WP2 kick-off meeting

  • Date: January 8, 2008, from 10.00-13.00h
    • A sandwich lunch will be organized at the end of the meeting.
  • Location: Universiteit Antwerpen, Campus Middelheim , building G, room G.005
  • Contact person: Bart Du Bois

How to get there

Campus Middelheim is located in Middelheimlaan 1, 2020 Antwerp. The entrance of building G is located in Floraliƫnlaan. There is a parking right in front of the G building, which can be recognized as the most modern building on the campus. This building is right beside the Middelheim hospital.

Travelling directions and maps are provided at this page in Dutch. You can use Google maps to find your way to building G.

Participants

  • UA
    • Serge Demeyer
    • Bart Du Bois
    • Hans Schippers
    • Jan Vlegels
  • FUNDP
    • Vincent Englebert
    • Nicolas Genon
    • Andreas Classens, Fall-in for Raimundas Matulevicius
  • KUL
    • Stefan van Baelen
  • VUB
    • Dirk Deridder
    • Ragnhild van der Straeten
    • Mathieu Braem

Context

To kick-start Workpackage 2, this meeting first of all intends to have the partners get to know each other. Secondly, we need to achieve commitment about our research goals. It has been about one year since the proposal was written, and we need to iterate on our initial perspective. By indicating where we are today and where we want to be after the project, we will clarify how we are going to organize the work.

Outcome

  • First step towards a cohesive network.
  • Commitment of partners.
  • Overview of most related work from each of the partners
  • Iteration on original proposal (e.g., tentative list of deliverables)

Agenda

  • Welcome and introduction
    • Objectives
    • Partners introduction
    • MoVES way of working
  • Introduction to the topics
    1. Semantics of modelling languages
    2. Concern-specific modelling languages
    3. Assessing the effect of model refactoring on quality of models
  • Kick-starting the work
    • Clarification of contribution intent
      • Current related work
      • Ambitions within WP2
    • Discussion of deliverables
  • Conclusions, next steps, next meetings
  • Any other business
  • Closing

Preparations

In preparation of the kick-off meeting, I would like you to clarify your related work. Please send me by email:

  1. a top 3 of papers (published by your team) most related to WP2
  2. a brief description of your previous and current activities which are concretely related to the 8 objectives I have recovered from the proposal. Note: partner names are indicated in case the proposal cited work published by the partner
    • Topic 1: Semantics of modelling languages
      1. formalize activity diagrams (VUB?)
      2. extend workflow modeling languages to enable modelling complex behaviour
      3. integrate feature models in UML (FUNDP?)
    • Topic 2: Concern-specific modelling languages
      1. define an ADL for aspectual architecture
      2. formalize the semantics of Problem Fames (FUNDP?)
      3. extend workflow modelling languages with croscutting concerns (VUB?)
    • Topic 3: Assessing the effect of model refactoring on quality of models
      1. composition model for refactorings
      2. refactorings for separation of concerns (TUD)
  3. a brief description of your personal ambition with regard to your WP2 contributions, and corresponding success criteria.

Minutes

The slides of the kick-off meeting can be downloaded.

Welcome and introduction

After introducing the objectives of the meeting (basically getting to know one another, identifying shared interests, updating the research goals and achieving commitment), we held a round robin to let each participant introduce him/herself and his/her research interests/background.

MoVES way of working

We agreed upon organizing a meeting at least every six months. We prefer to append the meeting at another event, e.g., before/after a workshop.

Introduction to the topics

The first topic, Semantics of modelling languages, turned out to be no longer in the focus of WP2:

  • The formalization of activity diagrams did not find any active research support.
  • The formalization of a workflow modelling languages semantics is merged into topic 2 (see below).
  • The integration of feature models in UML to model variability is handled in WP4.

Accordingly, we decided to drop topic 1.

The second topic, Concern-specific modelling languages, is also subject to changes. The initial goal to formalize problem frames semantics is dropped, since this is handled in workpackages 5 and 7.

Finally, topic 3, Assessing the effect of model refactoring on the quality of models, is reduced. There is no longer an interest in refactoring composition. Moreover, it is unclear whether refactoring to separation of concerns is still an active research topic in the project (@@TODO: ask TuD@@). However, Serge Demeyer proposed to introduce a new research goal, entitled mining software repositories.

Kick-starting the work

In this part of the meeting, we discussed where we are today, and where we want to be after the project.

Current related work

An overview of 10 of the most relevant papers/reseach topics was presented. Eight papers were submitted earlier, in preparation of the meeting. These papers are enlisted in the private section, and are available for download there. Additionally, two presentations were given on the research tools Padus and Cobro. These presentations are also available in the private section.

Quite some interest was raised for the MetaCASE tool. It seems interesting to apply this tool for modeling workflow.

Another potential interaction was identified based on the presentation of the AOP virtual machine (UA): is it feasible to implement a workflow model with an AOP VM? Stefan Van Baelen showed his interest in verifying whether AO models can be compose using the AOP virtual machine.

During the Cobro (VUB) presentation, Dirk de Ridder mentioned that he might provide some input to the PRECISE group with regard to implementation techniques for MetaCASE tools.

Lastly, after the introduction of UniTI (KUL) and MotMot (UA), we asked ourselves how traceability management differs between these two tools.

Accordingly, the following tool interests provides an initial agenda for a next meeting, which would consist of a series of tools demonstrations:

  • MetaDone (FUNDP)
    • Visualize models from Padus (of interest for VUB)
    • A PhD student at the KUL is working on architectural refinement using MetaDone
  • AOP VM (UA)
    • Implement workflow model (of interest for VUB)
    • Composing AO models (of interest for KUL)
  • Cobro (VUB)
    • Implementation technique for a MetaCASE tool (of interest for FUNDP)
  • UniTI + MotMot (KUL + UA)
    • How to manage traceability (of interest for KUL, UA and FUNDP)

Discussion of deliverables

We worked out a list of (joined) publications as the main deliverables for WP2:

  • Use MetaCASE to support modelling languages – e.g., case studies using modelling languages from the other partners (FUNDP).
  • Concern-specific modelling languages
    • Define an ADL for aspectual architecture (KUL)
    • Extend workflow modelling languages with crosscutting concerns (VUB)
  • Assessing the effect of model refactorings on the quality of models
    • Mutual interst in mining repositories between TuD and UA (@@TODO:to be confirmed this week@@)

Collaboration initiatives

We invited partners to share toy examples, case studies, tools and general material. E.g., the billing example (used by VUB for the introduction of aspects in workflow modelling languages) might be of interest for other partners as well.

In next meetings, we might share related courses, e..g, the ATOM3 experiment by the UA.

General tutorials, e.g., on Petri-nets, AOP, are welcomed in meetings, as well as invited speakers from inside and outside the workpackage.

Research visits are confirmed as a good way to organize concrete collaborations. FUNDP indicated that they have the funds to host a master student for 5 months at a foreign or Belgian university. However, the proposal needs to be submitted by January 20th. @@TODO: send email notification@@

Finally, the UA might collaborate with TuD with regard to mining repositories.

Conclusion, next meeting

We will organize the next meeting preferably before the yearly report (@@TODO: lookup date@@). We proposed to schedule the meeting once again at the UA, in either the last week of February or the first two weeks of March. The preference goes out to the last week of February, since one of the developers of a tool from the KUL will be unavailable in March.

For next meetings, we welcome invited speakers, and we will send out a Call for workshops or seminars. The idea behind this is that we show that we are a research network that actively disseminates its research.

Action points

The following action points were noted during the meeting:

  1. Schedule next meeting (UA)
  2. Create draft revision of new WP2 research outline (UA)
  3. Poll TuD for research interest in (i) refactoring to separation of concerns; and (ii) mining repositories (UA)
  4. Send email regarding student exchange (FUNDP)
  5. Check date of yearly report (VUB)
  6. Prepare presentations (ALL)
meetings/kickoff.txt · Last modified: 2008/01/28 09:15 by ven