Policy implementation and change management - the PERICLES way
In this blog post we present a practical example of how the PERICLES model-driven approach can help to implement, automate and manage change in a policy.
'Automatic policy application and change management’ was also the focus of one of the 'PERICLES in Practice' sessions at the final project conference ‘Acting on Change: New Approaches and Future Practices in LTDP’. In this presentation (available on Slideshare) we used the EUMETSAT example to demonstrate the change of a policy in a real life scenario applying the Policy Editor  and the Entity Registry Model Repository (ERMR).
Although there is a vivid discussion on the exact application of the term “policy” in the digital records management, archiving and preservation user communities, we take a high-level definition of policy  for the purpose of explaining how PERICLES proposes to implement change and manage the potential impact.
Policies can be used to specify the objectives and constraints of an organisation, and are commonly used to describe a preservation system’s driving principles.
We describe here a demonstrator showing two main concepts:
- How policies can be modelled, implemented and automatically applied with PERICLES tools.
- A general architecture for change management supporting policy change.
Within the PERICLES model-driven approach, a policy is considered an entity in the Digital Ecosystem Model (DEM), an ontological model representing a broad set of entities (digital objects, users, services, policies, processes). Through the use of the LRM and semantic technologies, we are tackling the problem that both policies and ecosystems are subject to continuous change.
We took an EUMETSAT data policy , expressing the embargo period for the data coming from satellites. The EUMETSAT, SEVIRI Images are data from the Meteosat Second Generation (MSG) geostationary satellite with Spinning Enhanced Visible and InfraRed Imager (SEVIRI).
For this scenario we built a model and a rule that can automatically apply the policy, and manage change in the embargo period being changed from 24 to 12 hours. This is effected thanks to a rule that automatically publishes the images based on a mutable parameter value.
We started by describing the scenario and policy as a DEM as shown in Fig. 1. Digital Ecosystem models can be created and edited using the PERICLES EcoBuilder tool.
Fig. 1. The Digital Ecosystem Model.
Once the model (DEM) has been created, containing all relevant entities, including policies, that contain their rule implementation and digital objects, it is sent to the PERICLES ERMR, a repository for both models and data. The DEM model is at a later stage opened in the PERICLES Policy Editor, where a user will might want to modify the release time from 24 to 12 hours. This change will in fact produce a second version of the model (DEM2), that will differ only in the parameter value.
Fig. 2: The PERICLES Policy editor.
A procedure in the policy editor will send the new DEM to the PERICLES LRM-S: this service will compute a semantic Delta (LRM-DELTA) that will be propagated to the ERMR through the PERICLES PERSIsT API in the form of a semantic query (SPARQL update query), updating in fact the copy of the model stored in the repository.
Fig. 3: The complete workflow.
The overall workflow is described in Fig. 3. Finally, the model is sent to the PERICLES SPIN rule engine. This service extracts from the models the SPIN rules, and executes them. This will in fact enact the policy implementation, in a way that can automatically manage change in the ecosystem model. In this specific example, the rule will release to a public repository all the data that is older than the specified embargo period (see Fig. 4).
Fig. 4: The public web portal - built on the ERMR repository.
The demo was developed by Fabio Corubolo (ULIV) with contributions from many of the project partners:
Johannes Biermann, Claudio Leone, Anna Eggers (University of Goettingen)
Marina Riga, Stratos Kontopoulos (CERTH/ITI)
David De Weerdt (Spaceapps)
Jean-Yves Vion-Dury (XRCE)
Jérôme Fuselier (University of Liverpool)
Simon Waddington (King's College London)
 More information on the Policy Editor coming soon.
 “A policy is a deliberate system of principles to guide decisions and achieve rational outcomes. A policy is a statement of intent, and is implemented as a procedure or protocol.” see: https://en.wikipedia.org/wiki/Policy
 The complete policy states that “Meteosat Data and Derived Products older than 24 hours are distributed on request from the EUMETSAT Data Archive in digital and graphical form via the associated operational service in formats which represent both full and partial spatial coverage as well as both full and partial spatial resolution.”, source: http://www.eumetsat.int/website/home/AboutUs/LegalInformation/DataPolicy/index.html