See also: IRC log
<antoine> Previous: http://www.w3.org/2005/Incubator/lld/minutes/2010/12/02-lld-minutes.html
<antoine> Scribe: Jeff
<antoine> scribenick: jeff_
<antoine> previous call: http://www.w3.org/2005/Incubator/lld/minutes/2010/12/02-lld-minutes.html
RESOLUTION: accept minutes of previous meeting
<scribe> ACTION: Remove PAUL WALK'S TOOL AND ANALYSIS from Agenda [recorded in http://www.w3.org/2005/Incubator/lld/minutes/2010/12/02-lld-minutes.html#action02] [DONE]
<scribe> ACTION: Antoine to add "LOD curriculum" into "recommendations" section of Outline, and benefits into an Introduction [recorded in http://www.w3.org/2005/Incubator/lld/minutes/2010/12/02-lld-minutes.html#action03] [DONE]
<antoine> http://www.w3.org/2005/Incubator/lld/wiki/index.php?title=RunningAgenda&oldid=2306
antoine: This week's focus is on the archives cluster.
<antoine> http://www.w3.org/2005/Incubator/lld/wiki/UseCases#Cluster:_Archives_and_heterogeneous_data
<emma> http://www.w3.org/2005/Incubator/lld/wiki/Cluster_Archives
karen: The use cases for archive linked data include targeting other institutions. ... Linked Data can be used to facilitate consortia building.
antoine: Is there overlap between library and archive data?
emma: Yes, linked data can be
used to federate access data across library archives and
museums
... by relating items in generalized terms of "subject", "people",
"places", etc.
... These broader terms allow archives to relate heterogeneous materials.
karen: the connections between disparate things can be made in topical ways.
<antoine> context?
emma: the archives cluster is unusual because it is less concerned with common metadata vocabularies and more concerned about context.
<marcia> +emma
karen: Archives have different kinds of goals.
<emma> http://www.w3.org/2005/Incubator/lld/wiki/Cluster_Archives
emma: the cluster document includes goals
and discussion from publisher and user POVs.
... DESCRIBE is about creating information
... SHARE is concerned with object descriptions beyond bibliographic items.
... The idea is to share information between
insitutions as much as communicate with end users
... with a need to map or share common models across different domains.
<antoine> +1 FEDERATE as a kind of DISCOVER/SEARCH
emma: FEDERATE: is a form of DISCOVER over
many sources and different domains.
... MANAGE: is different from
discovery in the sense of professional use (e.g. library management).
antoine: these goals seem to be higher levels than the other general goals that have been proposed.
karen: the DESCRIBE goal is more of a cataloging function and seen as broader than identification.
antoine: Does FEDERATE relate to
AGGREGATE?
... is AGGREGATE a "means" for FEDERATE?
<Asaf> there's federate, and there's federated search.
karen: there are subtle differences in meaning
<Asaf> to me, federate is the one-time establishing of an interface between two data stores, which would facilitate future federated searches.
emma: These seem to be different perspectives ... on how related resources are clustered.
Asaf: FEDERATE puts interface in
place
... could be a federated SPARQL query
karen: The difference between FEDERATE and AGGREGATE does need to be clearer.
<Asaf> FEDERATE: offer search capability over mutiple data sources
emma: FEDERATE assumes several stores and is thus different from AGGREGATE
karen: Presumably both fall under search and discovery.
antoine: AGGREGATE seems to be an act on its own.
<marcia> federated seemed to target to make deep web documents searchable
karen: FEDERATE is an action performed on more than one store.
<marcia> those not on surface to be aggregated by, for example, search engines
emma: The goals in this cluster may not align
with general goals
... they seem to be higher-level
<monica> one other way to define federate and aggregate is, federate is send a search term to remote source, and bring back only results, aggregate is to pool all possible metadata locally then do the search
antoine: it seems OK to add these goals to the general list and worry about reconciling them later
<marcia> I am in meeting
<marcia> *in meeting all day
<monica> that's a specific Z39.50 vs OAI-PMH distinction though so will probably apply to only a subset of our audiences
antoine: marcia is invited to follow up via email
<marcia> * OK.
antoine: Moving on to cases, Europeana may fall into several of these use goals.
emma: want to go through the concrete use cases again to make sure the mappings to various goals are accounted for.
<marcia> *need to go now. Will folow up on email.
antoine: all are encourage to send feedback
emma: A template for curating clusters would help guide the development of the document.
<emma> see citation cluster http://www.w3.org/2005/Incubator/lld/wiki/Cluster_Citations
antoine: citation cluster is well
aligned to suggestions last week
... 1) background
... 2) extracted use cases and scenarios
...3) relevant technologies and vocabularies
... but we need allow for some freedom
... it seems possible that all clusters could be aligned in this
general direction, though.
<edsu> kai: thanks
<edsu> emma pasted above
<antoine> http://www.w3.org/2005/Incubator/lld/wiki/Cluster_Citations
<scribe> ACTION: Citation cluster group to place Google Docs document into XG Wiki [recorded in http://www.w3.org/2005/Incubator/lld/minutes/2010/12/02-lld-minutes.html#action01] [DONE]
ed: We tried to take Google docs and
reformulate it into the Wiki
... scenarios section and a use case section
... vocabularies section
... and a section at the end for miscellaneous
antoine: sections seem to align well with archives cluster
ed: feedback is welcome
<scribe> ACTION: Monica and JISC colleagues to introduce and discuss JISC work on next call [recorded in http://www.w3.org/2005/Incubator/lld/minutes/2010/11/25-lld-minutes.html#action02] [CONTINUES]
antoine: what is the CKAN volunteer status?
ww: when you create your account
profile, you should fill in your name to make your identity clear.
... contact ww to volunteer your help as a curator.
antoine: need more
volunteers
... this will be a useful way to structure the community.
<scribe> ACTION: Volunteers to send login information (openid credentials) to William Waite to curate LLD group on CKAN [recorded in http://www.w3.org/2010/11/25-lld-minutes.html#action04] [CONTINUES]
<ww> kcoyle++
karen: is there a tag to bring together bibliographic datasets?
<edsu> thanks karen, i was missing that too :-)
<jodi> maybe it's in ww's instructions?
<jodi> I'm not sure--but if we choose one, let's add i there!
<jodi> it
karen: Is there a way to connect "linked" and "bibliographic" datasets?
ww: We need a convention.
... The current intersection between bibliographic and linked is
LLD
... There is a tool that checks properties to see
if the resources are dereferencable. It would be better to mark them as such, though.
<emma> twitter tag is lldata
<Asaf> ww: is the tool world-accessible?
<ww> http://ckan.net/group/lld
<Asaf> thanks
<ww> http://packages.python.org/curate/
ww: The tool is still rough, but should get better.
<edsu> sorry i missed the wiki page for instructions, i see the ckan group page links to http://esw.w3.org/TaskForces/CommunityProjects/LinkingOpenData/DataSets/CKANmetainformation ?
<jodi> http://esw.w3.org/TaskForces/CommunityProjects/LinkedLibraryData/Datasets/CKANmetainformation
<jodi> http://esw.w3.org/TaskForces/CommunityProjects/LinkingOpenData/DataSets/CKANmetainformation
<jodi> someone needs to be tasked with coming up with a tag, I guess
ww: The suggestion is that CKAN should inherit tags from others where possible and develop more specific ones as needed.
RESOLUTION: need to tag dataset with LLD in CKAN and add it to the LLD group
contact ww to become a curator by creating an account (indicating your real name) and asking him to add you.
karen: how can we take existing sets and add them to our group?
ww: They can be added to the LLD page. Instructions to follow...
antoine: That's all for this week. Meeting ajourned.
<jodi> bye!
<edsu> cheers
<aconstan> bye
<monica> bye