logseq/pages/@Streamlining the Engineering and Distribution of Graphical DSLs based on Recurring Patterns.md

tags:: [[/unread]], [[#zotero]]
title:: @Streamlining the Engineering and Distribution of Graphical DSLs based on Recurring Patterns
item-type:: [[journalArticle]]
original-title:: Streamlining the Engineering and Distribution of Graphical DSLs based on Recurring Patterns
language:: en
authors:: [[Daniel Sami Mitwalli]], [[Marco Krumrey]], [[Sebastian Teumert]]
library-catalog:: Zotero
links:: [Local library](zotero://select/library/items/8L6UM4C9), [Web library](https://www.zotero.org/users/1039502/items/8L6UM4C9)

- [[Abstract]]
	- In the context of general purpose LCNC platforms, where the visual modeling languages can be defined possibly easily and supported in a possibly systematic and uniform fashion, we examine the challenges of creating and distributing graphical Domain-Specific Languages (DSLs) and identify key patterns within the process that form the basis of Cinco Cloud, our cloud based LCNC platform. First, we outline essential functionalities for Integrated Modeling Environments (IMEs) for graphical DSLs, most of which can be derived from the highlevel DSL specifications. This recurring task can be simplified by automatically generating or configuring IMEs from these specifications, or by programmatically extending them with additional semantics. Second, we explores how an integrated language repository within the Cinco Cloud architecture can facilitate the distribution and reuse of DSLs, while streamlining the overall language engineering process. Architectural concerns are in fact at the core of the tension between uniformity and specialization. Finally, we discuss future opportunities for enhancing language versioning, improving discoverability, and integrating AI tools to support the design and use of DSLs.
- ### Attachments
	- [PDF](zotero://select/library/items/AMAMTJKN) {{zotero-imported-file AMAMTJKN, "Mitwalli et al. - Streamlining the Engineering and Distribution of Graphical DSLs based on Recurring Patterns.pdf"}}
- ### Notes
	- # Annotazioni 
	   (20/1/2025, 17:50:34)
	  
	  - “Abstract—In” (Mitwalli et al., p. 4) #ffd400
	  *The abstract is not effective in presenting what are the problems in the focus of the paper in making a clear sttement of what is the proposed solution of the work. *
	  
	   
	  
	  - “LCNC” (Mitwalli et al., p. 4) #ffd400
	  *This is the first occurrence of this acronym and as a such it is necessary to put it here in the extended manner. *
	  
	   
	  
	  - “possibly” (Mitwalli et al., p. 4) #ffd400
	  * *
	  
	   
	  
	  - “key patterns” (Mitwalli et al., p. 4) #ffd400
	  *what kind of patterns are you looking for? Patterns of what? *
	  
	   
	  
	  - “project 1” (Mitwalli et al., p. 4) #ff6666
	  * *
	  
	   
	  
	  - “In that sense, we will demonstrate the potential of Cinco Cloud in streamlining language engineering processes and discuss the future development of such repositories in the language engineering landscape.” (Mitwalli et al., p. 4) #5fb236
	  * *
	  
	   
	  
	  - “Several metamodeling tools address this problem by generating dedicated IMEs from a high-level specification that are then tailored to a specific set of graphical DSLs such as Eclipse Sirius Web [2] 2, WebGME [10] 3 and others [11, 1, 9, 7].” (Mitwalli et al., p. 5) #ffd400
	  * *
	  
	   
	  
	  - “[2] 2” (Mitwalli et al., p. 5) #ff6666
	  * *
	  
	   
	  
	  - “There, language engineers can specify graphical languages using two high-level textual specification languages:” (Mitwalli et al., p. 5) #ffd400
	  *The novelty is not clarified. *
	  
	   
	  
	  - “Meta Graph Language (MGL)” (Mitwalli et al., p. 5) #2ea8e5
	  * *
	  
	   
	  
	  - “Meta Style Language (MSL)” (Mitwalli et al., p. 5) #2ea8e5
	  * *
	  
	   
	  
	  - “TypeScript” (Mitwalli et al., p. 5) #5fb236
	  * *
	  
	   
	  
	  - “By abstracting the complexities of IME development, Cinco Cloud allows users to focus on designing DSLs and deploying intuitive graphical modeling tools.” (Mitwalli et al., p. 5) #ffd400
	  *Still not clear what are the limitations of existing approaches that authors aim at addressing, and thus what are the novelties of Cinco Cloud concerning the development of domain-specific modeling languages. *
	  
	   
	  
	  - “uploading the associated language to a central language repository.” (Mitwalli et al., p. 5) #ffd400
	  *THis is potentially interesting even though it is not clear what are the provided reusability mechanisms. *
	  
	   
	  
	  - “A domain expert simply selects the language through a form and a dedicated IME is automatically deployed in the Cinco Cloud, ready to use. Under the hood, a general purpose IME is deployed, the selected language(s) fetched from the central language repository into the IME, and thereby the IME is configured for the language(s).” (Mitwalli et al., p. 5) #ffd400
	  *What's new with what MetaEdit++ or other modeling platforms already provide? *
	  
	   
	  
	  - “These are provided automatically by Cinco Cloud 4” (Mitwalli et al., p. 5) #a28ae5
	  * *
	  
	   
	  
	  - “abstract and concrete syntax of a graphical DSL” (Mitwalli et al., p. 5) #5fb236
	  * *
	  
	   
	  
	  - “Eclipse Layout Kernel (ELK)” (Mitwalli et al., p. 6) #a28ae5
	  * *
	  
	   
	  
	  - “. Dyamic Components and Functionalities” (Mitwalli et al., p. 6) #ffd400
	  *Over the last few years there has been a proliferation of low-code platforms that permit to develop applications by reusing components that embody recurring behaviours. It is necessary to compare the propose approach with such a line of research. *
	  
	   
	  
	  - “CertBot 8” (Mitwalli et al., p. 7) #ffd400
	  * *
	  
	   
	  
	  - “Langium-based 11 language server” (Mitwalli et al., p. 7) #ffd400
	  * *
	  
	   
	  
	  - “easing the distribution and composition of languages, both from language-designer to language-designer, as well as from language-designer to domain-expert (a holistic approach)” (Mitwalli et al., p. 8) #ffd400
	  *This concept of language composition is not clear. It is relevant, but it is not clear how the proposed approach has addressed such a problem. *
	  
	   
	  
	  - “Sprotty Client 13 and GLSP server” (Mitwalli et al., p. 9) #a28ae5
	  * *
	  
	   
	  
	  - “As mentioned before, our language repository currently only supports a single version of each language.” (Mitwalli et al., p. 10) #5fb236
	  * *
	  
	   
	  
	  - “Although the current state of the central language repository is an independent microservice, it is not yet completely independent from the overall Cinco Cloud archetecture.” (Mitwalli et al., p. 10) #5fb236
	  * *
	  
	   
	  
	  - “archetecture” (Mitwalli et al., p. 10) #ff6666
	  * *
	  
	   
	  
	  - “an independently deployable service that can be integrated with other cloud-based archetectures for LCNC platforms, including a generalized API.” (Mitwalli et al., p. 10) #ffd400
	  *I'm not sure I got the plan of the authors. The repository is supposed to store languages developed with the Cinco Cloud and to enable their reuse. What does it mean enabling the integration of the repository in other LCNC platforms? How to to ensure the their interoperability/compatibility? *
	  
	   
	  
	  - “common patterns” (Mitwalli et al., p. 10) #ffd400
	  *it is not clear in the paper what are the common patterns that have been identified with respect to already existing primitives that existing (meta)modeling platforms already implement. *
	  
	   
	  
	  - “recurring patterns” (Mitwalli et al., p. 10) #ffd400
	  *if the authors are referring to the content given in Section III, it is not clear what's their novelty. *
	  
	   
	  
	  - “the externalization of a centralized language repository.” (Mitwalli et al., p. 10) #ffd400
	  *I see this as potential novelty of the proposed approach and as a such I would restructure the paper and rebalance the content and focus in this direction, instead of pushing the recuccring patterns concepts, that instead are recurring functionalities that are expected by any modeling environments nowadays. *