This paper presents an assessment case study on the evolutionary capability of a large software system using a knowledge-based approach. The knowledge-based assessment is based on interviews with selected stakeholders of a software system. We have used this to assess the capability of the software architecture to evolve in one large Nokia software system. We have found that this approach proves to be effective in large organizations where development teams are distributed in different time-zones, with cultural differences and with limited person-to person communication. The lessons learned and the advantages of using this approach are presented.

Mobile terminals have evolved from basic portable telephones to complex and diverse devices that encompass dozens of other features, ranging from tri-dimensional games to office suites with data transmission capabilities. Variability is value: mobile terminal manufacturers must succeed in fulfilling the requirements of hundreds of mobile telecom operators worldwide, and at the same time increase the value of their brand by adopting a common user interface style while offering the features that the target end-user category desires. This makes for practically infinite variability and creates a business problem. The complexity of the variability problem increases due to issues such as the ability to ‘plug and play’ and ‘feature descension’ (the down-scaling of high-end features and their introduction into lower-end models). The main lesson we have learned from our experience in this field is that the application of relatively simple architectural patterns usually eases up management of the complexity at the architectural level. However, tackling the variability problem at the technical level is ineffective unless the organization is able to ensure the application of the solutions. We analyze the main challenges that lie behind the variability problem in mobile terminals, at both technical and organizational level, and illustrate some of the solutions we have implemented together with our product developers and system architects. Our experience calls for more applied research in the area of variability management, as well as for a number of enhancements to academic curricula. Copyright © 2005 John Wiley & Sons, Ltd.

Software architecture assessments are a means to detect architectural problems before the bulk of development work is done. They facilitate planning of improvement activities early in the lifecycle and allow limiting the changes on any existing software. This is particularly beneficial when the architecture has been planned to (or already does) support a whole product family, or a set of products that share common requirements, architecture, components or code. As the family requirements evolve and new products are added, the need to assess the evolvability of the existing architecture is vital. I illustrate two assessment case studies I have recently worked on in the mobile telephone software domain: the Symbian operating system platform and the network resource access control software system. The former assessment has been carried out as a task within the European project ESAPS, while the latter has been performed solely by Nokia. By means of simple experimental data, I show evidence of the usefulness of architectural assessment as rated by the participating stakeholders. Both assessments have led to the identification of previously unknown architectural defects, and to the consequent planning of improvement initiatives. In both cases, stakeholders noted that a number of side benefits, including improvement of communication and architectural documentation, were also of considerable importance. I illustrate the lessons we have learned, and outline suggestions for future research and experimentation.

We present the results of a research work targeted to understanding CASE tools usage in Nokia. By means of a survey questionnaire, we collected data aimed to identify what features are most useful and best implemented in current CASE tools according to senior developers and managers. With the aid of both descriptive and inferential statistical data analysis methods, we found out that the features that are rated most useful belong to the graphical editing, version management and document generation categories. The statistical methods we use allow us to extend the results to the whole population with a certain degree of confidence. The analysis of the data seems to give the indication that there is a general level of dissatisfaction on the quality of currently available CASE tools. Also, there is evidence that some of the most advanced features (reverse engineering, code generation) are not deemed as useful as others. Further research should focus on extending the survey to other types of industries, and attempt generalization of the results. This may constitute precious feedback for the software tools industry in order to develop products that correspond more to industry needs.

The Use Case formalism is an effective way of capturing both business process and functional system requirements in a very simple and easy-to-learn way. Use Cases may be modeled in a graphical way (e.g. using the UML notation), mainly serving as a table of content for Use Cases. System behavior can more effectively be specified by structured natural language (NL) sentences. The use of NL as a way to specify the behavior of a system is however a critical point, due to the inherent ambiguity originating from different interpretations of natural language descriptions. We discuss the use of methods, based on a linguistic approach, to analyze functional requirements expressed by means of textual (NL) Use Cases. The aim is to collect quality metrics and detect defects related to such inherent ambiguity. In a series of preliminary experiments, we applied a number of tools for quality evaluation of NL text (and, in particular, of NL requirements documents) to an industrial Use Cases document. The result of the analysis is a set of metrics that aim to measure the quality of the NL textual description of Use Cases. We also discuss the application of selected linguistic analysis techniques that are provided by some of the tools to semantic analysis of NL expressed Use Case.

Nokia Research Center (NRC) has been leading Esprit project ARES (Architectural Reasoning for Embedded Systems) for two years. One of the main purposes of ARES is to demonstrate the use of architectural modeling in controlling properties of industrial scale products and product families. Two examples are documented. Forward architecting was applied in modeling the new architecture for our mobile phone family showing how to control the dynamic properties of the software. Reverse architecting was used to tune the performance of our telephone switch software. Both cases show that special purpose models can be constructed even for very large systems and they do give valuable insight to the properties of the software.