Download as PDF, PPTX
Uploaded byMarkus Voelter
Fusing Modeling and Programming into Language-Oriented Programming
The document discusses the integration of modeling and programming through the use of language-oriented programming, emphasizing the importance of model-driven approaches in various fields like software engineering and climate science. It highlights the role of tools like the JetBrains MPS language workbench in facilitating notational freedom, modularity, and complexity management, as well as the differences between traditional programming and model-driven techniques. Key takeaways include the benefits of automated execution and the necessity of aligning semantics to enhance collaboration and reduce coding efforts.
More Related Content
Similar to Fusing Modeling and Programming into Language-Oriented Programming
Fusing Modeling and Programming into Language-Oriented Programming
- 1. Fusing Modeling andProgramming Our Experiences with MPS ... into Language-Oriented Programming Markus Völter voelter@acm.org www.voelter.de @markusvoelter
- 2. Industry Theory ≤ Pragmatics WhileI value and conceptual purity... I value such as tool support & scalability more.
- 3.
- 4.
- 5. M1 M2 M3 L1L2 L3 def := … def := ∃ ⊢ #$α def := … .... LM ∃, ⊢, #, $, α <meta> <meta> <meta> <meta> <meta> <meta> LANGUAGEDEFINITION Language Workbench
- 6.
- 7. Reality (Software System) Model MODELKINDS describe understand explain analyze predict prescribe derive manufacture Modeling Model- Driven Climate Science ParticlePhysics Social Science FEM Models Manufacturing Construction Software Eng Automated
- 8.
- 9.
- 10. Model MODELDRIVENSE Language <abstracts> <meta> What-ness Program Programming Language Programming isjust model-driving using a programming language. And caring about particular concerns. P ⊂ M
- 11. MODELDRIVENSE Reality Model Language Execution Model Analysis Aligning the Semanticsof the two and is key.
- 12. MODELPURPOSE Reality Model Language Execution Model Analysis … Review Inspect Analyze Simulate … Diverse reasons whyyou would want to model beyond reducing the coding effort by generating the implementation from more concise models.
- 13. MODELPURPOSE Reality Model Language Execution Model Analysis … Review Inspect Analyze Simulate … Automated Automated Execution is essentialto ”connect” the models to the actual system. But:
- 14.
- 15. Programming vs. MD:What it can be
- 17.
- 19. Data changes. The TypeSystem supports temporal data natively.
- 21. Law changes overthe years. The payroll calculations have to follow.
- 22. Visualizations. Of relationships betweensubdomains.
- 29.
- 33.
- 34. MPS Language Workbench LanguageWorkbench Open Source, by Jetbrains Very Expressive Used for years in industry Vast Experience
- 35.
- 36. MPS: Notational Freedom 100%crucial for acceptance in domain! Rabbit Hole Modeling is more than diagrams!
- 37. Handling Complexity: LanguageModularity New Language GPL Extension Existing Domain Notation (Informal) Formalization Formalized Language Reuse GPL incl. Expressions and TS Add/Embed DS-extensions Compatible notational style Reduce to GPL Analyze Domain to find Abstractions Define suitable, new notations Rely on existing behavioral paradigm Reuse standard expression language Interpret/Generate to one or more GPLs Use existing notation from domain Clean up and formalize Generate/Interpret Often import existing „models“
- 38. Handling Complexity: GrowingLanguages New Language GPL Extension Existing Domain Notation (Informal) Formalization Formalized Language Reuse GPL incl. Expressions and TS Add/Embed DS-extensions Compatible notational style Reduce to GPL Analyze Domain to find Abstractions Define suitable, new notations Rely on existing behavioral paradigm Reuse standard expression language Interpret/Generate to one or more GPLs Use existing notation from domain Clean up and formalize Generate/Interpret Often import existing „models“
- 39. 35+ Modular CExtensions
- 40. Handling Complexity: GrowingLanguages New Language GPL Extension Existing Domain Notation (Informal) Formalization Formalized Language Reuse GPL incl. Expressions and TS Add/Embed DS-extensions Compatible notational style Reduce to GPL Analyze Domain to find Abstractions Define suitable, new notations Rely on existing behavioral paradigm Reuse standard expression language Interpret/Generate to one or more GPLs Use existing notation from domain Clean up and formalize Generate/Interpret Often import existing „models“
- 41. Partial Embedding ofKernelF Restriction Modular Extension Composition (Sep. of Concerns)
- 43. Language Workbench -Convecton Browser · Cloud · Collaborative Multi-Notation · Language Modularity
- 44. Language Workbench -Convecton
- 45. Language Workbench -Convecton
- 47. TU Delft itemis/Typefox CWI Amsterdam Solmi/Persiani Rascal TheWhole Platform Other Language Workbenches Metacase RWTH Aachen
- 48.
- 49. Language Great IDE Analyses Refactorings Testing Debuggers Abstractions Notations Syntax Coloring CodeCompletion Goto Definition Collaboration Migration Relevant Good Errors Type System! Aligned with Processes Write Tests Run them in IDE Report Back Animate Execution Simulators GOOD GREAT More than Language
- 50.
- 52.
- 53. Specification: KernelF Analysis: SMTSolving of Expressions
- 54. Specification: Medical Project Test:Run Tests directly in the IDE
- 55. Specification: NuSMV extensions(Dan Ratiu)
- 56.
- 57.
- 58. Everything on the sameplatform! No Tool Interop Problems! Common, Good ( ) IDE support. Properties from Mod‘ing + Prog‘ing . Language Composition & Reuse. and mostly free
- 59. ... is whatI call a Model { {Mogram Prodel Everything on the MPS platform!