SIMILE: objectives, status & demo Stefano Mazzocchi - MIT…

SIMILE: objectives, status & demo
       Stefano Mazzocchi - MIT Libraries




          "Practices of Knowledge Sharing" Workshop
              April 20-22 2004, Heraklion, Greece
Objectives
                SIMILE Goals

·   Make semantic interoperability of metadata a reality
    for digital libraries by:

    ·   providing reusable software for browsing,
        searching and mapping heterogenous metadata

    ·   using semantic web technologies

    ·   identifying issues, gaps and best practices
             SIMILE Vision

·   tools should help humans focus on their abilities,
    amplifying, not replacing them!

·   metadata quality depends on its heterogeneity

·   serendipitous discovery is a value that should not
    get lost

·   empower recombinant metadata
       SIMILE Participants


·   MIT Libraries (MacKenzie Smith)

·   MIT CSAIL (David Karger)

·   HP Labs (Mick Bass)

·   W3C (Eric Miller)
Status
                   Longwell


·   faceted metadata browser

    ·   aimed at end users

    ·   goal is to show max functionality with min
        complexity (maximize usability)
                     Knowle


·   RDF browser

    ·   aimed at semantic web specialists

    ·   goal is to enable cognitive estimations of
        complex models
                  Datasets


·   ARTStor

·   MIT OpenCourseWare

·   Wikipedia

·   CIA World Fact Book (in progress)
                  Schemata

·   Dublin Core

·   VRA

·   LOM

·   SKOS

·   SIMILE's own glue ones

·   LoC TGM (in progress)
           Achieved Results

·   Usable implementation of both Longweel and
    Knowle as web applications

·   Passed the 0.5 Megatriples wall

·   Successful use of XSLT2 as XML->RDF bridge

·   Use of the Levenshtein distance on literals to
    evaluate potential mappings between datasets
Demo
Open Questions
                 Scalability


·   How more complex can the model grow before
    saturating our computational capacity?

·   How can we design a distributed architecture and
    still be fast enough to be useable?
              Connectivity



·   How can we increase the connectivity when
    merging models with reasonable costs and
    without compromising perceived metadata
    quality?
               Provenance



·   How should provenance influence the reasoning
    on aggregated models?
                 Evolution


·   How can we deal with the evolution of models
    and their impact on previous inferenced
    interpretations?

·   Can time be another provenance or we need a
    different dimension?
             Disagreement


·   How well can the semantic web model cope with
    disagreement?

·   How do we distinguish disagreement from
    mistakes?
Thanks!
Q&A