Experimental IRC log swig-2008-03-28

Available formats: content-negotiated html turtle (see SIOC for the vocabulary)

Back to channel and daily index: content-negotiated html turtle

These logs are provided as an experiment in indexing discussions using IRCHub.py, Irc2RDF.hs, and SIOC.

00:00:00<timbl>I'm not convinced that it is useful to represent the relationship between an HTML page and the embedded images as container membership.
00:00:19<dorian>hmm
00:00:44<dorian>good point
00:01:13<timbl>You can have a separate container for, for example, all the resorces necessary to fully present a given main resource.
00:01:30<dorian>i mean, considering the semantics of any "embedding" can be considered a "suggestion" on the part of the markup author
00:01:55<dorian>hmm
00:03:16<danbri>timbl, re your blog post ... I had a chat with Brad Fitzpatrick earlier, re Google SG API
00:03:23<danbri>... its using Raptor for FOAF parsing now :)
00:03:44<timbl> oh really, excellent.
00:04:06<timbl>so i coul dpoint it at my foaf file
00:04:25<timbl>and then it puts it in a triple store?
00:04:32<timbl>Will it absorb arbitrary RDF?
00:05:04<dorian>so should containers be considered disjoint from other resources then? like a directory to a file, or a DAV container to atomic DAV resources, or a sioc:Container to a sioc:Item
00:06:12<timbl>There are some cases when I am happy to say something *is* an RDF collection or a class, but rare
00:06:22<timbl>Prefer to link them.
00:07:14<timbl>(danbri, which blog post?)
00:07:35<danbri>the one on planetrdf.com just now
00:07:48<timbl>about maching up linked data
00:07:53<timbl>s/ch/sh
00:08:00<dorian>so the notion of "containment" is not really a big deal then?
00:08:01<danbri>http://code.google.com/apis/socialgraph/docs/testparse.html
00:08:14<danbri>A:|New testparse service for Google SG API
00:08:26<danbri>A:The FOAF stuff also now updated to use Raptor
00:08:45<danbri>i thought you might wanna know, re the 'google could be superceded' misquote
00:09:02<timbl>Ah that ..
00:09:06<timbl>that was a pain.
00:09:55<timbl>Life, I suppose.
00:10:24<timbl>But anyway, back to the positive. If there a web form for testing the google parser?
00:10:27<danbri>yeah
00:10:37<danbri>web form, no, though i gues there could be easily
00:10:46<danbri>grabbing the perl script worked for me in macosx
00:10:53<danbri>just one arg, url of my foaf, worked
00:10:54<timbl>ok
00:11:06<danbri>it shows the kinda flattened 'triples' that sg thinks in natively
00:11:25<danbri>tends to be things like http://danbri.org/ contact http://tantek.com/
00:11:37<danbri>or urls of foaf files, where the person is implicitly the primary topic of the file
00:11:52<danbri>i am thinking to bring foaf closer to that model by saying that a homepage is an OnlineAccount
00:12:17<danbri>realising that SIOC already effectively does this, by having http://forums.example.com/userid be a User, where User is subclaass of OnlineAccount
00:12:36<dorian>ACTION trots off to ponder, thanks tbl
00:13:17<danbri>http://lists.foaf-project.org/pipermail/foaf-dev/2008-March/009106.html
00:13:32<danbri>B:|mapping of foaf:knows to xfn contact vs acquaintance
00:13:44<danbri>B:Discussion w.r.t. best map for Google's SG API
00:16:27<timbl>umm in these tripleque things , what is the middle thing?
00:16:31<timbl> [
00:16:31<timbl> "http://www.w3.org/People/Berners-Lee/card.rdf",
00:16:31<timbl> "me",
00:16:31<timbl> "http://www.w3.org/People/Berners-Lee/card.rdf#edd"
00:16:31<timbl> ],
00:16:45<timbl>in 'edges'
00:16:48<timbl>for example
00:22:09<danbri>middle thing is values from xn
00:22:11<danbri>er xfn
00:22:17<danbri>hence mapping discussion above
00:36:25<karlUshi>ACTION is reading the new mails of henry about RDFauth
00:38:19<karlUshi>I wonder what are the implications of showing only partial parts of the graph. Will it break RDF?
00:38:41<ericP>it should most emphatically not break RDF
00:38:49<karlUshi>ericP: yes agreed
00:38:55<ericP>that's the point of a monotonic language
00:38:59<karlUshi>but that's challenging, not easy
00:39:40<karlUshi>which part of the file you don't show or not. let's say RDF/XML representation
00:39:55<karlUshi>which attributes can you remove, which elements, etc.
00:40:34<karlUshi>Maybe it will have strong implications on the way you design the graph at the start. A bit like unit testing does for code.
00:40:59<ericP>ahh, if you want to remove parts of the XML tree that represents some RDF graph, you are probably eliminiating some triples and *changing* others
00:41:22<karlUshi>:) yep
00:41:30<ericP>one of my favorite non-mon examples is store ours
00:42:10<ericP>i write { :thisStore :openFrom "0800"; :openTill "1630" . }
00:42:23<ericP>then i discover that the sunday hours are different
00:43:51<ericP>i *want* to re-use the old data and say { :thisStore :openFrom "0800"; :openTill "1630" ; :openException [ :date "Sun" ; :openFrom "0900" ; :openTill "1330" ] . }
00:44:10<ericP>but that breaks the previous understanding of :openFrom and :openTill
00:44:27<TedThibodeauJr>and what do you do with the siesta nations? (closed for lunch/nap in the middle)
00:45:07<ericP>siesta nations are inherently non-mon. logitions have agreed on that for centuries
00:45:14<TedThibodeauJr>*laughs*
00:45:30<dorian>i guess you'd have to have a separate statement citing the range of open-ness
00:45:31<karlUshi>:))
00:45:32<TedThibodeauJr>so no RDF for them!
00:45:36<dorian>and then references to it?
00:46:48<timbl>karlUshi, "Maybe it will have strong implications on the way you design the graph at the start. A bit like unit testing does for code." -- no.
00:47:00<timbl>No implications on how you design a graph
00:47:24<karlUshi>if you cut the graph in the middle because there is an information you don't want to show.
00:47:25<timbl>Any graph can be subsetted in any way, and the subset is always implied by the original
00:47:47<timbl>if you cut the graph in hte middle you get 2 disconnected subgraphs
00:48:08<timbl>Which is fine and normal
00:48:13<karlUshi>ah ok :)
00:49:27<karlUshi>ACTION is trying to think about Venice and masks.
00:49:44<karlUshi>same body, hidden identity
00:49:53<timbl>pastebin with the resut of the google api parse: "Sorry, your post tripped our spam filter - let us know if you think this could be improved"
00:50:14<dorian>"how about by not tripping the spam filter?"
00:50:58<dorian>maybe the pastebin folks haven't read about face
00:54:31<timbl>The spam filter seems to be tripped my repettion of the same URI, so a munged version of the snippet is now at http://pastebin.com/d3af44637
00:55:19<timbl>opr rather http://pastebin.com/f8e24e59
00:55:55<timbl>and that would sugest to me that #cm and #edd are both related to the card.rdf file by a 'me' link
00:56:03<timbl>which suggests they are the same person
01:00:55<danbri>i asked if he could opensource the mapping code ,.. unlikely
01:01:01<danbri>but we could get testcases it seems
01:01:39<danbri>logger, pointer?
01:01:39<logger>See http://chatlogs.planetrdf.com/swig/2008-03-28#T01-01-39
01:03:19<karlUshi>http://en.wikipedia.org/wiki/List_of_pastebins
01:03:30<karlUshi>C:|List of pastebin
01:04:49<karlUshi>ACTION is usually using http://paste.ubuntu-nl.org/
01:48:45<danbri>http://www.wikinomics.com/blog/index.php/2008/03/26/wiki-collaboration-leads-to-happiness/
01:49:00<danbri>D:|Wiki vs Email collaboration
01:49:04<danbri>D:Re doc authoring/editing
07:07:17<dajobe>http://triplr.org/
07:07:30<dajobe>E:|Triplr 1 year old and JSON serializers callback parameter restored
07:07:59<dajobe>E:so RDF/JSON output can be more easily embedded and called from inside javascript
07:08:21<dajobe>E:more on the [http://triplr.org/when|triplr when] page
07:09:16<dajobe>E:->swig,rdf,json,triplr,raptor
10:00:45<chaals>dahuut
10:03:11<darobin>dahut!
10:04:09<chaals>da-robin-hut!!
10:04:49<darobin>chaals!
10:05:07<chaals>no, just someone who looks a bit like him.
10:05:27<darobin>right
10:05:27<chaals>oh, alright. yep, me :)
10:05:34<darobin>and happens to greet with "dahut"
10:14:04<kjetilkWork>OMG, both chaals and darobin is here
10:15:32<darobin>hey kjetilk!
10:16:32<kjetilkWork>:-)
10:21:42<danja>DanC, I just wanted to use an HTML doc as a namespace doc, with GRDDL-embedded transformation
10:22:32<danja>but it seems glean-profile doesn't recognise namespaceTransformation - http://www.w3.org/2003/g/glean-profile
10:22:46<danja>I can't think of any reason not to add this -
10:23:10<danja>tweaked version at http://hyperdata.org/xmlns/xrds/glean-profile.xsl
10:24:53<danja>there's an example which uses it at : http://hyperdata.org/xmlns/xrds/example-tweaked.xml
10:25:17<danja>namespace doc is at http://purl.org/stuff/xrdsx/
10:26:22<danja>(ignore the specific content there - it's only to demonstrate that grddl is available - the vocab isn't very webby)
10:26:55<danja>phenny, tell DanC see http://chatlogs.planetrdf.com/swig/2008-03-28.html#T10-21-42
10:26:56<phenny>danja: I'll pass that on when DanC is around.
10:27:12<danja>chaals!
10:30:23<chaals>danja!
10:50:05<danbri>$nick!
11:28:40<lordi>is anyone aware of a shop that has a RDF or GRDDL support? (for getting product or product type hierarchy)
11:29:18<lordi>product information*
12:17:44<kidehen>http://www.sitepoint.com/blogs/2008/03/26/microsoft-set-to-launch-semantic-web-light/
12:18:12<kidehen>F: They call it "Semantic Web Light", beats me, but they are coming around, albeit gradually :-)
12:27:11<mahound>:/
12:27:28<mahound>"semantic web light"?
12:30:02<Phae>"light" is always a concerning word, imho
12:30:12<mahound>ACTION agrees
12:30:34<Phae>i find it interchangable with "half-arsed"
12:30:42<mahound>:)
12:31:53<mahound>i think the second comment on the original article says it all
12:34:26<mahound>F: [here|http://savas.parastatidis.name/2008/03/24/19a7fa4a-560d-47b4-bbea-9a22e9824bb8.aspx] is the original blog post
13:08:48<danbri>http://savas.parastatidis.name/2008/03/26/21401781-24ec-4b36-8dc7-b9fca72c2e3d.aspx
13:09:01<danbri>G:|Relationships can have properties as well
13:09:40<danbri>G:Microsoft researcher on an age-old semweb issue; qualifying statements. I was talking with Damian about this just yesterday too.
13:10:34<danbri>G:I wonder whether named graphs address enough of these 'qualify the triples' use cases
13:10:57<danbri>G:eg. write assertions about the rdf/xml or data source, so they can be matched in SPARQL GRAPH queries
13:21:10<timbl>F:|Microsoft and "Research-Output" Repositories
13:40:33<timbl>G: I added a "just use N3" comment : Just use Notation3. You don't need to add lots more columns to your triple, once subgraphs become first class objects. Then they can be the subject (and/or object) or arbitrary statements. <21401781-24ec-4b36-8dc7-b9fca72c2e3d.aspx> dc:creator <#Savas>; dc:date "2008-03-26"; ex:whenPublished { <#Savas> foaf:organization <#MSFT>; geo:basedNear dbpedia:Seattle}.
15:26:54<HarryH>ACTION is working on bureaucracy and slides re Social Networking
15:27:15<Iskr>bureaucracy?
15:27:55<HarryH>You know, e-mailing people rather than hacking :(
15:28:08<Iskr>i see
15:28:47<HarryH>However, as soon as e-mail is done will try to merge myspace and FOAF file.
15:29:26<HarryH>Are there any good algorithms for identifying whether or not two FOAF files *likely* identify the same person other than just relying on inverseFunctionalProperties?
15:29:33<HarryH>You know, 80/20 sort of stuff.
15:29:54<Iskr>i dunno
15:30:03<HarryH>I've read lots about this problem *in general* but never anything about FOAF per se.
15:30:25<HarryH>It seems like you could say "Well, if the string names match and so does the birthday, then it's likely the same person"
15:36:17<Freso>HarryH: You should probably also be aware that "H. H." might just be same as "Harry H." might just be the same as "H. Haplin" might just ...
15:36:57<Freso>HarryH: (Or, in other words, two strings might mean the same, even if they're not identical.)
15:37:50<tommorris_>I'm working on doing name comparisons. Well, sorta working. More trying to avoid YouTube and my inbox and occasionally scrapping together bits of code.
15:42:22<Freso>tommorris_: Ouch. Good luck on that.
15:42:42<tommorris_>Tip: US Census Data.
15:55:18<mahound>tommorris_: is it about the anti-islamic video publiched by some sutch right wing guy?
15:55:30<tommorris_>no
15:55:40<tommorris_>it's just about weed and legalisation etc
15:55:52<mahound>oh, ok
15:56:01<mahound>s/sutch/dutch
15:56:28<mahound>s/publiched/published/
16:04:40<DanC>ACTION waves to HarryH , not really in office mode today
16:04:40<phenny>DanC: 10:26Z <danja> tell DanC see http://chatlogs.planetrdf.com/swig/2008-03-28.html#T10-21-42
16:05:56<DanC>phenny, mail public-grddl-comments about glean-profile and namespaceTransformation
16:06:42<DanC>phenny, tell danja that glean-profile and namespaceTransformation seems like a reasonable idea but doesn't fit on my desk today, so please queue it in public-grddl-comments@w3.org
16:06:42<phenny>DanC: I'll pass that on when danja is around.
16:07:54<HarryH>Freso, yep that's a good point - thinking name identity is tricky, but possibly helped by just lots of regexes.
16:08:22<HarryH>ACTION waves at DanC, don't worry - will e-mail you some draft slides next week.
16:08:29<DanC>ACTION used to write cobol programs to process tapes from name matching services
16:08:30<HarryH>ACTION is making progress, slowly but surely.
16:08:49<HarryH>In fact, if anyone *has* the mother-of-all name-matching regexes around, I'd appreciate it :)
16:19:47<kwijibo>anyone know of a vocabulary for describing a file system?
16:36:22<timbl>I feel that the identity match algorithms are a black art, very much a case of diminishing returns.
16:36:54<timbl>I like the southampton 'disambiguator' which allow you to claim to be certain instances and disclaim others.
16:37:00<timbl>For example in publications
16:46:02<bblfish>http://blogs.sun.com/bblfish/entry/rdfauth_sketch_of_a_buzzword
16:46:21<bblfish>H:| RDFAuth: sketch of a buzzword compliant authentication protocol
16:46:54<bblfish>H: RESTful, Resource Oriented, Hyperdata powered authentication protocol that is simpler than OpenId
16:47:18<bblfish>ACTION ligning up the buzzords :-)
16:48:54<bengee>you shouldn't call it RDfAuth if it's not following RDFAuth's approach ;)
16:50:19<bblfish>heh. well I explain at the end of the post that I still think I can convince you
16:51:16<bengee>'k
16:53:10<HarryH>Agree that identity matching algorithms are a black art, but want a simple example to show someone that turning their OpenSocial XML into FOAF helps them merge their friend lists rather than forcing them to do more work by matching their friends's URIs by hand :)
17:09:28<danbri>same name and same birthdate help
17:09:59<danbri>although my greatgrandma and her twin went by same name (they swapped) at different times of their lives
17:10:33<danbri>('marie madeleine' and vice-versa, fwiw)
17:25:18<HarryH>Sigh.
17:25:22<HarryH>There is no hope for the universe then :)
17:25:59<HarryH>I can imagine a possible world, a parallel Earth, where there is a 'zombie-harry' with all the same IFPs, yet a different identity.
17:26:02<HarryH>Would it be me?
17:26:12<HarryH>More precisely, should it get the same URI?
17:26:48<HarryH>Perhaps the chemical make-up of water in its body is not H20, but XYZ. But we don't talk about that "yet" in the SemWeb.
17:27:21<HarryH>I am referencing a quite well-known argument here, check the full paper, the "Meaning of 'Meaning'" by Hilary Putnam.
17:29:09<bblfish>yes, if you are a kripkean, then only that only causal transformations of the intial egg you came from are counterparts of you
17:29:56<bblfish>hopefully we don't need counterfactual logic soon for the semantic web
17:30:05<bblfish>I think we can get a lot done without that
17:32:52<HarryH>I'm more of a "If it looks like a duck, and talks like a duck, it's a duck" kinda neo-Fregean.
17:33:01<HarryH>I think that's the best we can hope for the SemWeb.
17:33:06<HarryH>And it's also philosophically correct.
17:33:55<bblfish>Ah, I'll just finish a few things I need to do, then read up on the literature, and get back to you ...
17:34:03<bblfish>;-)
17:36:46<HarryH>http://www.ibiblio.org/hhalpin/homepage/publications/indefenseofambiguity.html
17:37:16<HarryH>I: | In Defense of Ambiguity: a philosophical take on the identity issue on the SemWeb
17:38:06<bblfish>thanks
17:56:14<EtnaRosso>bye have a nice week end
18:10:08<danja>phenny, tell dajobe ack!
18:10:08<phenny>danja: I'll pass that on when dajobe is around.
18:10:11<phenny>danja: 16:06Z <DanC> tell danja that glean-profile and namespaceTransformation seems like a reasonable idea but doesn't fit on my desk today, so please queue it in public-grddl-comments@w3.org
18:15:50<dchud>edsu++
18:16:18<dchud>(sorry wrong channel, but the sentiment holds)
19:39:50<mhausenblas>yves around?
19:40:11<yvesr>mhausenblas: yes
19:41:29<mhausenblas>wow. just checked riese and found out it is broken
19:41:35<yvesr>hmm?
19:41:44<mhausenblas>well, it is fixed again
19:41:53<mhausenblas>however, I thought a bug, at first
19:42:25<mhausenblas>BUT; no: flacky data
19:42:47<mhausenblas>the UCI store of economy was broken
19:42:52<yvesr>where? did you try with the new shiny end-point to check if the bug hasn't been resolved yet?
19:43:28<mhausenblas>no. the 'bug was the folllowing line in the uci-store.rdf in economy:
19:43:30<mhausenblas>< rdf:resource="http://google.com/"/>
19:43:41<yvesr>the linked data on riese (relying on the small content.rdf static files) is seriously outdated anyway
19:44:01<yvesr>oh?
19:44:03<mhausenblas>sure.
19:44:15<mhausenblas>but that was a totally different problem ;)
19:44:17<yvesr>i guess someone must have explicitely stated that
19:44:23<yvesr>which is a problem with uci :)
19:44:24<mhausenblas>mhm
19:44:35<mhausenblas>the first UCI hack! ;)
19:44:40<yvesr>it is quite easy to state anything
19:44:53<mhausenblas>might be someone playing around with the REST API
19:45:04<yvesr>yes, probably
19:45:12<mhausenblas>fun, ain'T it
19:45:14<yvesr>:)
19:45:51<mhausenblas>IMHO there is at least one important lesson learned, here ;)
19:47:30<mhausenblas>so, now, after that bit of amusement back to - well you know what ... :)
19:47:40<danja>http://www.hueniverse.com/hueniverse/2008/03/putting-xrds-si.html
19:48:00<danja>J:| Putting XRDS-Simple in Context
19:49:19<danja>J: stuff for OpenID/OAuth
19:49:28<danja>J: "...picking RDF as the basis for web-wide service discovery is a hard sell. So is SOAP or other existing descriptor formats. XRDS has the advantage of already breaking through the XRI/OASIS world and being adopted by the OpenID community. It has momentum."
19:49:56<danja>J: "...yes, we can potentially drop XRDS-Simple and pick a whole other solution"
19:51:01<mhausenblas>danja, are you an OpenID wiz? :)
19:51:37<danja>J: (I just got into some fairly unproductive arguments over there - managed to give the impression I was pressing for RDF/XML when I was really only aiming for GRDDL + some tweaks to make the stuff more WebArch-friendly)
19:52:01<danja>mhausenblas, not remotely :-)
19:52:18<mhausenblas>I ask because I was about to add OpenID to http://143.224.254.32/irs/ (for the 'says' bit)
19:52:20<mhausenblas>ah
19:52:24<danja>trying to keep an eye on it though
19:52:35<danja>Sam Ruby has an excellent how-to
19:52:41<danja>.g sam ruby openid
19:52:42<phenny>danja: http://intertwingly.net/blog/2007/01/03/OpenID-for-non-SuperUsers
19:52:48<mhausenblas>yeah, saw it
19:52:56<mhausenblas>still, I think too heavy for me
19:53:18<mhausenblas>all I want is (I guess): Check if this person is the person it claims to be
19:53:26<mhausenblas>that is: no user managment, etc.
19:53:31<danja>dunno - seemed I only had to add a couple of header links and It Just Worked
19:53:48<mhausenblas>really? maybe I'm too dumb
19:54:05<danja>- ah, that was for logging in to other peoples sites
19:54:18<mhausenblas>yep
19:54:49<danja>there must be an easy doc around somewhere - I do need to read up more on this (and OAuth, catch up with bblfish)
19:55:02<bblfish>hi
19:55:04<mhausenblas>mhm. if you find it, please let me know ;)
19:55:08<mhausenblas>hi
19:55:13<danja>will do :-)
19:55:24<mhausenblas>thanx, danja :)
19:56:28<danja>bblfish - quick question first re. OAuth
19:56:31<danja>"My use case is OAuth. I need a format to document all the needed configuration and endpoints needed to successfully talk to an OAuth-protected resource."
19:56:43<bblfish>yes
19:56:57<danja>what *are* the needed config & endpoints - isn't the target URI enough?
19:57:03<danja>...to get started...
19:58:15<danja>sorry, biab - got to sort barky dogs
19:58:36<danja>how's your OpenId bblfish - mhausenblas is after using it
19:59:00<mhausenblas>hey, that was my Q ;)
19:59:05<bblfish>I started looking at OAuth, but as I understand it is more than what I am requiring now, and not machine friendly enough
19:59:18<mhausenblas>bblfish, can you peradventure gimme a hint?
19:59:19<bblfish>you mean RDFAuth?
19:59:52<bblfish>I need to think more about OAuth and read more about it.
20:00:35<mhausenblas>bblfish, I was asking about how to use OpenID for checking identity of users making statements (see http://chatlogs.planetrdf.com/swig/2008-03-28.html#T19-52-18)
20:00:56<bblfish>sorry just came back from pub, two wiskys and a beer later..
20:01:46<mhausenblas>hey, great ... I need a whisky too :)
20:01:55<bblfish>so I am not quite in a state to think too deeply
20:01:57<bblfish>:-)
20:02:19<mhausenblas>ACTION wondering if bblfish will be around TRIPLE-I this year again ...
20:02:28<bblfish>yes, should be
20:02:35<bblfish>I need to send in a presentation
20:02:37<bblfish>...
20:02:42<bblfish>thanks for reminding me
20:02:50<mhausenblas>welcome ;)
20:03:02<mhausenblas>so, we can continue our chat from last year ;)
20:03:12<mhausenblas>at least with RDFa we had some progress ;)
20:05:28<Shepard>bblfish: there's too much noise on foaf-dev at the moment so I haven't read the thread, just a short idea looking at your sequence diagram ...
20:05:46<bblfish>try looking at this: http://blogs.sun.com/bblfish/entry/rdfauth_sketch_of_a_buzzword
20:06:01<bblfish>it summarises the discussion
20:06:22<Shepard>did you consider having just one foaf file which will return a minimum set of data when requested without authentication but will return more data when there's an authorised user requesting it?
20:06:32<bblfish>yes
20:06:36<bblfish>that's what this is about
20:06:41<bblfish>my new diagram shows this
20:06:42<Shepard>as opposed to having a public and a private one
20:06:45<Shepard>ok, cool :)
20:06:48<bblfish>ah, yes
20:06:53<bblfish>that was the original diagram
20:06:59<bblfish>but it got people confused
20:07:12<bblfish>so in that post I suggest that step 1 and 2 could be fused
20:07:37<bblfish>How to do this in HTTP I am not sure. So I am looking for feedback from HTTP experts to tell me
20:07:40<Shepard>well the idea is to have just one resource, one uri
20:07:50<bblfish>yes I agree, that was my initial idea
20:07:54<Shepard>ok
20:08:11<bblfish>just want to hear back from HTTP and REST experts to know if that is ok
20:08:38<Shepard>I thought about splitting my foaf into different resources (contact data, relationships, ...) but got into hassle where to draw the line...
20:08:52<bblfish>yes. That is a semantic web question
20:09:03<bblfish>I'd be interested to hear people's point of view on that
20:13:24<bblfish>mhausenblas: ah yes I remember the conversation
20:13:38<bblfish>let me add iSemtnatics to my todo list
20:24:04<bblfish> ok. probably have to go. Sorry not to be able to answer the questions right now...
20:24:22<bblfish>(the more technical ones I mean)
20:25:33<Shepard>phenny, tell leobard: do you have something like this for nepomuk? http://groups.google.com/group/rdf-schema-dev/msg/43667b206274b6ba
20:25:34<phenny>Shepard: I'll pass that on when leobard is around.
20:28:08<mhausenblas>sure, thanks bblfish - we'll see us then ;)
22:30:42<KjetilK>hmmm, no bblfish
22:31:15<KjetilK>phenny: ask bblfish where his PGP key is, http://bblfish.net/people/henry/henry.gpg.asc is 404
22:31:18<phenny>KjetilK: I'll pass that on when bblfish is around.

Back to channel and daily index: content-negotiated html turtle