EU-funded project to make new software more trustworthy and adaptable
An EU-funded project addressing the challenges of designing and maintaining long-lived and trustworthy software systems has just got underway. The HATS ('Highly adaptable and trustworthy software using formal models') project has been funded with EUR 5.27 million from the Information and Communication Technologies Theme of the Seventh Framework Programme (FP7). The demands of software development are considerable. Given the costs of building large software programmes such as those used in e-commerce or e-government, such programmes must be long-lived and secure. And given their growing importance as part of the infrastructure of modern society, their trustworthiness (functional correctness or adherence to security policies, for example) is of utmost importance. In addition to being usable for a very long time, large programmes must accommodate a wide range of requirements and conditions, adapt to changing working and financial environments and be scalable to the needs of companies large and small. Yet the programmes that are in widest use today are based on millions of lines of code, which makes it impossible for anyone to single-handedly examine the entire programme's structure and anticipate all its possible interactions. Current approaches to reusing and maintaining various software components are unequal to the task of meeting the future needs of e-commerce, e-health and e-government applications. HATS researchers will spend the next three years contending with the challenges of ensuring security, adaptability and trustworthiness in the code on which large software programmes are based. According to the partners, led by Chalmers Technical University of Sweden, extending the current set of formal practices will not be enough. The consortium hopes to achieve a framework for developing methodologies and tools that will enable developers to automate the process of maintaining dynamically evolving software. HATS will also focus on security and trustworthiness by replacing what are normally informal processes with rigorous mathematical analyses of the meaning of the programming languages used ('formal semantics'). At the heart of the project is an Abstract Behavioural Specification modelling language, which can be used to describe software features and components (and where they occur) very precisely. This fills a serious gap in current software development practices; without such formal and in-depth descriptions, automated consistency checking and security enforcement would not be possible. 'The major challenge facing software construction in the next decades is high adaptability combined with trustworthiness,' according to the project partners. 'We propose to take an empirically successful, yet informal software development paradigm and put it on a formal basis. Specifically, we will turn software product family development into a rigorous approach.' Designing self-adapting software systems also falls within the scope of the project, which brings together 12 European partners with expertise in programming languages, distributed systems, software product lines, formal verification, and type theory.
Countries
Sweden