Files
logseq/pages/TOSEM-2024-0332_Proof_hi.md
T
2025-06-05 22:07:12 +02:00

142 lines
23 KiB
Markdown
Raw Blame History

This file contains ambiguous Unicode characters
This file contains Unicode characters that might be confused with other characters. If you think that this is intentional, you can safely ignore this warning. Use the Escape button to reveal them.
type:: [[REVIEWS]]
tags::
year:: 2024
venue:: [[TOSEM]]
full-title:: Stepwise Domain Modeling with Auto-Generated DSMLs: A Model Driven Method
date-start:: [[15-07-2024]] - 16:43
date-submitted:: [[17-07-2024]]
external-links::
status:: [[DONE]]
deadline-submission:: [[07-07-2024]]
file:: [[@TOSEM-2024-0332_Proof_hi]]
parent::
todoist:: https://app.todoist.com/app/task/tosem-2024-0332-reviewer-agreed-6VQxHHfQ4vQvQVcg
- ### [[Highlights]]
-
- # Annotazioni
- (17/7/2024, 15:09:35)
- - “Domain Modeling using automatically generated DomainSpecific Modeling Lan” ([“TOSEM-2024-0332_Proof_hi”, p. 1](zotero://select/library/items/WC6HAAWF)) ([pdf](zotero://open-pdf/library/items/4YYDZM6X?page=2&annotation=WW6MXSS9)) #5fb236
- - ““Intino” ([“TOSEM-2024-0332_Proof_hi”, p. 1](zotero://select/library/items/WC6HAAWF)) ([pdf](zotero://open-pdf/library/items/4YYDZM6X?page=2&annotation=Q9KD3QST)) #5fb236
- - “having three separate models.” ([“TOSEM-2024-0332_Proof_hi”, p. 2](zotero://select/library/items/WC6HAAWF)) ([pdf](zotero://open-pdf/library/items/4YYDZM6X?page=3&annotation=X44GF6GY)) #f0ff00
- *Why three? What are you referring to?*
- - “The final outcome is a unified model that represents the domain across multiple levels of abstraction despite being expressed with three different DSMLs. However, this model not only facilitates the automatic generation of the necessary DSMLs to build the model itself but also enables” ([“TOSEM-2024-0332_Proof_hi”, p. 2](zotero://select/library/items/WC6HAAWF)) ([pdf](zotero://open-pdf/library/items/4YYDZM6X?page=3&annotation=DI6D6QFN)) #f0ff00
- *Thre motivation for having three different languages is not clear.*
- - “[32],expressed” ([“TOSEM-2024-0332_Proof_hi”, p. 3](zotero://select/library/items/WC6HAAWF)) ([pdf](zotero://open-pdf/library/items/4YYDZM6X?page=4&annotation=FME2I2PA)) #ff4400
- - “• Model. A model is an abstraction that defines specific aspects of a system. Models are used to capture and convey domain knowledge, requirements, structure, behavior, and other characteristics of a system, facilitating a deeper understanding before moving to the implementation phase. • Metamodel.” ([“TOSEM-2024-0332_Proof_hi”, p. 4](zotero://select/library/items/WC6HAAWF)) #00b036
- - “metamodels. • Metametamodel.” ([“TOSEM-2024-0332_Proof_hi”, p. 4](zotero://select/library/items/WC6HAAWF)) #00b036
- - “metam” ([“TOSEM-2024-0332_Proof_hi”, p. 5](zotero://select/library/items/WC6HAAWF)) ([pdf](zotero://open-pdf/library/items/4YYDZM6X?page=6&annotation=PFGFFYAJ)) #ffd400
- *Ref [40] should have been used instead to metadepth, which is an approach that is related to the one presented by this paper.*
- - “Three fundamental features distinguish Intino from current MDE approaches. Firstly, it proposes the implementation of a single and unified model that integrates the three levels of abstraction, diverging from the traditional MDE practice of utilizing thre” ([“TOSEM-2024-0332_Proof_hi”, p. 6](zotero://select/library/items/WC6HAAWF)) ([pdf](zotero://open-pdf/library/items/4YYDZM6X?page=7&annotation=AMW3QDFC)) #ffd400
- *Why is this needed? It is important to motivate it.*
- - “Ψ : W → D,” ([“TOSEM-2024-0332_Proof_hi”, p. 6](zotero://select/library/items/WC6HAAWF)) ([pdf](zotero://open-pdf/library/items/4YYDZM6X?page=7&annotation=S5KQ8GFP)) #f0ff00
- - “Single Domain Model” ([“TOSEM-2024-0332_Proof_hi”, p. 7](zotero://select/library/items/WC6HAAWF)) ([pdf](zotero://open-pdf/library/items/4YYDZM6X?page=8&annotation=DMQTXISB)) #ffd400
- *How the proposed approach related to multimodeling. WHy is it relevant to have only one model, instead of defining different models at different layers (see metadepth)*
- - “elemental unit of modeling” ([“TOSEM-2024-0332_Proof_hi”, p. 7](zotero://select/library/items/WC6HAAWF)) ([pdf](zotero://open-pdf/library/items/4YYDZM6X?page=8&annotation=Y9ZSZI2J)) #ffd400
- *The used terminology is not always aligned with that from the modeling community. For intance, mograms, element unit of modelings are not used by the MDE community.*
- - “This transformation relies on a novel construct: every DSML is an extension of a mother language, analogous to the inheritance [7] in object-oriented programming, where a general class serves a the basis for deriving specialized classes. Therefore, the transformation involves specializing the mother language to create a DSML. This construct of mother language will be examined in detail in subsection 3.5” ([“TOSEM-2024-0332_Proof_hi”, p. 8](zotero://select/library/items/WC6HAAWF)) ([pdf](zotero://open-pdf/library/items/4YYDZM6X?page=9&annotation=XA7PLGUY)) #ffd400
- *we need some examples, sentences like this are not very clear to me.*
- - “three levels” ([“TOSEM-2024-0332_Proof_hi”, p. 8](zotero://select/library/items/WC6HAAWF)) ([pdf](zotero://open-pdf/library/items/4YYDZM6X?page=9&annotation=D6BTSPHG)) #f0ff00
- *why three_ It can be even more, isn't it
- again, the link with multimodeling is missing here.*
- - “lexicon” ([“TOSEM-2024-0332_Proof_hi”, p. 8](zotero://select/library/items/WC6HAAWF)) ([pdf](zotero://open-pdf/library/items/4YYDZM6X?page=9&annotation=5859CYPF)) #5fb236
- - “syntax” ([“TOSEM-2024-0332_Proof_hi”, p. 8](zotero://select/library/items/WC6HAAWF)) ([pdf](zotero://open-pdf/library/items/4YYDZM6X?page=9&annotation=I4TZN8Q3)) #5fb236
- - “semantic” ([“TOSEM-2024-0332_Proof_hi”, p. 8](zotero://select/library/items/WC6HAAWF)) #5fb236
- - “This means that M2 mograms can be represented with either λ2 or λ3” ([“TOSEM-2024-0332_Proof_hi”, p. 8](zotero://select/library/items/WC6HAAWF)) ([pdf](zotero://open-pdf/library/items/4YYDZM6X?page=9&annotation=QTI6H2ZE)) #ffd400
- *Why do we need such a complication?*
- - “flexibility” ([“TOSEM-2024-0332_Proof_hi”, p. 8](zotero://select/library/items/WC6HAAWF)) #ffd400
- *This can overcomplicate the modeling phases, isn't it?*
- - “Definition.” ([“TOSEM-2024-0332_Proof_hi”, p. 8](zotero://select/library/items/WC6HAAWF)) #ffd400
- *I think we need some explanatory examples to make such definitions concrete and related to real cases.*
- - “each step in this progression contributes to the completion of the domain model.” ([“TOSEM-2024-0332_Proof_hi”, p. 8](zotero://select/library/items/WC6HAAWF)) ([pdf](zotero://open-pdf/library/items/4YYDZM6X?page=9&annotation=85BYMV7C)) #ffd400
- *What do you mean? Examples, please!*
- - “Stepwise modeling process in Intino.” ([“TOSEM-2024-0332_Proof_hi”, p. 9](zotero://select/library/items/WC6HAAWF)) ([pdf](zotero://open-pdf/library/items/4YYDZM6X?page=10&annotation=LWDWPN5P)) #ffd400
- *Where is the mother language here?*
- - “the automatic DSML generation involves” ([“TOSEM-2024-0332_Proof_hi”, p. 9](zotero://select/library/items/WC6HAAWF)) ([pdf](zotero://open-pdf/library/items/4YYDZM6X?page=10&annotation=242ZYPP4)) #5fb236
- - “It defines an universal syntax, and provides a lexicon and a semantics that can be extended.” ([“TOSEM-2024-0332_Proof_hi”, p. 9](zotero://select/library/items/WC6HAAWF)) ([pdf](zotero://open-pdf/library/items/4YYDZM6X?page=10&annotation=X9I85VAJ)) #5fb236
- - “is a mother” ([“TOSEM-2024-0332_Proof_hi”, p. 10](zotero://select/library/items/WC6HAAWF)) ([pdf](zotero://open-pdf/library/items/4YYDZM6X?page=11&annotation=6X6P9EN4)) #ffd400
- *Why do we need such a definition with the "if" statement? Is it possible to have several mother languages?*
- - “λˆ is domain-agnostic” ([“TOSEM-2024-0332_Proof_hi”, p. 10](zotero://select/library/items/WC6HAAWF)) ([pdf](zotero://open-pdf/library/items/4YYDZM6X?page=11&annotation=FYZUKXMN)) #ffd400
- - “determine the domain (D).” ([“TOSEM-2024-0332_Proof_hi”, p. 10](zotero://select/library/items/WC6HAAWF)) ([pdf](zotero://open-pdf/library/items/4YYDZM6X?page=11&annotation=TZ34VIYC)) #ffd400
- *How can this analysis be performed automatically? Is it tool supported?*
- - “automatic creation of DSML” ([“TOSEM-2024-0332_Proof_hi”, p. 10](zotero://select/library/items/WC6HAAWF)) ([pdf](zotero://open-pdf/library/items/4YYDZM6X?page=11&annotation=WJLZF7KY)) #ffd400
- *AUTOMATIC? We really need running examples to help the reader understand the different steps of the proposed process.*
- - “if multiple domains share the same analytical constructs, it is possible to create a mother language capable of universally expressing th” ([“TOSEM-2024-0332_Proof_hi”, p. 11](zotero://select/library/items/WC6HAAWF)) ([pdf](zotero://open-pdf/library/items/4YYDZM6X?page=12&annotation=J5QIPHPC)) #5fb236
- - “the potential to use universal analytical constructs that possesses the versatility to be applied across heterogeneous domains; and a language whose syntactic structures support these constructs, thereby suggesting its applicability as a universal syntax across multiple DSMLs.” ([“TOSEM-2024-0332_Proof_hi”, p. 11](zotero://select/library/items/WC6HAAWF)) ([pdf](zotero://open-pdf/library/items/4YYDZM6X?page=12&annotation=QRWG3RUR)) #ffd400
- *This is vague ...*
- - “different domains can be analyzed through a set analytical” ([“TOSEM-2024-0332_Proof_hi”, p. 11](zotero://select/library/items/WC6HAAWF)) ([pdf](zotero://open-pdf/library/items/4YYDZM6X?page=12&annotation=K43UKFJT)) #5fb236
- - “Tara” ([“TOSEM-2024-0332_Proof_hi”, p. 11](zotero://select/library/items/WC6HAAWF)) ([pdf](zotero://open-pdf/library/items/4YYDZM6X?page=12&annotation=WV9F8TDC)) #5fb236
- - “analytical construct” ([“TOSEM-2024-0332_Proof_hi”, p. 11](zotero://select/library/items/WC6HAAWF)) ([pdf](zotero://open-pdf/library/items/4YYDZM6X?page=12&annotation=UAG97GIJ)) #5fb236
- - “properties, relationships, and semantic constraints” ([“TOSEM-2024-0332_Proof_hi”, p. 11](zotero://select/library/items/WC6HAAWF)) ([pdf](zotero://open-pdf/library/items/4YYDZM6X?page=12&annotation=T2IJTBZB)) #5fb236
- - “e.T” ([“TOSEM-2024-0332_Proof_hi”, p. 11](zotero://select/library/items/WC6HAAWF)) ([pdf](zotero://open-pdf/library/items/4YYDZM6X?page=12&annotation=BRCSB4WJ)) #ff6666
- - “In the ontological approach, entities and concepts are analyzed differently [60]. In contrast, the ontical approach treats concepts as entities, aiming to ensure uniform analysis across diverse levels of abstraction [59].” ([“TOSEM-2024-0332_Proof_hi”, p. 11](zotero://select/library/items/WC6HAAWF)) ([pdf](zotero://open-pdf/library/items/4YYDZM6X?page=12&annotation=IN33Q83E)) #ffd400
- *Without concrete examples, it's not easy to appreciate the sentences like this one. There are many sentences like it.*
- - “Properties” ([“TOSEM-2024-0332_Proof_hi”, p. 12](zotero://select/library/items/WC6HAAWF)) ([pdf](zotero://open-pdf/library/items/4YYDZM6X?page=13&annotation=QJY7FYEU)) #ffd400
- *Until section 5 included, the paper is not easy to follow. It's plenty of definitions without referring to concrete examples or links to existing technologies like EMF to make the discussion more concrete and help the reader to grasp the introduced concepts that are introduced only in an abstract manner.*
- - “descriptive,” ([“TOSEM-2024-0332_Proof_hi”, p. 12](zotero://select/library/items/WC6HAAWF)) ([pdf](zotero://open-pdf/library/items/4YYDZM6X?page=13&annotation=WFMPDEN9)) #5fb236
- - “Descriptive properties” ([“TOSEM-2024-0332_Proof_hi”, p. 12](zotero://select/library/items/WC6HAAWF)) ([pdf](zotero://open-pdf/library/items/4YYDZM6X?page=13&annotation=TAHT6MAC)) #5fb236
- - “Prescriptive properties” ([“TOSEM-2024-0332_Proof_hi”, p. 12](zotero://select/library/items/WC6HAAWF)) ([pdf](zotero://open-pdf/library/items/4YYDZM6X?page=13&annotation=H8RSHYTA)) #5fb236
- - “Relational properties” ([“TOSEM-2024-0332_Proof_hi”, p. 12](zotero://select/library/items/WC6HAAWF)) ([pdf](zotero://open-pdf/library/items/4YYDZM6X?page=13&annotation=XLFNUW3E)) #5fb236
- - “While descriptive properties allow for the detailed representation of each entitys unique features, prescriptive properties define the concept itself, ensuring that all entities classified under a particular concept share certain fundamental characteristics.” ([“TOSEM-2024-0332_Proof_hi”, p. 12](zotero://select/library/items/WC6HAAWF)) ([pdf](zotero://open-pdf/library/items/4YYDZM6X?page=13&annotation=X2PAC2X9)) #5fb236
- - “Fig. 5. Framework for model properties” ([“TOSEM-2024-0332_Proof_hi”, p. 13](zotero://select/library/items/WC6HAAWF)) ([pdf](zotero://open-pdf/library/items/4YYDZM6X?page=14&annotation=A3BTCI7V)) #ffd400
- *Complementing FIg. 5 with some concrete examples is necessary*
- - “TARA LANGUAGE” ([“TOSEM-2024-0332_Proof_hi”, p. 19](zotero://select/library/items/WC6HAAWF)) ([pdf](zotero://open-pdf/library/items/4YYDZM6X?page=20&annotation=EXFKPBDT)) #ffd400
- *Have you considered "metaDepth: A framework for deep meta-modelling" ???*
- - “NL (New Line).” ([“TOSEM-2024-0332_Proof_hi”, p. 19](zotero://select/library/items/WC6HAAWF)) ([pdf](zotero://open-pdf/library/items/4YYDZM6X?page=20&annotation=JRQ2MKTE)) #ffd400
- *Is this really necessary? It's the grammar then, why not using EBNF or alike?*
- - “Syntax” ([“TOSEM-2024-0332_Proof_hi”, p. 20](zotero://select/library/items/WC6HAAWF)) ([pdf](zotero://open-pdf/library/items/4YYDZM6X?page=21&annotation=DDSM3R6B)) #ffd400
- *Why do you need to distinguish lexicon + syntax, when it is possible to present the EBNF specification of the language?
- Sec 6.1 and 6.2 can be merged in my opinion.*
- - “relations with other mograms” ([“TOSEM-2024-0332_Proof_hi”, p. 20](zotero://select/library/items/WC6HAAWF)) ([pdf](zotero://open-pdf/library/items/4YYDZM6X?page=21&annotation=9VTYLXJY)) #ffd400
- *examples?*
- - “Fig. 8. Model Transformations.” ([“TOSEM-2024-0332_Proof_hi”, p. 23](zotero://select/library/items/WC6HAAWF)) ([pdf](zotero://open-pdf/library/items/4YYDZM6X?page=24&annotation=M7WYVSGV)) #ffd400
- *How the code generated at different stages of the process is glued together? What is the underlying execution environment?*
- - “7 SPECIALIZATION OF TARA” ([“TOSEM-2024-0332_Proof_hi”, p. 23](zotero://select/library/items/WC6HAAWF)) ([pdf](zotero://open-pdf/library/items/4YYDZM6X?page=24&annotation=YKANXZQH)) #ffd400
- *Also this section is affected by the same my problem affecting the other sections. Too high level presentation, verbose, and difficult to follow. They are not strict to the point and they are missing discussion on concrete examples and cases.*
- - “specialization is achieved through automatic model transformation” ([“TOSEM-2024-0332_Proof_hi”, p. 23](zotero://select/library/items/WC6HAAWF)) ([pdf](zotero://open-pdf/library/items/4YYDZM6X?page=24&annotation=7MA3MRT7)) #ffd400
- *This is not clear how it is performed.*
- - “By adopting this approach, we leverage the power of MDE to streamline and optimize the translation of abstract models into executable code. This ensures that the generated code is not only accurate and efficient but also adaptable to the evolving needs of software development in different programming languages.” ([“TOSEM-2024-0332_Proof_hi”, p. 23](zotero://select/library/items/WC6HAAWF)) ([pdf](zotero://open-pdf/library/items/4YYDZM6X?page=24&annotation=A6A6YNWS)) #ffd400
- *There are many sentences like this one that are vague and do not convey any technical message.*
- - “It is important to note that this paper does not directly address the process of code generation. This aspect remains an area for future research, where we plan to explore extending our method to include the automated generation of code.” ([“TOSEM-2024-0332_Proof_hi”, p. 23](zotero://select/library/items/WC6HAAWF)) ([pdf](zotero://open-pdf/library/items/4YYDZM6X?page=24&annotation=9B6SZ4YM)) #ffd400
- *Does it mean that the process shown in Fig. 8 has not been fully realized?*
- - “Currently, our focus is solely on demonstrating the process of generating lexicon and semantic rules.” ([“TOSEM-2024-0332_Proof_hi”, p. 23](zotero://select/library/items/WC6HAAWF)) ([pdf](zotero://open-pdf/library/items/4YYDZM6X?page=24&annotation=L2PIDMEJ)) #ffd400
- *Even in this case, my previous point applies!*
- - “Fig. 9. Aviation model by λ” ([“TOSEM-2024-0332_Proof_hi”, p. 25](zotero://select/library/items/WC6HAAWF)) ([pdf](zotero://open-pdf/library/items/4YYDZM6X?page=26&annotation=3REMFEGT)) #ffd400
- *what is the technology to develop such a language? How the used types (i.e., Manufacturer, Airline, etc. defined in Fig. 10) are imported?*
- - “Fig. 10. Aviation model by” ([“TOSEM-2024-0332_Proof_hi”, p. 27](zotero://select/library/items/WC6HAAWF)) ([pdf](zotero://open-pdf/library/items/4YYDZM6X?page=28&annotation=WX87BZ8Q)) #ffd400
- *Same here, what the technology used to develop this?*
- - “VALIDATION” ([“TOSEM-2024-0332_Proof_hi”, p. 29](zotero://select/library/items/WC6HAAWF)) ([pdf](zotero://open-pdf/library/items/4YYDZM6X?page=30&annotation=VTWHAGN7)) #ffd400
- *The validation should be about the provided framework, which needs to be discussed with respect to precise requirements. Expressiveness or domain independence are example of possible requirements. Then the validation is missing a direct comparison with existing technologies, like matadepth.*
- ### [[Comments]]
- ## Summary
- The paper presents "Intino", a novel model-based approach to support the development of Domain-Specific Modeling Languages (DSMLs). The method integrates three levels of abstraction within a domain into a single model instead of having three specifications. This integration is achieved through a stepwise modeling process using three distinct but interconnected DSMLs. The outcome is a model that represents the domain across multiple levels of abstraction. The paper also introduces the concept of a "mother language," a domain-agnostic language providing a common syntax across all DSMLs, which can be extended to create new DSMLs.
- ## Detailed Comments
- The paper is about an interesting topic, even though several issues require a major revision of the document, as detailed below.
- Page 1: The initial discussion on domain modeling using automatically generated DSMLs needs further elaboration. It would benefit from a clearer explanation of how automatic generation is achieved and its implications for domain modeling.
- Page 1: The introduction of "Intino" is somewhat generic. It would be helpful to provide a brief context or background to prepare the reader for the introduction of this new methodology.
- Page 2: The rationale behind having three separate models needs clarification. The paper should explain why three models are necessary and what each represents within the context of the domain. Additionally, the explanation for the necessity of three different DSMLs needs to be clarified. The paper should provide a more substantial justification for this choice and detail how it benefits the modeling process. My comment is justified by the existence of existing approaches that have been neglected by the authors, and that should be presented to motivate the work.
- Page 4: The definitions provided for model, metamodel, and metametamodel are helpful, but the paper would benefit from concrete examples to illustrate these concepts.
- Page 5: Reference [40] should be included to provide a more comprehensive discussion of related approaches, such as metadepth. See comment below on the evaluation
- Page 6: The motivation behind Intino's fundamental features needs to be clearly articulated. The paper should explain why a single unified model integrating three levels of abstraction is necessary and how it improves upon traditional MDE practices.
- Page 7: The relevance of having a single domain model versus multiple models at different layers needs further explanation. The paper should discuss how this approach compares to multimodeling and the advantages it offers.
- Page 7: The terminology used in the paper is not always aligned with standard MDE terminology. It is suggested to either use more widely accepted terms or clearly define and justify the new terms used.
- Page 8: The concept of transforming a mother language to create a DSML needs illustrative examples. Providing concrete instances would make the explanation clearer.
- Page 8: The flexibility introduced by allowing different DSMLs to represent various mograms needs justification. The paper should explain why this complexity is necessary and what benefits it brings to the modeling process.
- Page 9: The role of the mother language in the stepwise modeling process should be explicitly stated. The paper needs to clarify how the mother language fits into this process and its importance.
- Page 9: The claim of automatic DSML generation would be strengthened by providing running examples. These examples would help readers understand the different steps and appreciate the practicality of the proposed process.
- Page 10: The discussion on the potential use of universal analytical constructs is vague. The paper should provide specific examples or case studies to illustrate this potential.
- Page 12: Until section 5 included, the paper is not easy to follow. It's important to provide concrete examples or links to existing technologies like EMF in order to make the discussion more tangible and help the reader understand the concepts introduced in an abstract manner.
- Page 20: Why is it necessary to distinguish between lexicon and syntax when it is possible to present the EBNF specification of the language? Sections 6.1 and 6.2 could be merged, in my opinion.
- Page 23: Section 7 is also affected by the same issues that are affecting the other sections. It is presented at a high level, is verbose, and difficult to follow. It lacks specific examples and cases for discussion.
- Fig. 8: It appears that the process depicted in Fig. 8 has not been fully implemented. This is a significant limitation that should be addressed upfront. However, it is important to discuss convincingly that despite this missing implementation, the approach can still be used, validated, or compared with existing approaches.
- Fig. 9: What technology was used to develop the language shown in Fig. 9? How are the types used (i.e., Manufacturer, Airline, etc., as defined in Fig. 10) imported? The same comment applies to the content of Fig. 10.
- Page 29: The validation section is another area that needs improvement in the work. Section 9 should focus on the provided framework and discuss it in relation to specific requirements, such as expressiveness or domain independence. It is important to define research questions that also involve a direct comparison with existing technologies, such as metadepth.
- ### [[REVIEWS/Notes]]
- ### YELLOW CONCERNS
background-color:: yellow
- {{query (and [[ffd400]] [[TOSEM-2024-0332_Proof_hi]] )}}
collapsed:: true
- ### ❓️Questions
- {{query (and [[question]] [[TOSEM-2024-0332_Proof_hi]] )[[question]]}}
query-table:: true
query-properties:: [:block]