The Communicat - PowerPoint PPT Presentation

1 / 35
About This Presentation
Title:

The Communicat

Description:

An inventory of resources accumulated over time largely dictated by: ... First try: Unalog. Social bookmarking service created by Dan Chudnov. http://unalog.com ... – PowerPoint PPT presentation

Number of Views:50
Avg rating:3.0/5.0
Slides: 36
Provided by: library3
Category:

less

Transcript and Presenter's Notes

Title: The Communicat


1
The Communicat
  • contribute to the collective collection, comrade

2
What is a catalog?
  • An inventory of resources accumulated over time
    largely dictated by
  • Collection development policies
  • Budget
  • Preferences and biases of subject specialists (or
    lack thereof)
  • No real relationship to the outside world
  • No real way for the community to be directly
    involved

3
More than just inventory
  • Before drinking the Kool-Aid one must accept that
    public interface doesnt have to be 11 with
    inventory
  • Doesnt have to be included in ILS
  • Doesnt have to be included by default
  • Theoretically doesnt have to be included
    publicly at all

4
Now that Ive backed away from that
5
Lets rewind one year
  • Desire to capture individuals interests and
    associations between objects
  • Scope search context through lens created by
    interest/community
  • Snapshots of scholarly content aggregated into
    clusters or groups (by project, class, etc.)

6
First try Unalog
  • Social bookmarking service created by Dan Chudnov
  • http//unalog.com/
  • Built-in support for groups
  • Kicked around extending Unalog to support MODS
  • Installed local development instance gTunalog
  • A little more simplistic than we could use

7
So on with the vision thing
Catalog records
Webpages
Citations
Communicat
Interlibrary Loans
8
citation
citation
User
Reserve items
citation
annotation
User
Group Superconductor Research Article
User
Group Psychology 231
User
citation
Communicat
annotation
citation
ILL
bookmark
bookmark
bookmark
9
Emory Universitys Communicat
Joint research group
Georgia Techs Communicat
10
So what we thought
  • Dont worry about the end result right away
  • Allow users to gather resources and aggregate
    them however they want
  • It doesnt matter where things come from
  • We can use things we can identify (OpenURLs,
    MARC, DC, etc.)
  • Treat everything else like a bookmark

11
How to approach?
  • Submission needed to be in workflow users
  • We would also need to know a little about what
    the user was looking at/where they were
  • Link resolver would tell us a lot of that

12
Birth of the Ümlaut
  • Didnt think SFX would meet needs out of the
    box
  • Needed to come up with different handlers based
    on what something is
  • Grab data from a variety of sources
  • Give user a variety of places to store
    citations del.icio.us, Connotea, CiteULike,
    EndNote, etc.

13
Building a better link resolver
  • Broke ground mid-March last year with proof of
    concept
  • Ruby on Rails application
  • By April we knew we had something to work with
  • Finished mid-summer, launched before Fall
    semester (with no personalization/Communicat
    functionality)

14
What it does
  • Queries Crossref/Pubmed and grabs any metadata it
    can find
  • Queries SFX, Voyager catalog and consortial
    catalog (via xisbn, if necessary)
  • Queries Amazon, Google and Yahoo
  • Presents results to user

15
Then behind their back
  • Queries Connotea, CiteULike, Yahoos MyWeb and
    Unalog
  • Harvests metadata from any known OAI-PMH server
  • Makes associations between their citation and
    other things (via catalog, Amazon, other
    resolved citations, Citeseer references, etc.)

16
(No Transcript)
17
(No Transcript)
18
(No Transcript)
19
(No Transcript)
20
Todos
  • Personalization!
  • Was held up for a few reasons
  • Bugs needed ironing out at first
  • Fit into a larger framework
  • Seemed a bit overwhelming all at once
  • Redesign of backend in process, should be
    finished by end of month

21
Stage 2 Fancy Pants
  • AJAX Layer over WebVoyage
  • Aggregation of several components (improved
    search result display, faceting, spellcheck,
    etc.)
  • Replacement of Voyagers record save feature
  • Saved things from OPAC and Ümlaut should go to
    the same place
  • Track how people use the OPAC what do they
    click on? What resources do they look at?

22
Status
  • Mostly built (although new components will
    continue to be developed)
  • Librarians looking at current functionality
  • Launch Spring Break?

23
Next step Emperors clothes
  • Native interface for Fancy Pants components
  • Uses Solr/Lucene/Ferret for searching, calls back
    to Voyager for holdings and patron empowerment
  • When logged in, scopes relevance to user context
  • Expect development to really take off Fall 07

24
Meanwhile, in the classroom
  • LibraryCube web services bundle of library
    things to enable communication to non-library
    applications
  • Primary use case would be courseware
    specifically Sakai
  • Two way communication, LibraryCube aggregates
    library services for classroom purposes and
    classroom results for library purposes
  • Builds Communicat groups according to needs, can
    return results based on group dynamics

25
The Research Trail
  • Track users paths through information landscape
    and annotations surrounding them to the eventual
    scholarly output
  • Can be used for classroom, professional writing,
    graduate thesis/dissertation, etc.
  • A sort of überbibliography

26
Powered by GaTher
  • The engine that saves all of these items and
    creates associations between them
  • Breaking ground now for Fancy Pants
  • DCMI/RDF based model for storing items can point
    back to original (also would store original
    format MARC, OpenURL, etc.)
  • Initial interface would look very social
    bookmarker-y

27
Many faces of GaTher
  • Citation manager
  • Reserves system
  • Subject/Research Guides
  • Wishlist
  • Recommender service
  • Personal library
  • Group library
  • Project space
  • Student portfolio

28
Tying it all together
  • Lots of seemingly disparate services that have
    nothing to do with each other
  • To the whiteboard!

29
(No Transcript)
30
(No Transcript)
31
(No Transcript)
32
Its all about how you want to look at the blob
in the middle
  • The data behaves differently based on the context
    in which it is viewed
  • But its all basically the same data
  • Users dont really care they just want it to
    work
  • Focus on the ingester perspective first

33
Considerations
  • Another black box does no good
  • This doesnt have to supply the interface,
    necessarily, just data
  • In fact, its ok if all it does is aggregate
    citation managers and social bookmarks
  • Cant force users away from what they want to use
  • LMS integration is key, however captive audience

34
Redefining catalog
  • An aggregation of resources that is useful to a
    community
  • Includes resources and relationships that are
    tailored to the individual
  • Used things are more relevant/profile influences
    results
  • Inventory still just as important, but serves a
    different purpose

35
Thanks!
  • Any questions?
  • ross.singer_at_library.gatech.edu
Write a Comment
User Comments (0)
About PowerShow.com