The aim of the ALERT project is to develop methods and tools that improve FLOSS coordination by maintaining awareness of community activities through real-time, personalized, context-aware notification.

ALERT will create an active collaboration platform, i.e. a virtual actor would interact with other developers, process and recognize various kinds of interactions, suggest actions on the basis of these and remember and bring past interactions into the developers’ attention, thus enabling developers to work better together.

Frontpage

  • Efficient modelling of the more reactive coordination in FLOSS development that will improve the awareness of the group work and the responsiveness of individuals.
  • Efficient management of the awareness of team members that will enable interesting parties to be notified based on their interest/expertise as soon as something relevant happens without overloading them, interfering with and slowing down their work.
  • Efficient management of information relevant for FLOSS teams, including the semantic integration of information and its flow between all stakeholders that will support better understanding of the situations which a developer should react on.
  • Personalized and task-based access to information, by allowing developers to focus on activities to be performed to achieve a specific shared task and/or by including information about the presence and activities of other developers in the FLOSS.
  • Pilot, evaluate and impact the ALERT system in three important FLOSS communities (Mandriva, KDE and Morfeo) and disseminate the results in other FLOSS and relevant research communities.

In order to achieve the mentioned objectives, and to address the exposed problems, ALERT proposes a kind of interaction highway that will enable active, timely and personalized inter­action/coordination between all relevant artefacts and developers; in particular applying such an interaction highway in reducing the negative distance effects on collaboration in bug resolution.

The bellow figure shows the role of the interaction highway in managing implicit coordination in FLOSS development model.

The main research outcomes of the project will be:

  • Set of ontologies that model the conceptual dependencies between community, content and interactions.
  • Conceptual model of an (re)active collaboration platform.
  • Model for declarative specification of alerts.
  • Intelligent, pattern-based engine that can correlate, aggregate, and filter events in order to provide triggers for relevant notifications.
  • Module for integration of information from structured and unstructured sources.
  • Interaction highway set of services (e.g. search, recommendation, visualisation, etc.).
  • Module for capturing explicitly or implicitly the interest/expertise of the developer.
  • Module for capturing relevant information for the task at hand.

Besides, three use cases will be instantiated to address the three relevant aspects in FLOSS coordination: awareness, communication and task assignment.