Difference between revisions of "Code4lib/elag2010"

Latest revision as of 15:26, 5 October 2009

Planning for the code4lib 1-day preconference at ELAG2010 in Helsinki (Finland)

You can get more information about what ELAG looks like at the ELAG2009 website at http://indico.ulib.sk/elag2009 or the Elag2008 website which has more footage.

At this stage (2009-06-04) things are pretty much wide open about this : feel free to step in.

Initial suggested topic for the preconference : lucene/solr
- "A backend for your new catalogue in 15 minutes. Bring your data!"
- for example, various programming interfaces to Solr/Lucene could be explored: Java, PHP, Perl, etc.
- all mixed up: one integrated search interface to all your content and data (whatever it is)

Text mining -- the process of extracting information out of unstructured text -- might be an interesting topic. More specifically, text mining includes the processes of relevancy ranking search results, automatically classifying documents, automatically categorizing documents (not the same as classification), entity (name, place, organization, date) extraction, keywords and key phrase extraction, and summarization. With the increasing availability of full text books and journal articles, text mining provides a way to make sense of large corpuses of content in ways traditional cataloging are unequipped to deal with.

Linked data seems to be one of many people's favorite topics these days. Discussions and demonstrations on how such things can be implemented might a good topic for a pre-conference.

Incorporating ebooks into library "catalogs". Many of us are licensing ebooks in the same way we are licensing journal article content. How can these things be integrated into our "discovery" systems.

Compare and contrast "discovery" systems. What are the various technical characteristics of things like VUFind, Primo, XC, Blacklight, Scriblio, etc. How are they the same and different? How are they each empowering and/or limiting?

Pick a cool OSS distribution: Umlaut, VUFind, LibraryFind, MyLibrary, etc. Then do some hands-on with the selected tool. Call it a user-group meeting.

other idea (but maybe less "hands on"?): Drilling holes into that data silo. How to connect cool tools (eg NGCs/DIs) to legacy library systems? What's available, what's needed and what's practical?

A real geeky coding session for library hackers or a more guided but "hands on" approach? Please no "workshop" in the "28 minutes talks/2 minutes questions" style.

everyone is expected to come with his/her own laptop
we could be hosted here or there by the University of Helsinki.
GOOD wifi required : should we ask for a wired network too?
server : we don't think there's a need for a server, as we may as easily set something up on EC2, a home server, etc. But I (nicomo) could be wrong here : pls give your point of view if need be.
number of people : we're thinking about a cap at 50. Which seems a lot for a hands-on thing, maybe? That's the working assumption at least for now.

@@ Line 7: / Line 7: @@
 === Ideas for topics ===
-* Initial suggested topic for the preconference : lucene/solr ("A backend for your new catalogue in 15 minutes. Bring your data!"). Yes, for example, various programming interfaces to Solr/Lucene could be explored: Java, PHP, Perl, etc.
+* Initial suggested topic for the preconference : lucene/solr
+** "A backend for your new catalogue in 15 minutes. Bring your data!"
+** for example, various programming interfaces to Solr/Lucene could be explored: Java, PHP, Perl, etc.
+** all mixed up: one integrated search interface to all your content and data (whatever it is)
+* put your library in the cloud using Amazon EC2 et al.
 * Text mining -- the process of extracting information out of unstructured  text -- might be an interesting topic. More specifically, text mining includes the processes of relevancy ranking search results, automatically classifying documents, automatically categorizing documents (not the same as classification), entity (name, place, organization, date) extraction, keywords and key phrase extraction, and summarization. With the increasing availability of full text books and journal articles, text mining provides a way to make sense of large corpuses of content in ways traditional cataloging are unequipped to deal with.