Meetings/2008-10-01

= Meeting 2008-10-01 =

Details

 * When: Wednesday 1st October 2008 from 1830 BST (1730 GMT)
 * Where: #okfn irc channel on oftc.net (connect via Mibbit)

Agenda
Please feel free to add anything you'd like to discuss to the agenda...


 * Rob Myers: Open Art History
 * OKF projects on Cofundos?
 * Site overhaul

Participants

 * Mike Chelen
 * Rob Myers
 * Jonathan Gray
 * Rufus Pollock
 * Joshua Gay

IRC Log
IRC log will go here

(Time in CEST)

#!irc 19:06:21 	im not sure if someone has done this already, but i got started on a general OKFN calendar: http://www.google.com/calendar/embed?src=dnsgmdki4jfl78ib745i7485oo%40group.calendar.google.com 19:06:46 	its on google calendar so its available in html, xml, and ical formats 19:10:41 	hopefully the timezones adjust themselves for the user automatically 19:13:46 	oh cool 19:19:48 	i like having events in a calendar structure, theres no way i can keep up otherwise :) 19:24:38 	is there any page on the wiki where it would be okay to add a link? 19:26:01 	i made a separate one for the open textbook meetings as well, to help people get info on whichever events they are interested in 19:30:09 	hi mike! 19:30:22 	this is something we've been thinking of doing for a while... 19:30:27 	thanks for making the calendar! 19:30:52 	we should revamp a few revelant pages on the wiki.. 19:30:53 	e.g. http://okfn.org/wiki/Events 19:31:04 	also, i should make a general page on the main site for OKF events. 19:31:46 	hey there, sure thing 19:32:18 	it would be great to have a general calendar for open knowledge related events... 19:32:43 	i've been thinking of looking into open source packages we could install on our servers.. 19:32:53 	i don't know if there is a way to get the wiki software to output semantic event data, but its no trouble to set up google calendar 19:32:56 	hey robmyers! 19:33:06 	hiya 19:33:59 	it would be, google calendar can be set up so that any invitations sent to the calendar's email address are added automatically, that way many members of the community can contribute important / interesting events 19:34:20 	well.. perhaps i'll start something like.. http://okfn.org/events 19:35:40 	as well 19:36:10 	so we've got a record of which events are happening when in one place! 19:37:35 	good idea 19:37:53 	are we meeting now? :-) 19:40:29 	hey just to let you know, the wiki GUI editor is showing an error and won't load 19:40:50 	hmm.. thanks mike 19:41:03 	yes rob! 19:41:13 	goodgood 19:41:27 	sorry was just looking at our events pages and thinking about cleaning them up :-) 19:41:36 	how is everything going? 19:41:38 	heh 19:41:46 	is there an agenda? 19:42:04 	yep http://okfn.org/wiki/meetings/2008-10-01 19:42:29 	i've put open art history stuff first.. 19:42:32 	kewl 19:42:34 	oh wow 19:42:35 	ok 19:42:47 	you mentioned you were keen to chat about it! 19:42:50 	heh 19:42:57 	it would be great to hear how its going! :-) 19:43:13 	I've been researching standards for writing about artists and artefacts. 19:43:23 	so the Getty vocabulary and stuff. 19:43:27 	still working on that 19:43:35 	And I've been checking for similar projects. 19:43:40 	just to clarify - what kind of standards? 19:43:50 	oh how you describe an artwork 19:44:12 	so "large painting" (2008) 132"x20", oil on canvas, Rob Myers. 19:44:18 	ahh http://www.getty.edu/research/conducting_research/vocabularies/ 19:44:23 	there are standards. I was looking for machine readable ones as well 19:44:24 	yes 19:44:30 	brilliant 19:44:41 	I'm putting stuff on teh wiki page as I find it 19:45:25 	great! 19:46:01 	ah just looking at http://okfn.org/wiki/OpenArtHistory 19:46:35 	just looking - can you get a hold of the getty data very easily? 19:46:39 *	rgrp (~rgrp@92.17.26.1) has joined #okfn 19:46:53 	so we'd base the descriptions for artwork and biographical information on a known standard. preferably a machine readable one 19:46:55 	err I don't know 19:47:04 	but their description format is standard 19:47:27 	whatever we use should be free & open though yes 19:47:39 	hey rgrp 19:48:01 	robmyers: ok i see... 19:48:22 	if there is a standard which is easy for human & machine use, that would be great 19:48:29 	yes 19:48:43 	hi everyone 19:48:51 	a yaml format would be good for that 19:48:53 	heya rgrp 19:49:04 	i missed the start of this discussion 19:49:16 	but were we talking about metadata for descriing bontent (e.g. artworks) 19:50:20 	yes, for openarthsitory/openhogarth 19:50:24 	http://okfn.org/wiki/OpenArtHistory 19:50:29 	what about dublincore as a start? 19:50:31 	I've been looking for existing standards 19:50:45 	yes it's mentioned in some of the writing 19:51:20 	a human readable format might be good though. yaml or even javascript would be more readable than rdf 19:51:26 	imho 19:51:38 	and easier to get people working on 19:51:48 	true 19:52:02 	you can always pseudo convert dublic core 19:52:15 	oh? 19:52:16 	e.g. use yaml (or even just plain key/value pairs) based on dublin core 19:52:18 	e.g. 19:52:27 	yes yaml would be an ideal format 19:52:29 	dc-title: xyz 19:53:12 	we went down this route with econ data 19:53:15 	http://knowledgeforge.net/econ/svn/trunk/data/wheat-us/metadata.txt 19:53:43 	yes this kind of information is why it's good to be basing this on the work you guys have already done 19:54:23 	yes I'll definitely take a look at that, I think it's a good model. parsable by both machines and non-geeky people ;-) 19:55:02 	great! 19:55:37 	rdf can have standard form for ease of entry, but its true it can not be edited by hand as easily as yaml 19:55:46 	eys 19:55:47 	yes 19:55:49 	is there a standard which getty has adopted? 19:56:01 	for written descriptions. 19:56:25 	and they have a taxonomy/ontology I think, but we're not sure how open it is 19:56:45 	seems like they also have a registry of object ids for artefacts? 19:56:53 	oh right? 19:57:01 	the UN and others may have those as well 19:57:15 	if it's proprietary I may see an opportunity ;-) 19:57:54 	robmyers: I'm pretty sure Getty's ontology is proprietary in some way of other 19:58:15 	Object ID is a core standard for the identification of art, antiques and antiquities. It has been created as part of the Getty initiative to protect cultural objects and includes images among its categories. In the event of theft/loss Object ID is intended to allow the quick transmission of information about objects among museums, police and customs agencies, the art trade and collectors. 19:58:22 	ah. so an open ontology might be a good idea. I'll avoid looking at theirs then. 19:58:48 	they have a publication about their recent electronic calatloguing efforts at http://www.getty.edu/grants/pdfs/LA_Art_Online_Report.pdf but i havent read it all 19:58:52 	yes that's a bit outside of the scope of openarthistory 19:58:55 	oh ok cool 19:59:29 	If people could please add things they find to the open art hostory wiki page that would be very useful. there's a standards section. 19:59:30 	you should check to be sure -- I last looked at that sort of thing several years ago ... 19:59:37 	http://okfn.org/wiki/OpenArtHistory 19:59:41 	sure, I'll check 20:00:00 	it might be a good subject for advocacy, if they would consider migrating to an open standard 20:00:51 	yes 20:00:55 	definitely 20:01:28 	it would be great to hear about what your plans are for open art history! 20:02:15 	are you intending to initially collate scans and some metadata? 20:02:22 	oh well just openhogarth for now, as a pilot. but if more people get involved it would hopefully be a good resource for the study of art history by people and machines. 20:02:23 	yes 20:02:49 	great 20:02:54 	the metadata is the sort of thing you get in catalogues. title, year, artist, medium, size, any interesting ntoes. 20:02:55 	notes 20:03:27 	basically a little hogarth package :-) 20:03:33 	yes 20:04:16 	my parents had a couple of prints on the wall when I was a kid, so that's probably why Hogarth. ;-) 20:04:33 	great, just clarifying.. 20:04:46 	sure. no megalomania. yet ;-) 20:04:50 	sounds great 20:05:04 	so, perhaps we could have a todo list somewhere? 20:05:35 	yes. on the wiki page or do you have a todo list system somewhere? 20:06:47 	we do have http://okfn.org/wiki/TodoList 20:07:03 	but it might be more appropriate on the openhogarth wiki for now... 20:07:29 	yes I'll pop it on there later on 20:08:06 	cool! 20:08:23 	good to break it down to see what we need to do first! 20:08:26 	yes 20:09:10 	are you thinking of using knowledgeforge initially? 20:09:24 	there's a svn repo for putting images into.. 20:09:29 	yes I've set up a subversion repository there 20:09:50 	http://www.knowledgeforge.net/project/openhogarth/ 20:09:51 	cool stuff 20:10:02 	I couldn't see how to set the licence to PD 20:10:03 	great! 20:10:14 	so it's a bit misleading at the moment 20:10:21 	robmyers: yes i know 20:10:23 	:-) 20:10:29 	we need to sort out available set of licenses 20:10:43 	So yes the immediate task is to set metadata and attribution standards, then set project repository structure, then start uploading images 20:10:45 	i'd like to unify the license stuff across knowledgeforge/ckan/microfacts etc 20:11:09 	rgrp: that would be great! 20:11:27 	robmyers: sounds like a good way to proceed.. 20:11:36 	let me know if there's anything i can help with! 20:11:41 	k I'll add that to the wiki 20:11:43 	thanks! 20:11:57 	rgrp: how is it going with microfacts? 20:13:34 	a couple of things it'd be great to hear view on.. 20:13:43 	fine 20:14:13 	we were thinking of re-designing main site.. 20:14:18 	at moment I've switched slightly to getting next version of ckan done (major upgrade to switch from sqlobject to sqlalchemy) 20:14:31 	ahh. great! 20:15:36 	i was going to ask - do you think its worth putting any okf projects on cofundos? 20:15:43 	we discussed this a while back.. 20:15:53 	what's cofundos? 20:16:07 	http://www.cofundos.org/ 20:16:48 	its a platform for community funding/developing software/knowledge projects.. 20:16:55 	oh cool 20:16:57 	by soeren auer, who's on okf advisory board.. 20:17:26 	i was thinking we could list something small.. 20:17:34 	e.g. a browser plugin for ckan 20:17:47 	rgrp: do you think this would be hard to do? 20:18:47 	i don't think so 20:19:08 	it might be a nice discrete project to try running on cofundos.. 20:19:45 	or perhaps you can think of something else? 20:20:33 	what do you think robmyers? :-) 20:20:34 	what kind of plugin would be helpful for ckan? 20:20:56 	finance is good ;-) 20:21:03 	but yes what would a ckan plugin do? 20:21:52 	perhaps a bit like delicious plugin - to help make it easier for people to add packages? 20:22:06 	i guess the obvious thing would be something like the del.icio.us js plugin that auto submits relevant info 20:23:08 	perhaps also something like tag cloud for browsing tags? 20:23:27 	thats what came to mind, its possible to do that through a bookmarklet (i think) 20:23:36 	that might be a good project for cofundos or any other system, since it's relatively small and achievable, but still requires some technical collaboration 20:23:36 	yes it is really a bookmarklet 20:23:52 	effort would be to fix up ckan to take that data in easily 20:24:43 *	jgay (~jgay@terminus-est.gnu.org) has joined #okfn 20:25:20 	do you think it would be worth listing on cofundos? 20:26:33 	it sounds like a project with a lot of potential. but I don't know enough about cofundos, whether okfn would have the mindshare or draw to make it worthwhile yet. 20:26:41 	heya jgay 20:26:57 	robmyers: howdie 20:27:04 	hey joshua! 20:27:09 	jwyg: hey 20:27:34 	tag cloud might need to be done on the server side, if there is any way the server can offer up tag frequency as XML or JSON or something then someone could work on javascript to display it 20:27:52 	a data point bookmarklet would be cool 20:29:15 	making a bookmarklet is not too difficult, but it is a good thing for people to collaborate on because it doesn't require a whole lot of setup, just write some javascript and share it with the other people involved, who can try to improve it 20:29:24 	i dont have any experience with cofundos either :) 20:30:54 	ok.. 20:31:14 	i might draft a couple of ideas on the wiki.. 20:31:36 	in general, i think it would be great to give this a go! 20:31:40 	bookmarklet is i think the best cost/value proposition to get started with 20:31:50 	yes it sounds good 20:32:11 	id help work on a bookmarklet though, wherever it is organized =) 20:32:34 	it looks like they recommend tagging with relevant languages, components, etc... 20:33:03 	that would be great mike! 20:34:02 	what do you think i should list? 20:34:16 	Please enter as many appropriate tags as possible, in order to characterize your project: Which programming language, database, query language etc. has to be used. On which open-source components is it based or related to. 20:35:20 	any ideas? 20:36:33 	javascript 20:37:02 	prototype, they all use prototype ;-) and microfacts? 20:38:29 	(added todo to OpenArtHistory wiki page) 20:40:10 	actually we're using mootools for microfacts 20:40:20 	(though i think pylons still ships prototype for 'free' 20:41:01 	html 20:41:48 	added very brief outline to cofundos.. http://www.cofundos.org/json.php?output=html&project_id=130 20:42:06 	just javascript and html for now, unless you want to include any changes on the server side 20:42:39 	mootools is cool, yes 20:45:38 	i think you mean http://cofundos.org/project.php?id=130 20:45:39 	yes it does seem reasonable though i'm no great js expert ... 20:47:36 	ok I need to go now. bed time for my youngest. 20:47:40 	good meeting 20:47:41 	thats right mike :-) 20:47:48 	ok robmyers! 20:47:55 	good to chat to you! 20:48:06 	cheers 20:49:08 	i need to go to ;) 20:49:31 	good todo list on wiki page.. 20:49:33 	good to chat with everyone and the bookmarklet is a really good idea 20:49:35 	i should head too.. 20:50:27 	alright.. 20:50:44 	hopefully see you all next week! :-) 20:51:06 	mikechelen: good to see you.. 20:51:36 	'til next week ... 20:51:42 	mikechelen: do you think the cofundos proposal looks alright (albeit very brief)? 20:51:59 	thanks, great to hear about all this stuff going on with okfn :) 20:53:20 	jwyg, yes, well i added a comment, should it say "plugin" if all that is needed is a bookmarklet? 20:54:00 	a browser plugin can have many more features, but is also significantly more challenging to create 20:54:05 	ok.. 20:54:22 	i'll change the name :0) 20:55:50 	and add HTML? 20:55:57 	or anything else to tags? 20:56:23 	if there are more features that are thought of, it might be a good project to follow the bookmark 20:56:35 	great... 20:56:50 	i'd put HTML in the tags, but not in the title, "bookmarklet" is pretty self explanatory for people that might help 21:01:03 	good to see people 21:01:04 	o&o 21:01:05 *	rgrp has quit (Quit: leaving) 21:02:59 	it's up: http://howwikipediaworks.com/ 21:03:19 	Phoebe just IM'd me. Pretty awesome book. No Starch (subsidiary of O'Rielly) is publishing it under the GDFL 21:04:33 	brilliant! 21:04:43 	is this just out jgay? 21:05:59 	jwyg: the book has been out a bit, but the web version just came out, now. 21:06:17 	the book has maybe been out a couple weeks, I think 21:06:37 	brilliant! 21:06:39 	no starch is really cool in that they'll let authors choose their license 21:08:05 	that's fantastic" 21:08:07 	! 21:09:18 	jgay: i should shoot, but its great to see you here! 21:10:00 	i'll blog new launch of tbr tomorrow... 21:10:12 	meant to do today, but too much academic work.. 21:11:58 	jwyg: no worries 21:12:05 	thanks for your help. 21:13:54 	no problem 21:13:56 	:-)