logseq/pages/PAPERS___SOSYM-METAEDIT-EVOLUTION.md

type:: JournalPaper date:: 16-11-2023 - 15:54 full-title:: external-links:: todoist:: Todoist year:: 2023 date-start:: 16-11-2023 status:: DISMISSED , venue:: SOSYM priority::
leader:: progress:: {{renderer : todomaster}}

- ## MEETINGS
- ## READINGS
- ## TASKs
- ## QUESTIONS TO ANSWER
  {{query (and [[question]][[PAPERS/SOSYM-METAEDIT-EVOLUTION]])}}
- ## [[PAPERS/NOTES]]
	- ![How_MetaEdit_Supports_Co-Evolution_of_Modeling_Languages_Tools_and_Models.pdf](../assets/How_MetaEdit_Supports_Co-Evolution_of_Modeling_Languages_Tools_and_Models_1700146548743_0.pdf)
	- ![Evaluating_Tool_Support_for_Co-Evolution_of_Modeling_Languages_Tools_and_Models.pdf](../assets/Evaluating_Tool_Support_for_Co-Evolution_of_Modeling_Languages_Tools_and_Models_1700146564974_0.pdf)
	- [Evaluating_Tool_Support_for_Co-Evolution_of_Modeling_Languages_Tools_and_Models4SoSyMdraft.docx](../assets/Evaluating_Tool_Support_for_Co-Evolution_of_Modeling_Languages_Tools_and_Models4SoSyMdraft_1700146713360_0.docx)
		- **What we understood is that**
			- 1. We start with a formulation of the state machine metamodel in EMF and then in JJODEL
			- 2. We provide the metamodel with a concrete syntax on both platforms
			- 3. We apply the 12 refactoring on your demo paper on page 2 to both the metamodel and the concrete syntax
			- 4. Discuss each operation individually and try to come up with two tables, like table II in your demo paper on page 5, reporting the scores of EMF/Sirius and JJODEL
			- *Details from Steve*
				- To let others confirm the **results and scores (Table III** in the current paper), it would be useful to upload the initial state and the state after each scenario to GitHub or similar. We went through the 12 scenarios in turn from the start, so the GitHub timestamps also recorded how long each scenario took (Table IV in the current paper).
				- In this kind of evaluation, it’s common to write something like “one author performed the changes, and the other author checked them”. Maybe we could be the **checkers for each others’ work and evaluation** – either using the GitHub record or by demoing the changes to each other?
				- The **scalability** in terms of model size is also interesting (we simply duplicated the diagram content to make it 8 times larger, and then duplicated that diagram to give 72 diagrams – see text after Fig. 2 in the current paper, and the results on page 12). I think that stage can be left to last.
		- ![xtext-sirius-userguide.pdf](../assets/xtext-sirius-userguide_1700213627198_0.pdf)
			- State machine implementation with Sirius