In August 2017, the Enlarged Directorate endorsed a new policy which defines the framework of software dissemination activities at CERN, contributing to a more coherent approach in the licensing and dissemination of the Organization’s software assets. As a member of the CERN community, your work may lead to potential applications across a wide range of industries beyond CERN. In order to maximise the dissemination, there is a number of important implications to consider when developing software.
CERN is a collaborative environment, where developers from different organisations, institutes, and countries often contribute to various degrees of a given software project. Despite its uniqueness as a fundamental research laboratory, CERN does not differ from industry or academia with respect to software creation. The majority of the laboratory’s output is Components Based Software (CBS), and software that does not contain external components is the exception rather than the norm.
Development teams often focus on the desired outcome, employing the component that seems most suitable to satisfy the project’s requirements. As a result, code may not always be properly identified and documented. This approach is perfectly valid if the CBS is not to be distributed outside CERN, but from a dissemination perspective, it may contribute to complicate the path as the licensing scheme of the different components are not taken into account.
The new software dissemination policy therefore recommends ways of anticipating software dissemination in order to facilitate it. The possible dissemination paths are greatly influenced by the freedom to choose a licensing model, and falls into four different cases depending on the involvement of external contributors in the development, and the usage of external code components.
In order to assess the dissemination potential, every software technology is evaluated on a case-by-case basis, considering the needs and aspirations of the developer’s team. The Knowledge Transfer (KT) group is responsible for facilitating the transfer of CERN technology, undertaking a number of concrete actions to promote CERN software technologies, also providing support and expertise to assist in all steps of the dissemination process. Numerous technologies are brought to the attention of CERN’s KT group every year, leading to various collaboration agreements and knowledge transfer activities with industry.
In 2017, a licence agreement was signed between CERN and the leading global display manufacturer LG Display, giving them access to controls middleware software from CERN to be used in factory automation across their plants. The software was originally developed by the BE-CO group for the LHC to provide a common software communication infrastructure for the accelerator controls, but will now be adapted to its new application by LG-Display with the support of the development team.
Another knowledge transfer activity conducted was a four-day training course on machine learning, in which a team of experts from CERN’s EP-SFT group shared their expertise with Sanofi Pasteur, the vaccines business unit of the global life science company Sanofi. The aim was to improve vaccine production by using ROOT, the data analysis framework used to analyse HEP data, and the Toolkit for Multivariate Data Analysis (TMVA), a library of associated machine learning algorithms.
The possible disseminations paths for software technologies are many, and if you want to learn more about how KT can support the process you can access all CERN KT policy documents here.
The new CERN-wide policy applies to all CERN Software, addressing both proprietary licensing and open source licensing, and complements the more general framework ‘Policy on the Management of Intellectual Property in Technology Transfer Activities at CERN’ (the “CERN IP Policy”) while taking into account the recommendations of the ‘Final Report OSL-2012 – Main Volume’ by the ‘Open Source Licence Task Force’.
This article was published on the main CERN homepage.