Periodic Reporting for period 2 - Lowcomote (Training the Next Generation of Experts in Scalable Low-Code Engineering Platforms)
Período documentado: 2021-01-01 hasta 2023-06-30
Low-code development platforms (LCDPs) are software development platforms on the Cloud, provided through a Platform-as a-Service model, which allow users to build completely operational applications by interacting through dynamic graphical user interfaces, visual diagrams and declarative languages. They address the need of non-programmers to develop personalized software, and focus on their domain expertise instead of implementation requirements.
Lowcomote trains a generation of experts that will upgrade the current trend of LCDPs to a new paradigm, low-code engineering platforms (LCEPs). LCEPs will be open, allowing to integrate heterogeneous engineering tools, interoperable, allowing for cross-platform engineering, scalable, supporting very large engineering models and social networks of developers, smart, simplifying the development for citizen developers by machine learning and recommendation techniques. This will be achieved by injecting in LCDPs the theoretical and technical framework defined by recent research in Model-Driven Engineering (MDE), augmented with Cloud Computing and Machine Learning techniques.
The 54-month Lowcomote project trains the first European generation of skilled professionals in LCEPs. The 15 future scientists benefit from an original training and research program merging competencies and knowledge from 5 highly-recognized academic institutions and 9 large and small industries of several domains.
To reach its main scientific goal, Lowcomote integrates an interdisciplinary and inter-sectoral research program around three specific research objectives (RO) and three training objectives (TO). Lowcomote trains ESRs capable of:
* RO1: Enabling Low-code Engineering of Large-Scale Heterogeneous Systems (WP3), by smart development environments on the Cloud and precise integration of low-code languages with new domains.
* RO2: Developing a Large-scale Repository and Services for Low-Code Engineering (WP4), as a Cloud-based service able to handle a very large number of low-code artifacts, and automatically learn from them.
* RO3: Producing advancements in Scalable Low-Code Artifact Management (WP5), as new algorithms and reusable components.
The Lowcomote training program:
* TO1: Provides ESRs with scientific knowledge beyond the state of the art to strengthen career perspectives.
* TO2: Maximises exposure to different environments, in order to give ESRs the possibility to understand the whole chain of value in their domain. Multiple exposures enhance creativity, agility and value generation.
* TO3: Provides ESRs with communication, management and career development skills, enabling them to succeed in any sector.
Lowcomote trains a generation of experts that will upgrade the current trend of LCDPs to a new paradigm, low-code engineering platforms (LCEPs). LCEPs will be open, allowing to integrate heterogeneous engineering tools, interoperable, allowing for cross-platform engineering, scalable, supporting very large engineering models and social networks of developers, smart, simplifying the development for citizen developers by machine learning and recommendation techniques. This will be achieved by injecting in LCDPs the theoretical and technical framework defined by recent research in Model-Driven Engineering (MDE), augmented with Cloud Computing and Machine Learning techniques.
The 54-month Lowcomote project trains the first European generation of skilled professionals in LCEPs. The 15 future scientists benefit from an original training and research program merging competencies and knowledge from 5 highly-recognized academic institutions and 9 large and small industries of several domains.
To reach its main scientific goal, Lowcomote integrates an interdisciplinary and inter-sectoral research program around three specific research objectives (RO) and three training objectives (TO). Lowcomote trains ESRs capable of:
* RO1: Enabling Low-code Engineering of Large-Scale Heterogeneous Systems (WP3), by smart development environments on the Cloud and precise integration of low-code languages with new domains.
* RO2: Developing a Large-scale Repository and Services for Low-Code Engineering (WP4), as a Cloud-based service able to handle a very large number of low-code artifacts, and automatically learn from them.
* RO3: Producing advancements in Scalable Low-Code Artifact Management (WP5), as new algorithms and reusable components.
The Lowcomote training program:
* TO1: Provides ESRs with scientific knowledge beyond the state of the art to strengthen career perspectives.
* TO2: Maximises exposure to different environments, in order to give ESRs the possibility to understand the whole chain of value in their domain. Multiple exposures enhance creativity, agility and value generation.
* TO3: Provides ESRs with communication, management and career development skills, enabling them to succeed in any sector.
The project organized 10 one-week network events. The first 3 network events were focused on giving all the ESRs the base training in Model-Driven Engineering, Cloud Computing, and Data Science, required to address the 3 research objectives RO1-RO3 of the project. The ESRs also performed a wide set of individual training activities within their hosting institution. We organized three open workshops that gave the chance to ESRs to learn the state-of-the-art from well-known researchers and practitioners, and to exchange information about their research work. We provided all ESRs with specific training on a set of communication, management and career development skills.
Each work package produced a set of exploitable results:
* WP3 (Low-code Engineering of Large-Scale Heterogeneous Systems) produced components for large-scale heterogeneous low-code engineering (Dandelion), smart low-code by recommender systems (Droid), and applications of low-code in the domains of machine learning (Panoptes), IoT (CHESSIoT), augmented reality (ALTER).
* WP4 (Large-scale Repository and Services for Low-Code Engineering) produced a large-scale repository for low-code engineering artifacts (MDEForgeWL), and server-side components for analyzing user interaction events (Modeling Event Recorder), providing modeling recommendations (BORA), parse and generate models from DevOps artifacts (JSONSchemaDSL), performing low-code testing and test amplification (TAMO).
* WP5 (Scalable Low-Code Artifact Management) produced artifact management algorithms and components for static analysis of query languages (Epsilon static analyzer), query optimization (Epsilon query optimizer), a high-performance continuous consistency checker, a transformation engine on Spark (SparkTE), a tool for automatic chaining of model transformations.
The research underlying these tools has been disseminated in more than fifty publications in top-level academic conferences and journals. Our ESRs received two international awards and one national French award. We held a series of 4 industry-oriented events with participation of low-code vendors and users like Bosch, Blackbelt, TypeFox, Gitpod, Axellience, NASA, Boeing, Eclipse CDO and Dynatrace. Finally, we created the LowCode workshop series in the (A-rank) MODELS conference: the workshop series will outlive the project fostering the creation of an international academic and industrial community around low-code.
Each work package produced a set of exploitable results:
* WP3 (Low-code Engineering of Large-Scale Heterogeneous Systems) produced components for large-scale heterogeneous low-code engineering (Dandelion), smart low-code by recommender systems (Droid), and applications of low-code in the domains of machine learning (Panoptes), IoT (CHESSIoT), augmented reality (ALTER).
* WP4 (Large-scale Repository and Services for Low-Code Engineering) produced a large-scale repository for low-code engineering artifacts (MDEForgeWL), and server-side components for analyzing user interaction events (Modeling Event Recorder), providing modeling recommendations (BORA), parse and generate models from DevOps artifacts (JSONSchemaDSL), performing low-code testing and test amplification (TAMO).
* WP5 (Scalable Low-Code Artifact Management) produced artifact management algorithms and components for static analysis of query languages (Epsilon static analyzer), query optimization (Epsilon query optimizer), a high-performance continuous consistency checker, a transformation engine on Spark (SparkTE), a tool for automatic chaining of model transformations.
The research underlying these tools has been disseminated in more than fifty publications in top-level academic conferences and journals. Our ESRs received two international awards and one national French award. We held a series of 4 industry-oriented events with participation of low-code vendors and users like Bosch, Blackbelt, TypeFox, Gitpod, Axellience, NASA, Boeing, Eclipse CDO and Dynatrace. Finally, we created the LowCode workshop series in the (A-rank) MODELS conference: the workshop series will outlive the project fostering the creation of an international academic and industrial community around low-code.
Lowcomote provides a set of components for developing new open LCEPs, delivered as a Platform-as-a-Service, based on open standards at all levels (e.g. EMF, language-server-protocol, OpenStack). The platform is composed of two main parts, a frontend and a backend. A frontend provides low-code engineering languages and smart interfaces that end-users will exploit to produce their artifacts. A backend comes in the form of a LCEP repository providing server-side services for LCEP developers. Finally, the project produces the fundamental bricks used by frontend and backend for efficiently manipulating LCEP artifacts. The work in the first period produced theoretical bases, specifications and first prototypes for these components, in the second period the research was validated and implementations were made more mature.
We envision that LCEPs are going to radically transform the software construction landscape in the next few years, and will be the next step in the LCDP wave. Digitalization of the society and the complete industry is one of the key challenges of the EU in the upcoming decade. Providing novel interfaces for the development of software systems is one of the main enablers for the digital transformation. LCEPs are giving access to non-programmers in building business-ready applications either for industrial or academic purposes. The capacity of LCEPs to transform domain experts (e.g. mechanical, civil, industrial engineers) into citizen developers makes them an active part of the ongoing digital transformation, instead of passive receivers. By enhancing low-code technologies, Lowcomote will be strengthening the position of EU software companies worldwide and improving Europe’s competitiveness and growth.
Lowcomote’s ESRs have the competences needed to construct new LCEPs, upgrade LCDPs to LCEPs or extend existing LCEPs to address new specific industrial domains. ESRs of Lowcomote will also be prone to prevent citizen developers from developing low-code of low quality. ESRs have the potential to be employed in several fields like energy, transport, manufacturing, and will impact different sectors of industry. Lowcomote trains the first generation of LCEP experts, and they are in Europe.
We envision that LCEPs are going to radically transform the software construction landscape in the next few years, and will be the next step in the LCDP wave. Digitalization of the society and the complete industry is one of the key challenges of the EU in the upcoming decade. Providing novel interfaces for the development of software systems is one of the main enablers for the digital transformation. LCEPs are giving access to non-programmers in building business-ready applications either for industrial or academic purposes. The capacity of LCEPs to transform domain experts (e.g. mechanical, civil, industrial engineers) into citizen developers makes them an active part of the ongoing digital transformation, instead of passive receivers. By enhancing low-code technologies, Lowcomote will be strengthening the position of EU software companies worldwide and improving Europe’s competitiveness and growth.
Lowcomote’s ESRs have the competences needed to construct new LCEPs, upgrade LCDPs to LCEPs or extend existing LCEPs to address new specific industrial domains. ESRs of Lowcomote will also be prone to prevent citizen developers from developing low-code of low quality. ESRs have the potential to be employed in several fields like energy, transport, manufacturing, and will impact different sectors of industry. Lowcomote trains the first generation of LCEP experts, and they are in Europe.