The workshop is planned as a full-day event. It will include time slots for presentations and discussions. Authors of accepted submissions are asked to give a presentation of their research work. The presentations will be 15 minutes long for a short paper, and 20 minutes long for a long paper, with 15 minutes allocated for questions/discussion. Presentations should include research results, interesting research questions, and open issues to be discussed during the workshop.
Directly after each paper presentation,a pre-assigned discussant (assignments as indicated below) will provide a critical review of the paper. This task will be guided by a predetermined set of questions provided by the workshop organizers in. Discussion points of major interest can also be postponed to a dedicated forum which is planned at the end of the workshop.
The workshop program was organised as follows:
|09:00 - 09:10||Welcome & Introduction to ViSPLE 2008|
|09:10 - 10:00||Challenges for Graph Visualisation|
|10:00 - 10:30||On Controlled Visualisations in Software Product Line Engineering|
Florian Heidenreich, Ilie Savga and Christian Wende
|10:30 - 11:00||Break|
|11:00 - 11:30||Visualising Variability Relationships in Software Product Lines|
Ciarán Cawley, Steffen Thiel and Patrick Healy
|11:30 - 12:05||Visualizing Software Product Line Variabilities in Source Code|
Christian Kästner, Salvador Trujillo and Sven Apel
|12:05 - 12:35||Commonality and Variability Visualization using Multidimensional Scaling|
KwangChun Lee and Dan Hyung Lee
|12:30 - 14:00||Lunch|
|14:00 - 14:30||Flexible and User-Centered Visualization Support for Product Derivation|
|14:30 - 15:05||Three-Dimensional Feature Diagrams Visualization|
Pablo Trinidad, Antonio Ruiz-Cortés, David Benavides and Sergio Segura
|15:05 - 15:40||Visualization Support for Software Prouct Line Modeling|
Reinhard Stoiber, Tobias Reinhard and Martin Glinz
|15:40 - 16:10||Break|
|16:00 - 17:15||Discussion Forum|
|17:15 - 17:30||Wrap-Up & Summary|
Introduction to ViSPLE 2008
This workshop aims at elaborating on the idea of using visualisation techniques to achieve the economies of scale required to support variability management and product derivation in industrial product lines...
Challenges for Graph Visualisation
Visualization of graphs and networks is helpful in understanding and structuring complex relational information. The classical graph visualization methods (such as Sugiyama algorithms, force-directed algorithms, and topological approaches), developed about 25 years ago, are now commonly used in Software Engineering tools. However, these methods face significant challenges in three ways: (1) the scale of data is just too big for the classical methods, (2) the data has a time dimension, and
(3) the data is full of uncertainty. In this talk, we review these issues and describe some attempts to solve the problems that arise.
On Controlled Visualisations in Software Product Line Engineering
Authors: Florian Heidenreich, Ilie Savga and Christian Wende
Discussant: KwanChun Lee
Ongoing research in Software Product Line Engineering (SPLE) emphasises the derivation of a concrete product based on a given variant configuration as one of the most promising areas of the field. To allow for (automatic) derivation of products in SPLE, models that describe features and their variability in Software Product Lines (SPLs) -- for example feature models -- need to be connected with artefacts that are realising the features. It is crucial not only to support the developer in the complex task of defining such connections, but also to provide means to reason and analyse them -- for example visualisations. In this paper we present MappingViews, a novel visualisation technique that provides four different visualisations for connections between variability models and realisation models and show its realisation in our tool FeatureMapper.
Visualising Variability Relationships in Software Product Lines
Authors: Ciarán Cawley, Steffen Thiel and Patrick Healy
Discussant: Florian Heidenreich
Software Product Line Engineering is a development paradigm that focuses on the identification and
management of the commonalities and variability of a set of software products such that core assets can be developed and (re)used to derive individual product variants with a minimum of cost. In industrial product lines where it is possible to have thousands of variation points, the scale of variability can become extremely difficult to manage. In this position paper we elaborate on our ideas of focussing the representation and visualisation on the variability relationships that exist between different product line elements such as decisions, features and components and not on those elements that they relate. Further, we provide a conceptual three-dimensional visualisation technique to manage these relationships in the context of specific stakeholder tasks.
Visualizing Software Product Line Variabilities in Source Code
Authors: Christian Kästner, Salvador Trujillo and Sven Apel
Discussant: Reinhard Stoiber
Implementing software product lines is a challenging task. Depending on the implementation technique the code that realizes a feature is often scattered across multiple code units. This way it becomes difficult to trace features in source code which hinders maintenance and evolution. While previous effort on visualization technologies in software product lines has focused mainly on the feature model, we suggest tool support for feature traceability in the code base. With our tool CIDE, we propose an approach based on filters and views on source code in order to visualize and trace features in source code.
Commonality and Variability Visualization using Multidimensional Scaling
Authors: KwangChun Lee and Dan Hyung Lee
Discussant: Pablo Trinidad
As markets are fragmented into a plethora of submarkets, coping with various customers demands is
becoming time consuming and expensive. Recently software product line (SPL) approach has shown many benefits as opposed to single product development approach in terms of quality, time-to-market, and cost. Even though the opportunity for successful projects is at a project's start, early user involvement is still a rare phenomenon and software features visualization techniques to set the scope for a software product line through users' communications have not been explored. Since software product features are multidimensional and non numeric in nature, feature data has not been fully exploited as means of SPL scoping. In this paper, we propose software features visualization techniques to set the scope of a software product line based on multi-dimension scaling (MDS) of binary dissimilarities. A text editor software example is used to illustrate how product line scope can benefit from MDS.
Because of the size and complexity of product lines and the heterogeneity of the involved people, utilizing product lines in product derivation can be a tedious task. Different people have to understand different aspects of the provided variability. For example, sales people communicate variability to customers to elicit their requirements. Engineers typically resolve variability based on customers' requirements. These different people need different visualizations helping them best to understand and resolve the available variability. In this paper we present extendable tool support for product derivation that provides different visualizations for different users.
Visualizing and manipulating large feature diagrams is still an open issue for the SPL community. Few effort has been made on improving the techniques to get the most out of drawing space and current feature modeling tools either use file-system-like trees or 2D graphs that must be scrolled to locate features. The aim of this paper is presenting a new method to draw large feature models based on cone trees, a three-dimensional visualization technique to represent hierarchical information. In order to evaluate our proposal, we develop a prototype that generates standard 3D files so it can be easily integrated into existing tools. Finally, we present a roadmap for a future extension of our proposal with dynamic behaviour so large feature models handling might be improved.
Visualization Support for Software Prouct Line Modeling
Authors: Reinhard Stoiber, Tobias Reinhard and Martin Glinz
Discussant: Rick Rabiser
Product derivation from a product line model is a central activity in product line requirements engineering. A product line framework and/or approach should provide sophisticated visualization and product derivation support, otherwise large product lines become very hard to handle. In this paper we first present how we can benefit from existing single-system visualization techniques in product lines. Then we introduce new visualization aids and concepts that support the analysis of variability and the derivation of products in software product line engineering. We expect that these new visual support mechanisms will significantly ease the work of stakeholders and requirements engineers when negotiating a new product based on a software product line.