Bachelor and Master Theses

Model-based methodlogies are gaininig increasing interest from industry as an effective way to tackle the complexity of modern systems. These methodologies, despite having reached an adequate level of maturity (seen e.g. in Lowcode platforms and generative content creation), still suffer of remarkable tool usability issues, like steep learning curve, difficult customization, scarcely supported team development, etc. One of the basic issues related to tools is due to the technical foundations on which the modelling language workbench is conceived:

• generative modelling workbenches: the language engineer specifies the characteristics of the language for which automated generation mechanisms derive the corresponding modelling support (editors, validators, etc.);
• reflective modelling workbenches: the language engineer specifies the characteristics of the language directly extending the platform, such that the language continuously evolves through the changes.

In general the former platforms provide an easier way of managing the language creation procedure but are by far less malleable to modifications. Moreover, specifying editors in terms of tool components to be generated is typically not straightforward and requires extensive hand-tuning of generated supporting structures. The latter typically require a more extensive effort in the bootstrap phases to provide a minimal set of language engineering features, however once that base is provided the modelling support is flexible and lightweight.

JJODEL is a cloud-based, reflective language workbench that has been conceived since the beginning to provide a lightweight solution to meta-modelling (https://github.com/MDEGroup/jjodel/tree/master). The workbench provides many advanced features like positional editing, collaborative development, co-evolution, and more [1]. Despite the platform already provides such features, it is still a work-in-progress endeavour. In this respect, any language workbench cannot be considered complete until it does not provide support for model-to-model transformations (M2M).

This thesis project aims at developing a "driver" between JJODEL and a web-based service for M2M. In short, the thesis work is expected to design and implement a bridge between JJODEL and the M2M service; such bridge might include exporting (part of) models from JJODEL, communication with APIs provided by the M2M service, adequate back-propagation of transformation results to JJODEL (so importing part of models). The target M2M service is Epsilon Transformation Language (ETL) https://eclipse.dev/epsilon/doc/etl/

1. Juri Di Rocco, Davide Di Ruscio, Amleto Di Salle, Damiano Di Vincenzo, Alfonso Pierantonio, and Giordano Tinella. 2023. Jjodel - A Reflective Cloud-Based Modeling Framework. In 2023 ACM/IEEE International Conference on Model Driven Engineering Languages and Systems Companion (MODELS-C). IEEE Press, 55–59. https://doi.org/10.1109/MODELS-C59198.2023.00019  

- Excellent understanding of (meta-)modelling concepts, preferably having already taken a model-driven course;

- Experience in web technologies and programming like React, JavaScript, etc.; 

Suitable also for 2 students

This thesis is done in collaboration with the University of L'Aquila. A co-supervisor from there might be appointed to support the development on the implementation part.