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.
| 21:44:02 | <ChNepote> | Hi. Currently working on a SMOB publisher on WikiNi (french wiki engine). |
| 21:44:34 | <ChNepote> | Trying to understand the exemple of Alexandre Passant : http://apassant.net/smob/data/2008-05-22T13:41:43-07:00.rdf |
| 21:45:27 | <ChNepote> | And don't understand what does mean "<rdfs:seeAlso rdf:resource="type=user&user_id=alex"/>" in his file ??? |
| 22:10:06 | <tuukkah> | hello ChNepote! |
| 22:11:00 | <tuukkah> | ChNepote, what's your background regarding semantic web, rdf, sioc? |
| 22:11:48 | <tuukkah> | ChNepote, and what do you want to achieve? an extension to the wiki so that it can be used to publish smob data? |
| 22:21:44 | <ChNepote> | I am not a very beginner in semantic (I created websemantique.org with Karl Dubost and Eric van der Vlist a few years ago). |
| 22:22:20 | <ChNepote> | Yes, an extension to the wiki so that it can be used to publish smob data. |
| 22:27:35 | <tuukkah> | now, in theory you just need to produce the relevant foaf and sioc data. in practise, our current implementation has a lot of details you'll need to follow |
| 22:28:53 | <tuukkah> | if you know sparql, you can see the query we use here: http://smob.googlecode.com/svn/trunk/server/data.php |
| 22:29:09 | <tuukkah> | ($q = ...) |
| 22:29:45 | <ChNepote> | The FOAF exporter is allready written ; see http://websemantique.org/CharlesNepote/userdata for example |
| 22:30:00 | <tuukkah> | and the way we get the data is simply http://smob.googlecode.com/svn/trunk/server/load/index.php |
| 22:30:37 | <tuukkah> | for example, we don't do anything about rdfs:seeAlso |
| 22:31:56 | <tuukkah> | there is a set of semantic web publishing practises called "linked data", and we try to follow them - hopefully more in the future |
| 22:32:08 | <ChNepote> | OK. Thank you for to have shown me the sparql request. |
| 22:33:14 | <tuukkah> | the seeAlso you ask about is one practise for linking a resource to the rdf document containing information about it |
| 22:33:38 | <tuukkah> | it's clearer in the foaf:Person case, right? |
| 22:34:43 | <ChNepote> | I know those practises but I am still a bit confused with the use cases of seeAlso (in particular with sioc:User and foaf:Person). |
| 22:35:13 | <tuukkah> | i see |
| 22:35:46 | <tuukkah> | we can see alex uses this as his foaf:Person URI: http://apassant.net/alex |
| 22:36:25 | <tuukkah> | but to get the rdf data about http://apassant.net/alex, you need to follow the rdfs:seeAlso to http://apassant.net/foaf.rdf |
| 22:36:55 | <ChNepote> | yes, and http://apassant.net/alex can't be deferenced at that time... |
| 22:39:47 | <tuukkah> | apassant.net doesn't seem to work currently :-( |
| 22:40:24 | <tuukkah> | i think the sioc:User uri's are currently completely irrelevant and bogus |
| 22:41:03 | <tuukkah> | (in our implementation of smob) |
| 22:41:53 | <ChNepote> | Yes :( If I understand, I will use on the one hand : sioc:User http://websemantique.org/CharlesNepote and rdfs:seeAlso http://websemantique.org/CharlesNepote/userdata |
| 22:42:41 | <tuukkah> | this is just one way you can implement the linked data guidelines |
| 22:42:49 | <ChNepote> | And on the other hand : foaf:Person http://websemantique.org/CharlesNepote/userdata#me and rdfs:seeAlso http://websemantique.org/CharlesNepote/userdata |
| 22:43:20 | <tuukkah> | and for now, you could leave the sioc:User out as well as the seeAlso on the foaf:Person |
| 22:43:22 | <ChNepote> | (you can follow the links to see the result) |
| 22:44:38 | <ChNepote> | Yes, as you don't use it in the SPARQL query... |
| 22:45:06 | <tuukkah> | and loading the data we do by the url you provide to load.php... |
| 22:46:02 | <ChNepote> | ok |
| 22:46:12 | <tuukkah> | are you missing foaf:name? |
| 22:47:27 | <tuukkah> | (foaf:thumbnail you seem to have) |
| 22:49:06 | <ChNepote> | foaf:name can be precised by the user but it is not generated by the wiki as the wiki name is not a foaf:name... |
| 22:50:01 | <tuukkah> | so i should make the sparql a bit more intelligent? |
| 22:50:14 | <ChNepote> | Yes :) |
| 22:50:26 | <ChNepote> | Why don't you use sioc:name ? |
| 22:51:19 | <tuukkah> | i don't know |
| 22:51:20 | <ChNepote> | On a wiki for example, the sioc:name is always defined because it's the wiki name of the person... |
| 22:51:35 | <tuukkah> | we can at least fall back on sioc:name |
| 22:52:32 | <ChNepote> | The good rule, in my opinion, would be : foaf:name if it exist and sioc:name if it doesn't |
| 22:52:56 | <tuukkah> | exactly. but sparql is ugly for this :-( |
| 22:54:52 | <tuukkah> | i did add duplicate result removal outside of sparql, so maybe now all we need is union... |
| 22:55:13 | <ChNepote> | I see. In fact sioc:name should be better so... many services and application use a sioc:name (ie an account name) but don't the real (foaf)name of their users |
| 22:55:16 | <tuukkah> | or not, because that doesn't express the the preference |
| 22:55:50 | <tuukkah> | i see what you mean |
| 22:57:10 | <tuukkah> | our smob client does know the sioc:name but it doesn't publish it properly :-( |
| 22:57:49 | <tuukkah> | for now, i'll add the union, ok? |
| 22:59:14 | <ChNepote> | ok ! |
| 23:01:27 | <tuukkah> | something like union { ?post sioc:has_creator ? user . ?user sioc:name ?name } |
| 23:06:02 | <ChNepote> | (sorry for being disconnected) |
| 23:06:25 | <tuukkah> | no problem, that's normal on irc |
| 23:07:53 | <ChNepote> | I am not sure I understand your piece of SPARQL (not so familiar with complex queries) |
| 23:08:06 | <ChNepote> | (the union part) |
| 23:08:15 | <tuukkah> | me neither :-( |
| 23:08:28 | <DeVries> | Problem detected. |
| 23:08:45 | <tuukkah> | DeVries, great. can you also solve it ?-) |
| 23:09:49 | <tuukkah> | ChNepote, i think i got it working. did you try to upload example data to the demo server already? |
| 23:15:28 | <ChNepote> | No. I work in localhost for the moment without SMOB server. |
| 23:16:33 | <ChNepote> | I will upload a prototype in a day or two, or three... |
| 23:16:51 | <ChNepote> | (on the web) |
| 23:17:20 | <tuukkah> | ok cool! let us now then |
| 23:17:27 | <tuukkah> | and we'll do more testing |
| 23:19:07 | <ChNepote> | All right. I'll tell you here. I am also in contact with Alexandre (I'm french, nobody's perfect :) |
| 23:19:17 | <ChNepote> | See you soon. |
| 23:19:22 | <tuukkah> | see you :-) |
| 23:21:26 | <DeVries> | tuukkah: Why should I solve it? |
| 23:21:39 | <DeVries> | If there are no problems, what's left to detect? |
| 23:22:21 | <tuukkah> | it would be slightly more useful - shall we say constructive - if you solved it too :-) |
| 23:32:31 | <DeVries> | Then again, it's me. |
| 23:32:45 | <DeVries> | I cause exceptions for the pure joy of detecting them. |
Back to channel and daily index: content-negotiated html turtle