Experimental IRC log swig-2007-07-07

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.

14:52:07<timbl>ACTION wishes everyone a happy 7/7/7
14:53:00<vant>it's almost over here (tokyo) :(
14:54:06<timbl>Well, I should have said it at 7:07 in SOME timezone too!
14:58:43<timbl>A good time for adding new features, the date portends.
14:58:51<timbl>ACTION adds @base to his n3 parser
14:59:01<timbl>but not to the spec
14:59:46<danbri>"777 is a qabalistic dictionary of ceremonial magic, oriental mysticism, comparative religion, and symbology. It is also a handbook for ceremonial invocation and for checking the validity of dreams and visions. It is indispendisble to those who wish to correlate these apparently diverse studies. It was published privately by Aleister Crowley in 1909, has long been out of print and is now...
14:59:48<danbri>...practically unprocurable. "
15:01:39<timbl>However, the essential knowledge distilled from it is now available from a sparql end-point?
15:01:50<timbl>How are you, danbri?
15:02:22<danbri>actually it does appear to be largely tabular in form, consisting of contolled name lists and hierarchies... doubtless someone will get it converted!
15:03:08<danbri>i'm just fine thanks! off again to rome (asemantics) and netherlands (joost) soon...
15:03:13<danbri>and yourself?
15:03:47<timbl>Well, the eather in Boston at this time of year makes it silly to leave. very nice at the moment.
15:04:08<timbl>Good haveing students around making tabulator progress.
15:04:11<danbri>it's not airconditioner season yet?
15:04:26<timbl>It's open window season
15:04:38<timbl>(That means anyone can shoot a window)
15:04:53<cerealtom>danbri: any more thoughts about/progress with hosting for the t-shirt ontology?
15:04:53<timbl>(no reference to desktop standards)
15:05:01<cerealtom>and hi btw :)
15:05:07<danbri>hi!
15:05:11<cerealtom>hey :)
15:05:31<danbri>sorry, nothing specific w/ tshirt thingy yet, been focussed on the foaf side o things
15:05:56<danbri>but if you're happy with the ns going into storage alongside foaf, whenever that happens (long-term storage ie. frozen / preservation) we could just proceed...
15:06:18<danbri>i'm going to go to the DCMI conference later this year, talk about that with DC folks
15:06:24<cerealtom>yeah, i think kasei and i were happy with the arrangements/what you could offer
15:06:33<cerealtom>and keen to go ahead :)
15:06:59<danbri>ok cool. next steps - can you email me a couple preferred usernames for svn/cvs?
15:07:20<cerealtom>sure, will do. that would be great.
15:07:20<timbl>t-shirt ontology? serious or not so?
15:07:30<cerealtom>timbl: sure, serious!
15:07:40<cerealtom>for t-shirt size and colour prefs
15:07:55<danbri>(derranged crowley rantings - http://www.thelemapedia.org/index.php/Tree_of_Life:777 ...wiki but not semwiki)
15:08:00<cerealtom>so hopefully somewhere down the line there'll be no more oversized conference t-shirts
15:08:06<danbri>I think US and UK sizes are different...
15:08:41<cerealtom>indeed. we figured that some approximation was better than nothing, without getting into the rather personal territory of "preferred chest size", if you see what i mean
15:09:10<danbri>you do also need gender - medium female shirts are smaller
15:10:10<danbri>reminds me - http://www.theonion.com/content/video/gap_unveils_new_for_kids_by_kids
15:10:49<timbl>Well, to actually produce an ontology which would allow a shop to generate the my measuremnets -> your (shops) size mapping woukld be useful
15:11:30<danbri>sliperyslope into 3d avatar portability...
15:12:10<timbl>:) welll more like http://www.landsend.com/cd/popup/0,,copyName=pop_size_1001_mn_tops.html
15:12:36<cerealtom>yeah, point taken danbri (and lol about theonion vid). we decided to try and keep it fairly simple, on the basis that conf organisers were unlikely to cater to infinite different options, and that most people would be happy to have something closer to their preference than wildly out, which is often the case
15:12:40<cerealtom>cf ISWC2006!
15:12:51<cerealtom>agreed timbl
15:12:55<timbl>There are 18 similar tables just for that store http://www.landsend.com/cgi-bin/search.cgi?store=le&action=newSearch&search=sizes&imageField.x=0&imageField.y=0
15:13:17<cerealtom>latest draft is at http://kasei.us/2007/06/tshirt.rdf
15:14:13<timbl>There are often issues with transatlantic misunderstandings of "large" for clothes as well as fro food.
15:14:28<cerealtom>timbl: can you say some more on that?
15:15:17<cerealtom>you mean as in north american large = euro huge?
15:15:24<danbri>we can also plot our physical decline as a community, over the years...
15:15:32<timbl>Well, UK people when registering for a conf in the US may click 'large' and find that they expected a UK large but got alarger US large.
15:15:44<cerealtom>lol danbri.
15:15:48<cerealtom>yes, agreed timbl
15:16:05<timbl>Actually there was a famou blog thred (who?) about onference tshirt sizes which I am sure you have followed
15:16:23<cerealtom>yes, i think so, but cant remember the source either
15:16:46<cerealtom>http://headrush.typepad.com/creating_passionate_users/2006/12/tech_tshirts_ar.html
15:16:49<cerealtom>?
15:17:32<danbri>timbl, re tabulator ... is there a full sparql impl coming together there?
15:19:04<cerealtom>re the different interpretations of large, yes, its not perfect, but we figured that getting closer to what people liked was preferable to doing nothing, or getting overly complex. anyway, interested to hear if you think this is a deal breaker/insurmountable following our current approach
15:19:04<timbl>Well, Yosi is working on SPARQL in cwm. That migyt convert I guess. But teh tabulator internal sparql system is not complete.
15:19:13<timbl>It does have optional thouugh
15:19:20<timbl>re tshirts: here http://headrush.typepad.com/creating_passionate_users/2006/12/tech_tshirts_ar.html
15:19:53<cerealtom>i hope we might be able to do something with the t-shirts for ISWC2007, though lead time might be too short
15:21:01<timbl>for exampel see "Most XXL short-sleeve shirts fit pretty well, but believe it or not, I find shows often run out of them. This is especially galling if the show asked for my t-shirt size on the registration form (which, really, every show should if they're handing out shirts.)" in comment to that blog
15:22:04<timbl>Just require that people give their FOAF and if they have t-shirt size on their FOAF then they won't need to give it
15:22:21<timbl>But probably must people won't share it publically
15:22:29<danbri>oh, hadn't heard much of cwm dev lately... i thought in some ways tabulator was the new cwm
15:22:51<timbl>Tabulator is different.
15:23:09<timbl>It is very 'live' .. loos up URIs whenever it needs to
15:23:26<timbl>A bit like cwm with lots of mode glags turned on, but more so.
15:23:43<timbl>but the n3 parser i converted from the python one with pyjs
15:23:53<danbri>oh, interesting
15:24:23<danbri>hadn't heard of pyjs
15:25:24<timbl>http://groups.google.com/group/mochikit/msg/d0f967fd9c914fb8
15:25:52<timbl>B:|Steve Wedig on pyjs
15:25:56<cerealtom>timbl: agreed re the sharing of this info, which was why we went for preferred size, on the basis that expressing a preference is less personal that size myTshirtSize. Some people may prefer a size very different to what would fit them, as they give conf t-shirts to their kids/partner/gym buddy etc. so i'm hopeful that people will bother to express it in their foaf
15:26:31<cerealtom>on the basis that you cant infer much about them physically from this
15:27:25<cerealtom>anyway, we can give it a go, and if its a miserable failure just put it down to experience, and remind ourselves that it was fun hacking it over a beer in innsbruck
15:28:12<cerealtom>http://www.flickr.com/photos/gromgull/534331504/
15:28:54<danbri>ACTION finds http://goeurope.about.com/cs/shopping/l/bl_clothes_size.htm
15:29:04<danbri>but it only puts us labels against the numbers
15:31:26<cerealtom>presumably a combination of foaf:based_near and foaf:gender could get us somewhere nearer to people's actual preferences/expectations, although we may also need a "preferred form factor" (!) or whatever, to express the preference for male or female fit/sizes
15:31:51<cerealtom>so people can say i'm foaf:gender "male", but prefer a womans fit t-shirt (!)
15:31:54<cerealtom>;)
15:31:55<timbl>B: Steve Wedig's converter from python to Javascript.
15:41:59<danbri>we need a button for "I'd like an XL tshirt but i'd prefer to fit a medium"....
15:45:42<cerealtom>lol danbri. isnt the second statement something for the DietPlan ontology?
15:46:27<danbri>or a todolist
15:47:35<timbl>B: Nifty.
15:48:46<timbl>B: Used this to convert cwm's n3 parser (in python) to Tabulator's (in js)
15:49:53<cerealtom>danbri: indeed!
16:09:42<timbl>cerealtom, I'd suggest just a singular label for a class, llike "color"
16:10:03<timbl>For example, the tabultor can make a plural if it needs to, and also can capitalize.
16:10:20<timbl>Singular is the norm
16:10:51<timbl>You have hex rep
16:10:52<timbl>"http://www.w3.org/2000/10/swap/test/n3/n3-rdf.tests"
16:11:02<timbl>sorry
16:11:14<timbl>I mean Hexadecimal representation of a colour
16:11:46<timbl>that is a long label .. more like a s:comment. Maybe "hex code" for a label to be used on a form?
16:12:29<timbl>In the descriptions I would say that it is SRGB (or whatever it is) Chris Lilley is a good person to involve if doing color
16:13:07<cerealtom>timbl: re colours, sure, fair point. it was mainly tongue in cheek as neither the brit and the american hacking it wanted to accept the others spelling ;)
16:13:29<timbl>Oh, i'm not worried about the 'u' :)
16:13:37<cerealtom>:)
16:13:55<cerealtom>cool. kasei and i will argue about it again and decide on one label, whichever it is!
16:13:57<timbl>Is 'http://xmlns.com/tshirt/ExtraExtraLargeSize' a class of t-shirts?
16:14:16<timbl>Wuld I say that my particular shirt is a member of the class
16:14:55<timbl>In that case, the superclas would be tshirt. In gace you have a superclass http://xmlns.com/tshirt/Size
16:15:24<timbl>maybe you have classes of classes ./. ExtraLArge is a class of shirt, and it is type Size.
16:15:53<timbl>I'm tabulating around teh ontology and just mentioning things as I find them
16:16:13<cerealtom>great, thanks!
16:16:32<cerealtom>just refamiliarising myself with the modelling
16:19:25<timbl>Looks useful with that superclass sorted out ... look forward to it turnin up at a conference
16:23:44<cerealtom>thanks timbl. as i say i hope it may turn up at ISWC2007, but will need to speak to the rest of the Org Ctte and see what kind of uptake we get from potential delegates
16:23:51<cerealtom>as for the modelling...
16:24:57<danbri>ACTION wants to buy some Sulphur Hexafluoride
16:25:07<cerealtom>http://xmlns.com/tshirt/XXLarge is an instance with the type http://xmlns.com/tshirt/ExtraExtraLargeSize, if that makes sense
16:25:17<danbri>(after seeing http://www.youtube.com/watch?v=mb3ktPn1MQk)
16:26:00<cerealtom>so that way we can say that someone has http://xmlns.com/tshirt/preferredSize -> http://xmlns.com/tshirt/XXLarge
16:26:03<danbri>re colours, apparently http://en.wikipedia.org/wiki/X11_color_names --- x11 and css/svg colours are pretty close
16:26:24<danbri>"It is not known who originally compiled the list. The list shows neither a continuity in selected color values nor in color names, and many color triplets have multiple names. Despite this, graphic designers and others got used to them making it practically impossible to introduce a more stringent and logical alias list."
16:26:29<danbri>the names are great, at least...
16:27:44<cerealtom>timbl: re colours, thanks for the tip re Chris Lilley, will follow up
16:28:59<cerealtom>yes, brilliant. i like the name BlanchedAlmond, if not the colour
16:30:22<cerealtom>this is a nice set danbri. a good candidate for replacing the 16 currently defined
16:30:27<cerealtom>kasei?
17:06:57<danbri>http://users.utu.fi/micnel/BEC/BEC%20Neg%20Key%20Words.htm
17:07:30<danbri>E:|Words found to occur with unusual INfrequency in the Business English Corpus
17:08:07<danbri>E:From [http://users.utu.fi/micnel/business_english_lexis_site.htm|Mike Nelson's Business English Lexis Site]
17:10:02<danbri>E:Another [http://en.wikipedia.org/wiki/X11_color_names|strangely evocative] list...
17:11:51<danbri>E:"The analysis showed that Business English was semantically divorced from lexis concerned with personal issues, society, family, house and home and personal activities. Also used far less in business was lexis referring to distinctly negative states, and words used to express deep, reflective and emotive feelings were found to occur significantly less - the word ‘truth’ for example,...
17:11:53<danbri>...was found to occur nine times less in Business English than in general English."

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