S-CASE: Scaffolding Scalable Software Services
The main goal of S-CASE is to provide a cloud-based realm of services and tools for software developers to enable rapid software prototyping based on user requirements and system models, provided in multimodal formats. It will allow the extraction of system specifications and low-level architecture, as well as the discovery and synthesis of composite workflows of software artefacts from distributed open source and proprietary resources. In this context, based on the current work programme and probing on the 2016-17 work programme, the S-CASE partners are focusing their discussions against five main axes:
- Boosting public sector productivity and innovation through cloud computing services
- Collective Awareness Platforms for Sustainability and Social Innovation
- Big data
- Internet of Things and Platforms for Connected Smart Objects - User-driven web
The most important result of the project is the methodology defined and the engineering framework developed to (semi)-automatically generate RESTful web services out of multimodal requirements. The requirements may come in the form of textual requirement in semi-structured documents, UML diagrams and storyboards. The developed methodology employs natural language processing (NLP) and automated software engineering (ASE) primitives in order to transform the provided information into a Computation Independent Model (CIM), which is further elaborated to a Platform independent Model (PIM), a Platform Specific Model (PSM), and finally, software code. Up to this point the ASE and NLP methodology have been defined, while two developer tools have also been released: a tool for importing, annotating and incorporating 3rd party web services into S-CASE services, and a tool for collecting and (semi-)automatically annotating textual requirements from software projects. Additionally, a search engine for software assets has been presented.
The project is still at its first year, thus a discussion on setting up start-ups and/or spin-offs would be too premature. However, all partners have identified a number of potential exploitable items, which will be periodically refined and revised throughout the course of the project. Additionally, a set of exploitation KPIs will be defined by the end of year 1 (October 2014). The main exploitable items that have been identified, apart from the S-CASE platform as a whole (which aims to accelerate the overall agile software prototyping process), are the Automatic Semantic Annotation and developer tools, the Open Source Resource Finder and the Question Answering Mechanism. Next we plan on conducting a market research on the S-CASE niche, i.e. multi-modal requirements, model driven engineering and restful web-services. From this analysis we plan on identifying key potential industrial actors, S-CASE customers and users.
Being early in the project, we have probed for future collaborations and for adopting work already done in other projects. Specifically, we have closely investigated work done in projects MARKOS, 4CaaSt, and MODAClouds and consider that the MARKOS project could potentially aid us in the development of the open source resource finder. We have established communication and have started working with the MARKOS API. This approach is expected to lead to tight collaboration. Recently we started collaboration with project STORM CLOUDS, aiming to exchange know-how and expertise. S-CASE members have been participating in meetings and providing feedback whenever requested. With respect to standards contribution, we have already decided to adopt the WSDL and WADL standards for describing S-CASE produced and consumed web services and we are aiming to contribute in that direction. Additionally, we are also working towards standardising work on UML profiles for RESTful application development.