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.
| 11:43:42 | <Wikier> | hi |
| 11:53:04 | <CaptSolo> | hi |
| 11:56:23 | <thosh> | hello |
| 12:02:53 | <thosh> | in the specification there is sioc:related_to: |
| 12:02:53 | <thosh> | "related_to - Related Posts for this Post, perhaps determined implicitly from topics or references." |
| 12:03:08 | <thosh> | so it has a domain and range of sioc:Post |
| 12:03:23 | <thosh> | but a property like that would also be useful for e. g. sioc:Thread - I could use such a thing for vBulletin's threads |
| 12:03:49 | <thosh> | unfortuanetly it can't easily be solved by changing related_to's domain and range to sioc:Item, because Thread is a sioc:Container |
| 13:36:01 | <CaptSolo> | hi thosh |
| 13:36:50 | <CaptSolo> | thosh: if so, we may have to drop domain and range of sioc:related_to altogether |
| 13:37:31 | <CaptSolo> | at least in RDFS it is not possible to say that a range of a property is ClassA _or_ ClassB |
| 13:38:20 | <CaptSolo> | for some other SIOC properties we "solve" this by dropping domain or range |
| 13:38:26 | <CaptSolo> | --- |
| 13:39:01 | <CaptSolo> | thosh: did you just update WP SIOC plugin to 1.24 or also patched the line to fix foaf:mbox_sha1sum bug? |
| 13:40:43 | <CaptSolo> | tuukkah: another cool feature of IRC log page would be if we could search logs for some text |
| 13:46:13 | <CaptSolo> | --- |
| 13:46:15 | <CaptSolo> | http://www.progbox.co.uk/wordpress/?p=461#comments |
| 13:46:25 | <CaptSolo> | an author is talking about printing his blog as a book |
| 13:46:59 | <CaptSolo> | has crossed my mind too |
| 13:47:44 | <CaptSolo> | and one of the commenters is saying he'd like a script too (so that he can adapt it) but that he is using Drupal and not WordPress |
| 13:47:54 | <CaptSolo> | now, this is where SIOC could help |
| 14:31:01 | <thosh> | CaptSolo: re. sioc:related_to - maybe there is a property saying somehting is "similar" or related to another thing in another ontology? |
| 14:31:44 | <tuukkah> | rdfs:seeAlso is something like that |
| 14:31:53 | <thosh> | CaptSolo: I didn't modify your WP SIOC plugin, just installed it. |
| 14:35:36 | <thosh> | tuukkah: yeah, kind of. still I'd like to have something more specifc. at the moment the various sioc exporters use rdfs:seeAlso just to show crawlers where to find more tripels for a given resource (as in paging) |
| 14:36:38 | <tuukkah> | thosh, indeed |
| 15:00:52 | <tuukkah> | so CaptSolo, is this a fair or an unfair implementation of search?-) |
| 15:17:38 | <CaptSolo> | tuukkah: what is? |
| 15:17:41 | <CaptSolo> | this |
| 15:17:59 | <tuukkah> | what i put up now |
| 15:18:06 | <tuukkah> | on the index page |
| 15:18:58 | <CaptSolo> | tuukkah: i see :) |
| 15:19:07 | <CaptSolo> | it's fair, a good start |
| 15:19:25 | <CaptSolo> | although i'd expect more from a search |
| 15:19:40 | <CaptSolo> | e.g., matching lines (+ maybe come context) |
| 15:20:21 | <CaptSolo> | saying that a word is mentioned somewhere in the 1000 lines long irc log is a good start, but still leaves some work to be done to find the actual mentions |
| 15:26:42 | <tuukkah> | that's right |
| 15:44:03 | <RKlyne_work> | CaptSolo: "at least in RDFS it is not possible to say that a range of a property is ClassA _or_ ClassB" - Are you sure about this? Wouldn't the triples [(x, rdfs:range, ClassA), (x, rdfs:range, ClassB)] do it? |
| 16:12:14 | <tuukkah> | and scor, now we have... a table with columns =) |
| 16:12:49 | <scor> | tuukkah: OMG, this is awesome! :D |
| 16:13:34 | <scor> | I like web usuability! |
| 16:13:58 | <tuukkah> | scor, what do you mean? |
| 16:14:52 | <scor> | tuukkah: I mean as it is this page is more readable. a human eye is able to flow the information in a quicker way now! |
| 16:15:36 | <tuukkah> | i see. so you prefer readable pages to unreadable pages =) |
| 16:18:03 | <scor> | tuukkah: yes, of course. make the web machine readable does not imply make the web human non readable! :p |
| 16:36:45 | <CaptSolo> | scor, tuukkah: what pages areyou talking about? |
| 16:36:55 | <tuukkah> | CaptSolo, the logindex still |
| 16:37:25 | <CaptSolo> | RKlyne_work: no. what those triples will do is state that range is ClassA _and_ ClassB |
| 16:37:38 | <CaptSolo> | e.g., if ClassA is a house and ClassB is a car |
| 16:38:17 | <CaptSolo> | then using a property with range defined as you wrote there would imply that the object of that statement is both a car and a house |
| 16:38:34 | <CaptSolo> | we stumbled upon that issue in the past. |
| 16:38:52 | <CaptSolo> | and had to remove statements which we erroneously thought to express "or" |
| 16:39:01 | <CaptSolo> | tuukkah: ah, the table! |
| 16:39:05 | <CaptSolo> | looks nice :) |
| 16:39:36 | <CaptSolo> | there's only 1 day which has log for #haskell2 |
| 16:40:28 | <scor> | CaptSolo: about http://tuukka.iki.fi/tmp/logindex.html |
| 16:40:29 | <tuukkah> | yes, it appears because of some testing i've done |
| 16:40:39 | <scor> | I find it more readable now |
| 16:40:44 | <scor> | more appealing |
| 16:40:54 | <tuukkah> | CaptSolo, gone |
| 16:53:32 | <CaptSolo> | scor: re. readable - well, sometimes table structure is less readable |
| 16:53:59 | <scor> | CaptSolo: less readable for who? human or machine? |
| 16:54:18 | <CaptSolo> | scor: it gives a better overview for when the logs are collected. |
| 16:54:22 | <CaptSolo> | oh, it's a good change :) |
| 16:54:25 | <CaptSolo> | thx, tuukkah |
| 16:54:42 | <scor> | CaptSolo: I don't get your point |
| 16:54:58 | <CaptSolo> | scor: what i mean is that a column repeating same thing on and on ("sioc, sioc, sioc, ...") becomes boring for a human |
| 16:55:09 | <CaptSolo> | and brain starts to get tired from it, filter it out |
| 16:55:52 | <CaptSolo> | just a general observation. i don't see how that can be taken into account to improve it. so don't consider this a critique. it's just an observation. |
| 16:56:18 | <scor> | maybe, but then you could just say SIOC in the <thead> with some nice green V |
| 16:56:59 | <scor> | but my point is it's visually easier to spot which days have been active or not (no logs or logs) |
| 16:57:13 | <CaptSolo> | scor: that's true |
| 16:57:30 | <scor> | CaptSolo: ;) |
| 16:57:34 | <tuukkah> | do browsers support thead so i should use it? |
| 16:58:23 | <CaptSolo> | tuukkah: dunno re. thread, my reply weas to it being visually easier to spot |
| 16:58:43 | <scor> | well, it's part of the standards http://www.w3.org/TR/html401/struct/tables.html#h-11.2.3 |
| 16:59:10 | <scor> | although I don't know what the browser exactly do with it. it depends. it's also used in CSS for theming |
| 16:59:34 | <CaptSolo> | scor: it's good as it is |
| 16:59:56 | <scor> | sure, I'm just knitpicking now! |
| 17:03:13 | <tuukkah> | i could put a tbody per year ;-) |
| 17:17:04 | <CaptSolo> | http://sw.deri.org/svn/sw/2005/08/sioc/media/ |
| 17:17:25 | <CaptSolo> | this is a directory with different SIOC-related multimedia materials |
| 17:17:34 | <CaptSolo> | mainly figures, images |
| 17:17:42 | <CaptSolo> | (in case if you did not know about this yet) |
| 17:18:31 | <CaptSolo> | vector graphics files often are in a .CDR format which is John's preferred illustration editing app |
| 17:20:58 | <tuukkah> | CDR? |
| 17:25:45 | <thosh> | tuukkah: it's a corel draw format |
| 17:26:15 | <tuukkah> | ok, right |
| 21:39:56 | <alder> | is there a little example in rdf/xml with two blog posts annotated with sioc where maybe the one references the other? |
| 21:41:20 | <alder> | with has_reply or related_to ? |
Back to channel and daily index: content-negotiated html turtle