From Athenawiki

Jump to: navigation, search

Contents

Purpose

Here is the description of the logical processes concerning terminology within the project Athena. This document is a draft and its content shall both help write the WP4 deliverables and drive the experiment we are planning. So: For discussion.

Introduction

Here are a few remarks necessary to keep in mind by reading this document.

Museums

Among all the possible content providers we focus on the European museums:

  • Which have descriptions of the objects composing their digital collections,
  • Which have used a terminology to express these descriptions,
  • Which have made available or intend to make available their repository for an harvesting by Athena, hence by Europeana.

Use cases

For such an institution, we have listed different scenarios that we can distribute into 2 major categories. The first main category takes into account 3 cases where the processes are fully achieved through Athena and Europeana without any update of the Athena Thesaurus. The second main category deals with 3 other cases where an evolution (modification/update) of the Athena thesaurus has to be achieved.

No update cases:

  • The institution has used a standard fitting with Europeana for describing its collections
  • The institution has used the Athena Thesaurus without modification for describing its collections
  • The institution has used another terminology which is already registered into the Athena Platform and mapped with the Athena Thesaurus, and does not aim to update anything

Necessary update cases:

  • The institution has used another terminology which is already registered into the Athena Platform and mapped with the Athena Thesaurus, and just aims to update its mapping
  • The institution has used another terminology which is already registered into the Athena Platform but which is not mapped with the Athena Thesaurus yet, and aims to do this mapping
  • The institution has used another terminology which is not already registered into the Athena Platform, and aims to register it and map it with the Athena Thesaurus

Athena Thesaurus

A definition is proposed in the wiki part Athena Thesaurus.

Principles

Updating

In a nutshell, hereafter we have represented the processes of these use cases according to the principle of updating. In this way the question we tried to answer step-by-step was: Is there someone aiming to create or modify something? Then we identified the possible “who” of each action among: the institution, the Athena platform, the Europeana platform. This must absolutely be discussed and validated, especially by the WP7 leader. Any updating of the Athena Thesaurus is motivated by a change of the mapping between its elements. In certain case, this change is due to a new terminology that some institution has mapped with the Athena Thesaurus. And the mapping of a new terminology requires the validation of its form: the terminology must be well-SKOSified.


Processes

No update cases

Use case 1

       The institution has used a standard fitting with Europeana for describing its collections
This is the simplest case in the sense that it does not require any specific action from the Athena terminology scope. Indeed the institution has used a standard already compliant with Europeana requirements to describe the objects of its collections.


Use case 2

        The institution has used the Athena Thesaurus without modification for describing its collections	
This case is also very simple, like a particular case of the previous one. Here the institution has used the Athena Thesaurus to describe the objects of its collections. Now, as far as the Athena Thesaurus is used without modification in-house, it is compliant with Europeana requirements. This is one guarantee that Athena can provide.


Use case 3

         The institution has used another terminology which is already registered 
         into the Athena Platform and mapped with the Athena Thesaurus, and does not aim to update anything

Here is the last simple case we have identified. The institution has used another terminology to describe the objects of its collections rather than the compliant standards nor the Athena thesaurus. But it has already registered it into the Athena Platform and mapped it with the Athena Thesaurus. So all the descriptions expressed with this terminology are exploitable by Europeana for access and retrieval. This is due to the mapping with the Athena Thesaurus. Since no update is foreseen, harvesting is in order as before.


Necessary update cases

Use case 4

         The institution has used another terminology which is already registered
         into the Athena Platform and mapped with the Athena Thesaurus, and just aims to update its mapping

Here is the first case of updating of the Athena Thesaurus. The institution has used another terminology to describe the objects of its collections rather than the compliant standards or the Athena thesaurus. It has already registered it into the Athena Platform and mapped it with the Athena Thesaurus.

But the institution needs to refine the existing mapping. So the Athena Platform must check if the new mapping is correct. If yes, the Athena Platform submits the update of the Athena Thesaurus to Europeana, and harvesting keeps possible as before. If no, the institution must refine again and again its mapping until the platform validates the result. In case of cancellation, the last valid mapping version is still applied.

Use case 5

        The institution has used another terminology which is already registered 
        into the Athena Platform but which is not mapped with the Athena Thesaurus yet, 
        and aims to do this mapping

Here is the case of updating of the Athena Thesaurus in which a first mapping has to be made between with the institution terminology. The institution has used another terminology to describe the objects of its collections rather than the compliant standards or the Athena thesaurus. It has already registered it into the Athena Platform, however it did not map it with the Athena Thesaurus yet.

To do so, the institution is invited to look at the existing terms of the Athena Thesaurus to find equivalent ones to its own terminology terms. Thanks to a domain organisation of the Athena Thesaurus and a graphical display of all its controlled vocabularies, the research would be more effective. If it finds relevant terms, the institution maps its terminology terms with those, then the Athena platform controls if the mapping is correct like it was a simple updating. If the institution does not find relevant terms, it can propose its own ones to enrich the Athena Thesaurus. Once again a validation process (with moderation) is necessary. The following of the process is like within the previous case.

Use case 6

       The institution has used another terminology which is not already registered
       into the Athena Platform, and aims to register it and map it with the Athena Thesaurus

Here is the final case of updating of the Athena Thesaurus in which a complete process of registration and mapping is necessary. The institution has used another terminology to describe the objects of its collections rather than the compliant standards or the Athena thesaurus. It has not already registered it into the Athena Platform.


The registration is mandatory before any mapping with the Athena Thesaurus. This is a 2-step process. First the Athena Platform checks if the terminology is semantically and syntactically valid (it means: if the file can be interpreted). Then it checks if the terminology is well-SKOSified. If these two requirements are not satisfied, the Web service of the Athena platform does not allow the mapping. If they are satisfied, the institution can map its terminology with the Athena Thesaurus as presented in the previous case.


Issues

Thanks to this process representation, we have listed for the time being 4 issues to discuss with Athena partners:

  • Workflow: We are planning to deliver finally a workflow specification of collaborative production of the Athena Thesaurus; the moderation process appears as the tricky point of such a workflow;
  • Versioning: We keep in mind that such a service must ensure at every moment that all previous versions of the Athena Thesaurus are still working for harvesting. Each update (modification/deletion/addition) should be detailed and archived.
  • Platform(s): We wonder how to technically support SKOSification and mapping tasks: is it in the perimeter of the Athena platform? Shall we expect that Europeana will provide similar Web services (and hardware) we could duplicate? (→ To discuss especially with WP7 and EuropeanaConnect or v1.0)
  • IPR: We consider to let the Athena Thesaurus free of rights for use and (controlled) modification; a Creative Commons “By:” license might be useful → To discuss especially with WP6
  • Sustainability: What Athena and/or Europeana can ensure as a service after the project? What if we propose at the end of the project an Athena Thesaurus Web service for the online use of the Athena Thesaurus for description of collections?
This page was last modified on 20 March 2010, at 10:45.This page has been accessed 18,096 times.