Section 6 contains information about related work, and in Sect. This section will also include propositions for much needed validation of claims about model transformation languages synthesized from the literature review. 5, we discuss the results of our findings. 3, we will detail the methodology used for the conducted literature review. 2 introduces the background of this research, model-driven engineering and model transformation languages. The remainder of this paper is structured as follows: Sect. We hope our results can provide an overview over what MTLs are envisioned to achieve, what current research suggests they do and where further research to validate the claimed properties is necessary.
There is a lack of research interest in evaluation and especially verification of claimed advantages and disadvantages. There is a number of claims that originate in claims about DSLs without proper evidence why they hold for MTLs too. There is insufficient or no evidence for a large portion of claims. The current literature claims many advantages and disadvantages of model transformation languages.Ī large portion of claims are very broad. Afterwards, we categorized claims and systematized the evidence to produce (i) a categorization of claimed advantages and disadvantages into 15 separate categories (namely analysability, comprehensibility, conciseness, debugging, ease of writing a transformation, expressiveness, extendability, just better, learnability, performance, productivity, reuse and maintainability, tool support, semantics and verification, versatility) and (ii) a systematic representation of which claims are verified through what means. As a first step, during the review we selected 58 publications from which to extract claims and evidence for advantages and disadvantages of model transformation languages. based on the research questions we defined (see Sect.
To systematize information from the literature, we performed a systematic literature review The study can also be of interest to practitioners who wish to gain an overview over what research claims about MTLs compared to a practitioners view of the matter. This study is intended for researchers to (i) raise awareness for the current state of research and (ii) incentivise further research in areas where we identified gaps.
Insights into the state of verification of aforementioned advantages and disadvantages For this purpose, we performed a systematic review of claims and evidence in the literature.Ī systematic review and overview over the advantages and disadvantages of model transformation languages as claimed in the literature We do not intend to provide a complete overview over the current state of the art in research. The goal of this study is to identify and categorize claims about advantages and disadvantages of model transformation languages made throughout the literature and to gather available evidence thereof. Nowadays, such an abundance of claims runs through the whole literature body that one can be forgiven when losing track of which claims verifiably apply and which are still purely visionary. Such claims are reiterated time and time again in the literature, often without any actual evidence. Model transformation languages (MTLs), being domain-specific languages, have ever since been associated with advantages in areas like productivity, expressiveness and comprehensibility compared to general-purpose programming languages (GPLs) , have been an integral part of model-driven development. Ever since the dawn of model-driven engineering at the beginning of the century, model transformations, supported by dedicated transformation languages