Monday, April 15, 2013

The AMOR Manifesto

Introduction

Ontologies are becoming increasingly more important in many different domains. However, there are still plenty of barriers to the reuse of ontologies, being the main one the lack of open availability of existing ontologies.

Agreeing on a common set of ontological commitments is costly and all this effort is wasted when others are forced to develop a similar ontology from scratch or to re-engineer an ontology from other available artefacts (documentation, figures, etc.).

The goal for the AMOR Manifesto is to articulate the vision of an ecosystem of Accessible, Machine-readable, Open, and Reusable ontologies that facilitates the reuse of these ontologies or parts of them. To this end, a set of principles have been distilled, based on the 5 star scheme defined for Linked Open Data.

Principles

The ontology is available on the web (whatever format) but with an open licence
★★ All the above, plus: available as machine-readable structured data (e.g., CycL instead of image scan of a table)
★★★ All the above, plus: non-proprietary format (e.g., OBO instead of CycL)
★★★★ All the above, plus: use open standards from the W3C (RDF Schema and OWL)
★★★★★ All the above, plus: reuse other people’s ontologies in your ontology

These principles define a set of minimal restrictions to ensure that ontologies are accessible on the web using open standards. It is not in the scope of this Manifesto to impose further quality restrictions on ontologies (e.g., regarding the usability of Linked Data vocabularies).

Invitation

Everyone is invited to support the principles of the AMOR Manifesto and to find new ways to make this vision a reality.

If you support this manifesto and the principles included in it, please leave a comment below.

7 comments:

  1. I like how this manifesto and http://blog.hubjects.com/2012/02/is-your-linked-data-vocabulary-5-star_9588.html combine together. It would have been nice to include some examples for each star rating. In any case, nice work!

    ReplyDelete
  2. Holao Raúl

    AMOR = LOV I guess :) At http://lov.okfn.org we try indeed to apply similar principles in vocabulary selection, trying to add only ***** vocabularies ... and if they are not, push the vocabulary editors to improve them.
    In fact we do not look very closely at the first star, the more so that many vocabularies don't mention any licence at all.

    ReplyDelete
    Replies
    1. Yes, ontology licensing has been the "great unknown" up to now.
      Let's try to change it! :)

      Delete
  3. If you are interested in such issues, you should follow this event :
    http://dcevents.dublincore.org/IntConf/index/pages/view/vocPres

    ReplyDelete
  4. Great idea, I think this kind of star systems fit perfectly with many others that need to be implemented in science and society :)

    ReplyDelete