeting Closed

18:24:29 <bengee> thx crschmidt

18:24:29 <dc_swig> Replaced comment M13.

18:24:34 <crschmidt> logger: chump M

18:24:34 <crschmidt> M:See [http://www.ilrt.bris.ac.uk/discovery/chatlogs/swig/2005-04-20#T18-24-34|discussion]

18:24:35 <dc_swig> Added comment M14.

18:24:41 <crschmidt> (silly me, failing as a botwrangler)

18:24:49 <jesperll> crschmidt: thx

18:24:50 <crschmidt> No problem! Thanks to all for coming!

18:25:04 <crschmidt> I'll write up the meeting, but I have dayjob requirements to take care of right now :)

18:25:49 <crschmidt> ocelma: the URI you gave is non-public

18:26:07 <crschmidt> http://mtg127/simac/user_profiles <- that one

18:26:44 <libby> cheers chris, all!

18:26:58 * crschmidt also waves to DanC, who snuck in

18:27:12 <libby> libby has changed the topic to: Semantic Web Interest Group hack-n-chat - UTF-8 charset please - Weblog: http://swig.xmlhack.com/ - Logs: http://ilrt.org/discovery/chatlogs/swig/ - please identify with NickServ

18:27:14 * DanC tuned in when libby asked about xtech

18:27:30 <DanC> ^add http://www.w3.org/People/Connolly/travel-sched

18:27:31 <julie> Adding http://www.w3.org/People/Connolly/travel-sched to my database...

18:27:35 * jesperll heads for telly and comfy chair

18:28:03 <julie> Added 1403 statements from http://www.w3.org/People/Connolly/travel-sched. Model size is 2146681.

18:28:06 <DanC> ^q select ?pg where (?who foaf:nick "DanC") (?e caltz:url ?pg) (?e cyc:socialParticipants ?who)

18:28:07 <julie> http://www.xtech-conference.org/, http://www.w3.org/2004/12/allgroupoverview.html, http://www.w3.org/2001/tag/ (x2), http://www.w3.org/2004/12/rules-ws/

18:28:25 <ocelma> hmmm...yep! heres the correct one: http://mtg127.upf.es/simac/user_profiles

18:28:28 * libby waves at DanC

18:28:39 <libby> danc do you use iphoto at all?

18:28:40 <crschmidt> much better :)

18:28:48 <DanC> yes, I use iPhoto for family stuff.

18:29:06 <DanC> e.g. http://dm93.org/200406nh-trip/Page2.html

18:29:36 * crschmidt needs to start annotating his photos *sigh*

18:29:43 <DanC> libby, have you told the web you're attending xtech?

18:30:05 <DanC> I haven't found a cost-effective way to annotate photos.

18:30:12 <libby> no, not yet dan...

18:30:14 <crschmidt> me neither

18:30:21 <DanC> well, I do use iPhoto keywords

18:30:28 <crschmidt> wrote some python for it at one point, but didn't use it

18:30:33 <DanC> and I looked at F-spot the other day. looks promising.

18:30:43 * libby made something danc: sorta works: http://planb.nicecupoftea.org/archives/001292.html

18:30:53 <DanC> my backlog of unpublished photos is overwhelming these days

18:31:04 <libby> I generated some file here: http://libby.asemantics.com/2005/04/pictures/

18:31:10 * mortenf does publish first, ask questions later...

18:31:41 <CaptSolo> DanC, chris: same with me. and the bakclog is huge.

18:31:51 <crschmidt> i have the photos all published

18:31:56 <crschmidt> but no good annotations

18:32:04 <crschmidt> I did do one full set of phtoos with RDF descriptions

18:32:04 <libby> there's also this guy: http://www.holygoat.co.uk/applications/iphoto-rdf/iphoto-rdf

18:32:06 <DanC> yea, saw that, libby... the appliscript breadcrumbs were especially nice. It sure sucks that the album.xml doesn't have all the data.

18:32:07 <crschmidt> but they weren't very complete

18:32:16 <CaptSolo> morten: good practice. and you have a good anotation/query tool for that

18:32:35 <mortenf> not good, but almost sufficient :)

18:32:45 <libby> it's quite fast to use, needs munging as to where you put the photos online though

18:32:58 <libby> morten's stuff is neato

18:33:02 <DanC> I just want a way for the web to automatically sync with iPhoto. If I tag a photo as world-access, I just want it to magically appear in the web. But then I also want to tell stories around sets of photos.

18:33:13 <crschmidt> I'm really not liking gallery2 either, the photo URIs it provides are majorly suck

18:33:24 * DanC was blown away by morten's faceted photo thingy

18:33:31 <CaptSolo> chris: did not go for gallery either

18:33:38 <crschmidt> gallery was okay, because it just saved the photos

18:33:44 <CaptSolo> and grouping of photos and telling stories about them would be nice

18:33:45 <mortenf> thanks :)

18:33:54 <CaptSolo> i wonder can i somehow integrate that into the blog?

18:33:57 <libby> the holygoat guy might be worth chatting to danc...he;'s going an iphoto5 version

18:33:59 <crschmidt> that I can deal with: I can create RDF around http://crschmidt.net/albums/foo/bar.jpg and foo/bar.thumb.jpg

18:34:05 <CaptSolo> cenrtainly the best place to tell stories

18:34:18 <CaptSolo> but not the best place to post photos as individual posts

18:34:24 <crschmidt> but I have no idea if http://crschmidt.net/gallery2/main.php/d/18-2/100_1991.jpg is generated, or stored, or what

18:34:48 <crschmidt> don't know if the URI might change, there's no indication anywhere that i can find

18:34:55 <CaptSolo> look in the source, luke :)

18:35:05 <DanC> iphoto is pretty good for exporting... stuff like http://dm93.org/200406nh-trip/Page2.html comes out of iPhoto and then I added the story manually. But if I want to re-crop a photo, or add one, or fix a timestamp, I lose. iPhoto doesn't remember the connection to the exported photos. It doesn't sync, just export.

18:35:10 <CaptSolo> chris: if you find the function that generates the filenames, you could use that

18:35:37 <libby> right, there's no way to get the path of the exported stuff form iphoto, shame

18:36:55 <CaptSolo> see you all

18:37:01 <libby> cu!

18:37:05 * CaptSolo disappears

18:37:55 * DanC wanders off...

18:38:00 <CaptSolo> DanC: nice photos

18:38:08 <DanC> tx

18:39:16 <CaptSolo> they really are

18:45:00 <crschmidt> DanC: I've done that drive before, maine->canada->chicagoland

18:45:58 <danja> grr, DOAP-a-matic don't work

18:46:18 <libby> yeah

18:46:23 <libby> I tried it earlier

18:46:28 <libby> looks like balbinus is away

18:46:54 <danja> so can I take a screenshot of all those filled in fields...

18:47:13 <libby> copy the one I just did: http://libby.asemantics.com/2005/04/pigsty/pigsty-doap.rdf

18:47:19 <danja> ta

18:47:23 <crschmidt> doap-a-matic isn't all that complex, http://sdlroads.sf.net/doap.rdf should match up to it pretty well

18:47:56 <libby> and link it in the <head> of the homepage, in a foafy way

18:48:03 <danja> ok

18:48:09 <libby> :)

18:48:15 <danja> (and add to triplestore ;-)

18:48:23 * libby has a SWBPWG plan to persuade everyone to craete doap files

18:48:28 <crschmidt> <link rel="meta" type="application/rdf+xml" title="DOAP" href="example" />

18:48:39 * crschmidt needs to make a few more, he supposes

18:48:42 <libby> and classify them as pertaining to the semweb, where appropriate

18:48:51 <crschmidt> i've got four trac projects now for semwebby stuff

18:48:57 <crschmidt> oh! fun stuff, i totally forgot my foafnaut stuff

18:50:48 <danja> I was thinking of nicking FOAF-a-matic as an input source

18:51:02 <libby> think this'll do for category semweb? <category rdf:resource="http://www.w3.org/2001/sw/"/>

18:51:07 <danja> maybe I should hack for DOAPyness

18:51:39 <danja> looks reasonable

18:52:01 <libby> certainly it'd be useful to have a doapa-a-matic working

18:52:32 <danja> I tweaked FOAF-a-matic for Pet-a-matic already, dead straightforward

18:52:59 * libby was thinkng of hacking one for SWBP stuff - it'd be much the same though: http://esw.w3.org/topic/SemanticWebBestPracticesTaskForceOnApplicationsAndDemos

18:53:01 * danja adds to to-do

18:53:06 <libby> :)

18:53:53 <danja> heh, one hell of a WikiWord

18:54:06 <libby> yeees... :)

18:54:25 <libby> if you look lower down, you'll see the plan

18:54:39 <libby> we were catalogiung apps and demos manually, but it was too slow and boring

18:55:11 <libby> see http://esw.w3.org/topic/SemanticWebDOAPBulletinBoard

18:55:12 <danja> nice plan

18:55:15 <libby> only 2 there!

18:55:20 <libby> yeah

18:55:31 <libby> I shoudl ahve maiiled around to get agreement about it but not done it yet :/

18:55:35 <libby> bad /me

18:55:56 <crschmidt> there's a DOAPBullitenBoard on the foaf wiki too

18:56:02 <crschmidt> which isn't semanticweb only

18:56:09 <libby> ah yep

18:56:22 <libby> was that from when I made this one, or before?

18:56:24 <crschmidt> haven't publicized it though, forgot about it

18:56:36 <crschmidt> i did it when you made this one

18:56:41 <crschmidt> cause i was stepping on your toes while editing

18:56:49 <libby> phew, glad I didn;t duplicate

18:56:53 <libby> right

18:56:59 <libby> toes fine!

18:57:11 <libby> just very specific to one particular thing

18:57:27 <crschmidt> ^todoItem popularize DOAP foaf wiki thingy

18:57:29 <julie> crschmidt Make my foafnaut generator work across a model, and generate the whole kit and kaboodle, rather than just being a one-shot script. 2005-04-11T19:42:21Z, crschmidt popularize DOAP foaf wiki thingy 2005-04-20T18:59:01Z, crschmidt Read through http://rdfweb.org/2002/01/photo/, revamp for current state. Include information on http://www.kanzaki.com/docs/sw/img-annotator.html , flickr2rdf, both of which are much more 'state of the art'. Fini

18:57:31 <julie> sh up http://crschmidt.net/semweb/depiction/ and include that as well. 2005-04-06T16:27:14Z, crschmidt document built ins. 2005-04-05T20:17:56Z, crschmidt document builtins 2005-04-06T13:11:20Z, crschmidt Add support in julie for decrypting FOAF files based on a wot:encryptedTo 2005-04-11T11:33:13Z

18:57:39 * libby woudl really just like to see more people using doap

18:57:44 <crschmidt> hm, can delete some of those

18:57:53 * danja blogged

18:58:28 <crschmidt> danny: note that the SemanticWebDoapBullitenBoard is just for semweb apps

18:58:38 <danja> noted

18:58:47 <libby> heh, cheers danny

18:58:59 <crschmidt>http://rdfweb.org/topic/DOAPBulletinBoard

18:59:00 <dc_swig> T: http://rdfweb.org/topic/DOAPBulletinBoard from crschmidt

18:59:06 <danja> thanks for the recent comments btw crschmidt

18:59:10 <crschmidt> T:| DOAPBulletinBoard

18:59:10 <dc_swig> Titled item T.

18:59:15 <danja> sanity-checks ;-)

18:59:16 <crschmidt> T: Link your DOAP files here!

18:59:16 <dc_swig> Added comment T1.

18:59:48 <crschmidt> danja: no problem! i like commenting on your blog so long as it's not RSS related

18:59:55 <danja> hehehe

19:00:26 * danja coontemplating RSS-free zone

19:00:42 <crschmidt> At this point, I just hate RSS and everything RSS related

19:00:48 <crschmidt> although julie can now parse RSS

19:01:32 <crschmidt> hm, should add atom too

19:01:33 <danja> I'm pretty tired of the same-oldness

19:01:42 <crschmidt> yeah, pretty much

19:01:47 <crschmidt> that's why rss 1.1 just died

19:01:52 <danja> right

19:01:55 <crschmidt> too much of the same idiotic arguments

19:02:02 <danja> yup

19:02:14 <danja> better go eat - bye y'all

19:02:22 <crschmidt> seeya

19:02:29 <libby> cu danja

19:02:33 <libby> darn

19:03:13 <libby> T: if it's a semweb application or demo you could also [http://esw.w3.org/topic/SemanticWebDOAPBulletinBoard|add it here on SemanticWebDOAPBulletinBoard]

19:03:14 <dc_swig> Added comment T2.

19:05:18 <crschmidt> hm, need a SVNRepostiroy example

19:05:50 * crschmidt can't spell today

19:05:53 <libby> there are some linked form the doap site

19:06:32 <libby> hoowever not all of them pass through the validator as rdf

19:06:33 <libby> http://svn.usefulinc.com/svn/repos/trunk/doap/examples/

19:08:15 * libby bugs balibinus about doap-a-matic

19:09:29 <crschmidt> ^add http://crschmidt.net/semweb/redland/phpwrapper/doap

19:09:29 <julie> Adding http://crschmidt.net/semweb/redland/phpwrapper/doap to my database...

19:09:30 <julie> Added 21 statements from http://crschmidt.net/semweb/redland/phpwrapper/doap. Model size is 2146708.

19:11:00 <crschmidt> ^q select ?l where (?p doap:shortname "phpredland") (?p doap:license ?l)

19:11:02 <julie> Query returned no results

19:11:20 <crschmidt> ^q select ?l where (?p doap:shortname "redlandphp") (?p doap:license ?l)

19:11:21 <julie> http://usefulinc.com/doap/licenses/gpl, http://usefulinc.com/doap/licenses/asl

19:13:57 * crschmidt adds a couple more files to the DOAP list on the FOAF wiki

19:17:10 * crschmidt decides that 53 tabs is too many, starts closing

19:26:46 <daniel_larsson> The doap-a-matic returns an empty HTML page :/

19:26:58 <crschmidt> daniel_larsson: yeah, we were discussing that

19:27:13 <libby> have asked balbinus to fix

19:27:27 <libby> meantime, copying an existing one is probably the easiest

19:27:36 <daniel_larsson> *nod*

19:27:53 <libby> e.g. http://svn.usefulinc.com/svn/repos/trunk/doap/examples/redland-doap.rdf

19:28:05 <libby> some of those other examples don;t validate as rdf ...

19:28:12 <libby> but that one's ok

19:28:31 <libby> hm, edd?

19:30:59 <libby> T:doap-a-mtaic is temporarily broken: best bet is to copy an example such as [http://svn.usefulinc.com/svn/repos/trunk/doap/examples/redland-doap.rdf|redland doap] or [julie|http://crschmidt.net/julie/doap.rdf] or [http://libby.asemantics.com/2005/04/pigsty/pigsty-doap.rdf|pigsty]

19:30:59 <dc_swig> Added comment T3.

19:31:20 <libby> T3:doap-a-matic is temporarily broken: best bet is to copy an example such as [http://svn.usefulinc.com/svn/repos/trunk/doap/examples/redland-doap.rdf|redland doap] or [julie|http://crschmidt.net/julie/doap.rdf] or [http://libby.asemantics.com/2005/04/pigsty/pigsty-doap.rdf|pigsty]

19:31:21 <dc_swig> Replaced comment T3.

19:34:58 * crschmidt starts working on a replacement, http://crschmidt.net/semweb/doapamatic/

19:35:09 <libby> good plan :)

19:35:23 <crschmidt> doap a matic has broken one too many times for my personal liking

19:35:40 <libby> it's good to have alternatives

19:35:42 <crschmidt> yah

19:35:46 <crschmidt> no offense to balbinus

19:35:51 <crschmidt> stuff is hard to do right and keep up

19:35:54 <libby> borkages happen to the best of us

19:35:56 <crschmidt> i should know, my site was down for days

19:48:29 <jsled> libby: around?

19:48:54 <libby> hello

19:48:57 <libby> sorta

19:49:03 <jsled> hi. :)

19:49:05 <jsled> libby: re http://planb.nicecupoftea.org/archives/001293.html ... what do you mean by "[needing] CC properties " ...

19:49:56 <libby> hm, well I was thinking of cross-searching I guess, implies reusing pictuires for different things

19:50:35 <jsled> are the existing CC properties inappropriate?

19:51:19 <jsled> http://66.102.7.104/search?q=cache:oCUIGVWv0cwJ:web.resource.org/cc/+creative+commons+RDF+schema&hl=en&start=4&client=firefox [too bad the actual URL isn't working...]

19:51:50 <libby> no I think they'll be fine - I just meant I hadn't made my code able to access them

19:51:58 <jsled> got it. mis-read, then.

19:52:10 <libby> no, I qrote it too quickly!

19:52:13 <libby> wrote

19:54:32 <libby> though actually jsled I remember having some issues with the CC schema previously, not used it recently: http://esw.w3.org/topic/W3PhotoVocabs

19:55:02 <libby> the problem I think was that the identifying property for the person was name+date of birth, which seemd invasive and non-unique

19:55:08 <libby> but maybe it's changed now

19:55:30 <libby> before, we added some foaf stuff

19:56:51 <jsled> identifying property as per CC?

19:57:08 <libby> yep

19:57:19 <libby> but as I say, not checked lately

19:57:22 <jsled> boo.

20:02:02 <libby> e.g. http://creativecommons.org/technology/metadata/example1.rdf has a dc:creator as a cc:Agent but the only identifier for them is a dc:title

20:02:35 <crschmidt> I think that's related to the legal stuff behind copyright

20:02:56 <libby> we used cc:Agent but with a foaf:name and foaf:mbox_sha1sum, which seems more useful

20:03:00 <crschmidt> I doubt that there's anything wrong with adding extra identifiers: I added my homepage when I did mine

20:03:03 <crschmidt> yeah

20:03:08 <libby> yeah

20:03:39 <libby> maybe we shoudl try to find out about that. CC will be very important I think

20:03:43 <libby> for thsi sorta stuff

20:03:54 <libby> (not that it isn;t alreday very important :)

20:06:27 <crschmidt> doapamatic no has name, shortname, description, short description, homepage, and maintainer props

20:06:33 <crschmidt> i'd say it's usable now

20:07:30 <sh1mmer> crschmidt it loaded about 2.06m triples in around 40 hours

20:07:51 <libby> license/screenshots useful too...

20:08:00 <crschmidt> yeah, working on it

20:08:12 <libby> but niceone chris. maybe link to the schema and examples so people can add a bit more if they ewant to?

20:08:17 <libby> heheh

20:08:20 <crschmidt> when I get home tonight, yep

20:08:25 <libby> cool

20:08:39 <crschmidt> i'm going to make it more complete, too, this is one of my 15 minute hacks

20:08:43 <crschmidt> rapid development at its finest

20:08:50 <crschmidt> every time you refresh, a new field to fill out! ;)

20:09:11 <libby> woohoo :)

20:10:13 <crschmidt> complete with bugs...

20:10:46 * libby grins at http://www.wasab.dk/morten/2005/04/photos/sparqling-days/2/image-006.html

20:11:00 <libby> wonder what, if anything, I was lookign at?

20:11:50 <crschmidt> are there any programming languages with spaces in the name?

20:13:38 <daniel_larsson> Simula 67 :)

20:14:29 <crschmidt> is there any open source Simula 67 code? :p

20:14:45 * crschmidt will probably do something more roboust for languages later, is going with space seperated at the moment

20:14:46 <daniel_larsson> The object oriented language, which was an improvement to most of its successors...

20:14:51 <benja_> Common LISP

20:15:12 <crschmidt> anything that anyon in here is going to create a DOAP file for?

20:15:34 <jsled> crschmidt: how does it matter?

20:15:59 <crschmidt> jsled: because right now the code is asking for a space seperated list, to save me time and effort in doing something more complex

20:16:03 <daniel_larsson> So... you might need to hack the DOAP file after you generate it, no big deal :)

20:20:51 * crschmidt adds support for multiple licenses

20:36:41 <crschmidt> CVS and SVN repository support

20:39:21 <libby> :)

20:39:31 <libby> you can stop now chris ;)

20:48:10 <crschmidt> boss just asked what is on my shirt

20:48:15 <crschmidt> (wearing SWAD-E)

20:48:44 <libby> what did you say?

21:01:57 <crschmidt> libby: coworker offered up "geekstuff"

21:02:05 <crschmidt> libby: I said "geekstuff, yeah. RDF, specifically"

21:02:29 <crschmidt> he asked "RFID?" I responded "No, much cooler than that." he said "Hm, RFID is a better marketing term."

21:04:00 <libby> heh

21:14:27 <LotR> crschmidt: is comma-seperated that much harder?

21:16:15 <DanC>http://www1.ietf.org/mail-archive/web/ietf-announce/current/msg01133.html

21:16:16 <dc_swig> U: http://www1.ietf.org/mail-archive/web/ietf-announce/current/msg01133.html from DanC

21:16:27 <DanC> U:|Last Call: 'The Atom Syndication Format' to Proposed Standard

21:16:27 <dc_swig> Titled item U.

21:20:33 <crschmidt> LotR: if i did comma seperated, i'd probably have had to strip spaces

21:20:44 <crschmidt> which would have been an extra line of code at the time :)

21:20:50 <crschmidt> (i'll be doing it when I get home tonight)

21:21:50 <LotR> split (/,\s+/, $line) :)

21:22:14 <crschmidt> not in php ;)

21:22:31 <LotR> ah, there's the real problem ;-p

21:22:51 <crschmidt> right, because Perl is so perfect *rolls his eyes*

21:23:16 <crschmidt> I dealt with Perl for quite a long time. for small webpages, I prefer PHP immensely

21:23:20 * KjetilK stares at crschmidt

21:23:31 <KjetilK> :-)

21:23:45 <crschmidt> yeah, yeah. dahut your way right back to irc.perl.org ;)

21:23:51 <KjetilK> :-)

21:24:03 <KjetilK> somebody is missed there, BTW

21:24:36 <KjetilK> but seriously, once you start separating code from markup, I think you'll see the light... :-)

21:24:37 <crschmidt> damn bot

21:25:15 * crschmidt likes having his code and markup tightly integrated

21:25:20 <KjetilK> hmmmm

21:25:29 <crschmidt> so much easier :)

21:25:41 * crschmidt fixes the doapamatic page so it's valid XML

21:26:01 <KjetilK> I found that when creating larger apps, it caused the app to collapse under maintainability problems

21:26:13 <KjetilK> take for example PHPNuke/PostNuke

21:26:16 <crschmidt> well, yeah, but my webpage isn't a large app

21:26:22 <KjetilK> yup

21:26:25 <crschmidt> When I'm doing that kind of stuff, I seperate it quite well

21:26:26 <KjetilK> then PHP can be ok

21:26:42 <KjetilK> allthough XSP is better... :-)

21:26:46 <crschmidt> data retrieval functions get the data, and pass it through a template

21:26:58 <KjetilK> yup

21:27:18 <crschmidt> $replies = get_message_board_replies(); foreach($replies as $v) { print_message_board_reply($v); }, that kind of thing

21:27:35 <KjetilK> yup

21:28:11 <KjetilK> I have two small PHP thingies, but I guess we're sliding OT now

21:30:02 <benja_> if anybody has time to try out an RDF browser, have a look at http://himalia.it.jyu.fi/~benja/2005/04/fenfire-snapshot-2005-04-20.jar

21:30:26 <benja_> (can be run with java -jar)

21:33:00 <crschmidt> how to I add data on a mac?

21:33:56 <benja_> if that was a reply to me :-) you can load a Turtle file by giving its name on the command line, and an xml file by giving '--xml filename' on the command line

21:34:04 <benja_> and you can use Ctrl-G to get data from the web

21:34:18 <crschmidt> hm, ctrl-G says "no gnowsis"

21:34:23 <benja_> (loads the current node's URI and follows seeAlso links)

21:34:25 <mattmcc> KjetilK: Writing large maintainable applications in PHP is perfectly doable. That said, PHP's low learning curve does naturally lead to the creation of large apps by people who don't know how to write maintainable code, regardless of language. :)

21:34:25 <mattmcc> Indeed, most popular PHP apps are poster children for how not to do things.

21:34:27 <benja_> crschmidt: harmless :)

21:34:50 <benja_> I think we looked into it and the message was hard to suppress

21:34:58 <benja_> but it only means you don't get data from that end

21:35:03 <crschmidt> oh, hm

21:35:07 <crschmidt> i was expecting a popup or something

21:35:11 <crschmidt> asking for a uri

21:35:12 <benja_> (Gnowsis is Leo Sauermann's desktop RDF thing)

21:35:22 <benja_> crschmidt, ok, when I have time ;-)

21:35:32 <crschmidt> well, I Just want to browse my own data :)

21:35:41 <benja_> ah, you mean on Ctrl-G

21:36:12 <benja_> there's a hack -- hit Ctrl-J, enter URI on command line and hit enter (goes to the node), hit Ctrl-G

21:36:51 <crschmidt> ctrl-J doesn't seem to do anything for ... ah, i see it

21:37:01 <KjetilK> mattmc: yup, I guess so; I personally really prefer to strictly separate code from markup

21:38:11 <crschmidt> Perl doesn't really force your hand in that way: just look at LiveJournal

21:38:27 <KjetilK> nope, that's correct

21:38:43 <KjetilK> Perl forces nobody to nothing, that's part of the philosophy

21:38:56 <crschmidt> who basically reinvented PHP type stuff with templating, so that they could intersperse code and markup :)

21:39:19 <KjetilK> Actually, I'm not that fond of Perl as a language, but I'm really fond of Perl as a community

21:39:52 <KjetilK> Perl has some weaknesses in OOP methods dispatching, for example, that I feel is rather limiting

21:40:39 <DanC> perl is crack. the perl community is organized crime. ;-)

21:41:35 <benja_> "http://crschmidt.net/wordpress foaf:maker http://crschmidt.net/foaf.rdf#crschmidt"

21:41:41 <benja_> no, other way around

21:41:59 <benja_> "http://crschmidt.net/foaf.rdf#crschmidt foaf:maker http://crschmidt.net/wordpress"

21:42:15 <DanC> the nice thing about perl is that somebody else seems to be willing to write all the perl code that needs to be written. I don't have to write any of it. ;-)

21:42:26 * DanC thinks it's { ?DOC foaf:maker ?WHO }

21:42:45 <KjetilK> hehe

21:42:52 <DanC> ^q select ?C where (foaf:maker rdfs:domain ?C )

21:42:53 <julie> rdfs:Resource

21:42:59 <DanC> ^q select ?C where (foaf:maker rdfs:range ?C )

21:42:59 <julie> foaf:Agent

21:43:14 <benja_> so crschmidt was made by wordpress

21:43:38 <benja_> scary =)

21:44:52 <crschmidt> hm

21:45:27 <crschmidt> oops? :)

21:45:51 * crschmidt killed that triple now

21:45:57 <crschmidt> that URI is long gone anyway

21:46:03 <benja_> =)

21:46:34 <daniel_larsson> You outlived your parent

21:46:38 <crschmidt> heh, heh

21:47:45 <crschmidt> ^q select ?p where (?p foaf:made <http://crschmidt.net/wordpress>)

21:47:45 <julie> Query returned no results

21:47:54 <crschmidt> ^q select ?p where (?p foaf:maker <http://crschmidt.net/wordpress>)

21:47:54 <julie> http://crschmidt.net/foaf.rdf#crschmidt

21:48:57 <crschmidt> ^q select ?p where (?p foaf:maker <http://crschmidt.net/wordpress>)

21:48:58 <julie> Query returned no results

21:49:43 <crschmidt> the maker v. made thing always gets me

21:50:57 <benja_> oh, now I can read weblogs again! :-) (I'd decided some time ago not to use an aggregator until I got my RDF-based one to work, to spur efforts)

21:54:01 <libby> chris, you got a link for doap-a-matic 2, or is that coming later?

21:54:11 <crschmidt>http://crschmidt.net/semweb/doapamatic/

21:54:11 <dc_swig> V: http://crschmidt.net/semweb/doapamatic/ from crschmidt

21:54:22 <crschmidt> V:| DOAP-a-Matic, Mark 2

21:54:23 <dc_swig> Titled item V.

21:54:30 <libby> ta

21:54:36 <crschmidt> V: Since balbinus's is offline, I decided to throw one together

21:54:36 <dc_swig> Added comment V1.

21:54:59 <crschmidt> V: Is less featured, but does put together a decent file to start with

21:54:59 <dc_swig> Added comment V2.

21:55:04 <crschmidt> (did I not link it before? I meant to)

21:55:42 <crschmidt> yeah, i did before too

21:55:53 <libby> T:or you could use [http://crschmidt.net/semweb/doapamatic/|chris' version] or [http://beta.semanticweb.org/|bengee's if you register]

21:55:53 <dc_swig> Added comment T4.

21:56:00 <libby> oh, didn;t see it

21:56:18 <crschmidt> no problem

21:57:05 <sbp> crschmidt: if you don't select a license it uses <license rdf:resource="http://usefulinc.com/doap/licenses/"/>

21:57:12 <sbp> seems weird

21:57:18 <crschmidt> hm

21:58:52 <sbp> ah, it's only if you select "select one or more"

21:58:57 <crschmidt> fixed it now

21:59:18 <sbp> I don't see that

21:59:28 <benja_> ^add http://fenfire.org/doap.turtle

21:59:28 <julie> Adding http://fenfire.org/doap.turtle to my database...

21:59:29 <julie> Adding that URL failed (XML parser error - Document is empty).

21:59:33 <sbp> ah yes

21:59:39 <benja_> *shrugs*

21:59:47 <crschmidt> benja_: she doesn't do turtle like that

21:59:59 <dajobe-lap> where did that suffix come from :) .ttl is the approved one ;)

22:00:33 <crschmidt> if you give it .ttl, I can parse it

22:01:28 <crschmidt> julie's got a parse_anything, which attempts to do simple heuristics, mostly based on extensions

22:01:33 <benja_> dajobe-lap, civil disobedience ;-)

22:01:51 <benja_> crschmidt, I'm giving the approved MIME type?

22:02:03 <crschmidt> benja_: yeah, yeah. I don't do mime types yet

22:02:15 <crschmidt> it also doesn't use the ^add method, because I'm still experimenting with it

22:02:16 <benja_> aha. hmm =)

22:02:26 <dajobe-lap> not actually approved. mime type. turtle itself has no real stamp of officialdom from anyone

22:02:48 <benja_> dajobe-lap: well, in the same vein as .ttl is approved

22:05:26 <benja_> dajobe-lap, btw, I still don't get how "character* with excapes as defined in N-Triples" can mean that characters in the range 128+ do not have to be escaped (although they have to be in N-Triples) but quotes " do have to be (because they have to be in N-Triples)

22:07:05 <dajobe-lap> nobody else had that problem

22:07:13 <dajobe-lap> n-triples is 7 bit ascii - 128-255 don't exist

22:07:21 <dajobe-lap> turtle is utf8 - so you don't escape 128-255

22:08:06 <crschmidt> ^parse http://fenfire.org/doap.turtle

22:08:08 <julie> Model size increased by 30 to 2146737 by adding turtle data.

22:08:14 <benja_> =)

22:08:19 <crschmidt> I s till think you should change the extension

22:09:06 <DanC> extensions are the server-owner's business. you should be paying attention to the mime type, crschmidt

22:09:43 <sbp> http://fenfire.org/doap would be better from a URI design point of view, too

22:09:50 <crschmidt> I'm passing the url off onto Redland. Someone other than me should be paying attention to the mime type :)

22:09:59 <benja_> sbp, true

22:10:08 <crschmidt> ^parse http://crschmidt.net/blog/feed/atom

22:10:09 <julie> Error: Using property attribute 'version' without a namespace is forbidden.

22:10:15 <crschmidt> ^reload modules

22:10:15 <julie> Reloaded extras.

22:10:16 <crschmidt> ^parse http://crschmidt.net/blog/feed/atom

22:10:21 <julie> Model size increased by 38 to 2146776 by adding RSS data.

22:10:26 <crschmidt> have i ever mentioned that I love python?

22:10:35 <dajobe-lap> twice today

22:10:45 <crschmidt> make that three, please

22:11:07 * crschmidt heads home

22:12:01 <benja_> dajobe-lap, I think nobody has had a problem because the non-grammar part of the spec makes the intent clear =-)

22:12:33 <benja_> anyway...

22:20:43 <darobin__> darobin__ is now known as darobin

22:45:05 <libby> heh heh: /me discovers how to set my ichat message to what I'm watching on tivo

22:46:51 <JibberJim> hmm, so what happens if your tivo is recording Debbie Does Dallas for its own pleasures whilst you're watching a futurama DVD, people will get entirely the wrong idea.

22:47:31 <libby> hm, good point

22:47:41 <libby> our tivo is a little eccentric

22:48:24 <evamen> Libby, Jim!!! :-)

22:48:35 <libby> evamen!!

22:48:51 * evamen wondering if tivo is TV (it isn't at my dictionary)

22:49:08 <libby> it's a harddisk tv recorder

22:49:12 <JibberJim> evamen!

22:51:28 <evamen> Mmm... interesting... HD tv recorder!!!

22:51:41 <libby> it's danbri's really :)

22:52:22 <libby> it's got a web interface, so I hacked a script to grab the 'now watching' page every 10 mins and put it in ichat

22:52:41 <evamen> :-o How much technology in that house!!!

22:53:05 <libby> :)

22:53:16 <libby> can't live without it now!

22:55:41 <jsled> tivo? No kidding. Easily one of the best consumer electronics devices ever.

22:55:52 <libby> yes indeed

22:59:42 <deltab> JibberJim: I imagine it'd be more interested in Debian Does Dells