How it works

Analysis of feature modularity instead of just code modularity

Code modularity is one of the keys to keeping a software maintainable and extendable in the long-term. Low modularisation is to be considered as technical debt. Modularisation concerns both code and architecture and is a concept to avoid too strong couplings within the code base, as these lead to increasing problems in the maintenance of a software. The common software analysis tools consider the couplings between individual programme artefacts in the code, but leave out coupling issues in the software architecture. However, it is the technical debts in the software architecture that would most meaningfully describe the fundamental technological weaknesses of a software.

Cape of Good Code analyses the modularity of features with the DETANGLE Analysis Suite in order to also consider the role of the software architecture with respect to technical debt. The analysis of the couplings between features in the code provides significantly more meaningful results than just examining the couplings between artefacts of a programming language. This enables more targeted insights into the technological state of the software, e.g. with regard to the effort required to extend it with new features.  

By considering the importance of the software architecture for the criticality of technical debt, a meaningful prioritisation of restructuring measures can also be derived from the data of a DETANGLE analysis.

Our software analysis tool DETANGLE® analyses the relationships between features and code as well as between features themselves over the entire lifetime of the software. It therefore captures technical debt based on the modularity of features and not only on the modularity of code. With this concept, all the necessary information becomes available to initiate feature-centric restructuring measures directly at the right places in the code. We recommend monitoring the effectiveness of the measures with renewed analyses on a regular basis.

Feature modularity and the concepts behind are absolutely innovative and groundbreaking in the field of software quality analysis. The underlying intelligent mathematical models are the results of our long-term practical experience and research in this field.

Time-based analysis

All types of software quality analysis strive to identify or predict maintainability and extensibility constraints. Ultimately, the aim is always to measure how easily a system can be extended with new features. Existing approaches to code analysis attempt to make statements about the changeability of a system by simply analysing the current state of the code without measuring what really matters: the actual changes to a system over the course of its life cycle.

Our research and projects show that taking into account the progression of the software development process is prerequisite for gaining usable insights into the technological status quo of the software architecture and for predicting critical system errors that would limit its extensibility.

DETANGLE® analyses the entire commit history of a project. It tracks code changes and relates them to each other, rather than just comparing the results of static code analysis of different snapshots. We have developed solutions with DETANGLE® to consider parallel development branches in distributed repositories or between multiple repositories to analyse the software architecture for extensibility with new features.

DETANGLE® can predict the technological future of the code from its history.

Your development data and IP. How secure are these with our analysis?

To start with, the appropriate contractual regulations regarding the confidentiality of IP, data and other relevant information and documents are agreed between the client and Cape of Good Code. These regulations formally and legally ensure that the information and data from the customers software used in the analysis are no longer accessible to the consultant after completion of the project.

From a technical point of view, the DETANGLE® analysis software is installed, configured and executed on the client's computer. All analysis results remain on the client's computer. Remote access to this computer can be encrypted and secured via a VPN connection. The software analysis can be carried out on-premise or remotely without the need that the analysed code has to be stored on an external computer.