Agenda March 21 Scientific Discourse call
Dial-in & IRC Information
- Time: 15:00 GMT / 10am EST / 7am PST
- Dial-In #: +1.617.761.6200 (Cambridge, MA)
- Dial-In #: +33.4.26.46.79.03 (Paris, France)
- Dial-In #: +44.203.318.0479 (London, UK)
- Participant Access Code: 42572 ("HCLS2")
- IRC Channel: irc.w3.org port 6665 channel #HCLS2 use IRC direct link or (see W3C IRC page for details, or see Web IRC)
- Mibbit quick start: Click on mibbit for instant IRC access
- Duration: 1hr
IRC Transcript:
09:56 *** paoloC joined #hcls2
09:57 *** barend joined #hcls2
09:58 *** mscottm joined #hcls2
09:58 *** Tudor joined #hcls2
09:59 Anita HI all! Can you all dail in?
09:59 *** Joanne joined #hcls2
09:59 Anita Details here...http://www.w3.org/wiki/HCLSIG/SWANSIOC/Actions/RhetoricalStructure/meetings/201 10321
10:01 Anita Hi Paolo, Barend and Scott - are you on the phone?
10:01 *** Howard joined #hcls2
10:04 *** andrewg joined #hcls2
10:05 *** andrewg quit (Quit: http://www.mibbit.com ajax IRC Client)
10:05 *** bbalsa joined #hcls2
10:05 *** david_r_newman joined #hcls2
10:06 david_r_newman Is anyone else having problems with the conference bridge. I am gettting "All lines are busy" on the UK bridge number
10:07 *** andrewg joined #hcls2
10:08 Anita Hi David Newman did you get in?
10:08 *** andrewg quit (Quit: http://www.mibbit.com ajax IRC Client)
10:08 Anita Anyone else having problems on accessing the phone lines?
10:08 Anita Barend: now start measuring actual citations of nanopublications
10:08 *** andrewg joined #hcls2
10: 09 *** Gully joined #hcls2
10: 09 *** Zakim joined #hcls2
10: 09 david_r_newman Anita: No I am using Skype from the US number
10: 09 *** RRSAgent joined #hcls2
10: 09 RRSAgent logging to http://www.w3.org/2011/03/21-hcls2-irc
10: 09 Anita Thanks - that's working ok then?
10: 09 david_r_newman Anita: But I am now in
10: 10 Howard Very nice sequence from slide 5 to 6 showing how nanoPubs solve current problems with strictly narrative papers
10: 10 Joanne it's hard to hear Barend, but am getting most of it. ... would be useful, once he goes throught he overview to go into a specific example and walk us through something
10: 10 Anita If you want to describe the variants in the underlying databases there are hundreds of thousands - easily mine 40.000 nanopublications from an abstract but there are really only 600 in them (Scott/Barnd did I get that right?)
10: 10 david_r_newman Yes, I hopefully won't drop out from lack on credit. Think I have enough for this call.
10:11 Anita No - paper is textual review of LOPDs mine text manually 100.000 nanopubs; only 40.000 make it into the paper
10:12 Anita If you look in general for variants: in literature frequency of variants is very low - is only in the databases
10:12 mscottm BareBarend was referring to LOVD - Leiden Open Variant Database
10:12 *** philippe joined #hcls2
10:12 Anita Scott can you pse (co-)scribe - I'm not sure I'm catching all the nuances - thanks!
10:13 Anita Good David glad you're here - nr should be toll-free though??
10:13 mscottm I'll do my best. I find it also difficult to hear Barend even though he's about a 20 min walk away. !
10:13 Joanne Someone (Gully?) is asking how the nanopublicaions are structured
10:13 Anita Gully: how is nanopub represented? Barend: we're just making that - Scott will explain details!
10:14 mscottm http://www.mendeley.com/research/the-anatomy-of-a-nanopublication/
10:14 Gully nanopub.org is the site where this will be put
10:15 *** matthias_samwald joined #hcls2
10:16 matthias_samwald (just joined the call)
10:18 Anita Details of RDF is to be determined - Herman Verhagen did mining of LOVD - Leiden Open Variant Database: http://www.lovd.nl/2.0/
10:19 Gully Scott: do you have a group on Mendeley for nanopublication developers?
10:20 Anita A certain SNP has a variation frequency in a population - "cognitive deficits relate to BACA-1 gene" - see many things that refer to this core assertion
10:20 *** barend quit (Quit: http://www.mibbit.com ajax IRC Client)
10:20 Anita Joanne: can we get a bit of background? What makes something a nanopublication?
10:22 Anita Eric (LastName?) - nanopub is an assertion, with provenance and context, which defines when it is true or not true -
10:22 Anita Is there a difference between provenance and context?
10:23 Anita Typical things you reference in a research paper - Author, journal, date, DOI - are different from conditions critical to understanding of the data and the assertions
10:23 Anita Or: this SNP has a frequency of 2.5% only in a given population
10:24 Anita Does this mean that the reference is the provenance?
10:26 Anita Anita: do you take the opinion of the author or the experimental data as the context?
10:27 Anita Eric: either the data or the opinion - so nanopublication has no hedging: "Protein A interacts with Protein B"
10:27 Howard I like finer grained sciDiscourse; is there a good way to link claim to warrant?
10:28 Anita Provenance and context will tell you how believable this is -
10:29 Howard I am not clear on the link from P&C to warrant: the warrant is theory or model dependent.
10:36 Anita Joanne - yes, so is this like fine-grained structure?
10:36 Anita Anita: yes-ish: I'm not clear on how nanopub comes out of a paper. Is it only for databases?
10:37 Anita Eric: much more exposed data in databases than in papers - easier than text-mining literature. So: with high-throughput biology and unambiguous data, we want to make a system for publishing data points.
10:38 Anita Eric: create a system that is recognised by e.g. Nature and Thompson to publish thier *data* as nanopublications
10:40 mscottm Erik Schultes
10:40 Anita Like to get rid of the idea of 'supplemental data': but publish data as a first-order citizen and then publish paper as an afterthought
10:40 Howard The paper is the pre-thought: what made the data count as data
10:41 Anita Joanne: went to earth science data conference: they publish data all the time! Let's not duplicate efforts.
10:41 mscottm Re: Gully's question "Scott: do you have a group on Mendeley for nanopublication developers?" - it sounds like a suggestion. Should we start one?
10:44 Anita Joanne are you still talking next week about the GeoData stuff?
10:45 Gully Re: Mendeley group. Yes, would you like to start it or shall I?
10:46 Joanne Anita -- I'm not sure that I will have time to prepare (sorry to keep putting this off, but am working through a backlog created by travel and moving)
10:46 Joanne http://www.sciencemag.org/content/331/6018/705.abstract (this is the science article I was talking about -- it's focus is on data visualization). I'll post the link next to the GeoData2011 workshop
10:49 Joanne GeoData 2011 -Geo-Data Informatics: Exploring the Life Cycle, Citation and Integration of Geo-Data http://tw.rpi.edu/web/Workshop/Community/GeoData2011
10:49 Howard Erik: nanoPubs have "flavors" or semantic types: these types may vary within biology domains
10:49 Joanne Note- the term citation is used instead of publication here - but we're talking about the same thing
10:50 Howard Erik: microCitations within a hemoglobin database can be used to track usage of the dataset
10:51 paoloC Anita: I have to drop out in a minute, just wanted to say that if nobody else has further remarks/comments about ORB, the draft is ready to go.
10:51 Anita LODD just published data publishing provenance recommendations - just been submitted; now a paper
10:52 mscottm http://www.w3.org/blog/hcls
10:55 mscottm mscottmarshall@gmail.com
10:56 Gully thanks
10:56 *** paoloC quit (Quit: CGI:IRC)
10:57 *** philippe quit (Quit: http://www.mibbit.com ajax IRC Client)
10:57 david_r_newman Creative Commons is a good starting point for licensing - http://creativecommons.org/ns#
10:58 *** paoloC joined #hcls2
10:58 paoloC Nanopublications and SWAN: http://www.slideshare.net/paolociccarese/alzswan-hyque-and-nanopublications
11:00 Gully http://www.w3.org/wiki/HCLSIG/SWANSIOC/Nanopublications-Subtask
11:01 Anita List of ontologies is here: http://www.w3.org/wiki/HCLSIG/SWANSIOC
