SlideShare a Scribd company logo

Adaptable Model Versioning using Model Transformation By Demonstration

Philip Langer, profile picture
Philip Langer

This document summarizes the contributions of the author's PhD dissertation on adaptable model versioning using model transformation by demonstration. The contributions include: (1) an adaptable model versioning framework, (2) a method of specifying composite model operations called "model transformation by demonstration", (3) an operation detection process, and (4) an approach to conflict and warning detection. The operation detection process involves matching models, detecting atomic and composite operations, and deriving operation specifications. The conflict detection approach considers both operation-based and state-based differences between models.

Technology
1 of 58
Downloaded 29 times
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
Adaptable Model Versioning using Model Transformation By Demonstration PhD Defense Vienna, 21st of December 2011 Philip Langer Business Informatics Group Institute of Software Technology and Interactive Systems Vienna University of Technology Favoritenstraße 9-11/188-3, 1040 Vienna, Austria phone: +43 (1) 58801-18804 (secretary), fax: +43 (1) 58801-18896 office@big.tuwien.ac.at, www.big.tuwien.ac.at
Introduction Context Adaptable Model Versioning 2
Introduction Adaptable Model Versioning Managing the evolution of software artifacts performed by a distributed team. Vo Vr1 Merge Process Developer 1 Check Out Check In Repository Vm Check Out Check In Developer 2 Vo Vr2 t0 t1 t2 t3 3
Introduction Adaptable Model Versioning Software models in the context of model-driven engineering. Vo Vr1 Merge Process Developer 1 Check Out Check In Repository Vm Check Out Check In Developer 2 Vo Vr2 t0 t1 t2 t3 4
Introduction Adaptable Model Versioning Impedance Mismatch 5
Introduction Adaptable Model Versioning  Deficiencies of existing model versioning systems  Modeling editor/language-specific vs. imprecise versioning  Tightly bound to a specific modeling editor or modeling language  Generic and unaware of the specifics of modeling languages  Specification of composite operations (e.g., refactorings)  Currently very challenging task  Requires specific knowledge which untrained users don’t have  Absence of information on applied composite operations  Operation recording  inherently puts restrictions on the modeling editor  Model differencing  no support for detecting composite operations  Insufficient conflict detection  Composite operations are not considered  Important types of conflicts and warnings are neglected P. Brosch, G. Kappel, P. Langer, M. Seidl, K. Wieland, and M. Wimmer. “The Past, Present, and Future of Model 6 Versioning”, in Emerging Technologies for the Evolution and Maintenance of Software Models. IGI Global, 2011.
Introduction Adaptable Model Versioning  Generic but adaptable model versioning  Supporting all EMF*-based modeling languages  Adaptation points to incorporate language specifics  Advantages of generic and language-specific versioning systems  AMOR project  Funded by FFG FIT-IT Semantic Systems  Gerti Kappel, Martina Seidl, Manuel Wimmer, Petra Brosch, Horst Kargl, Philip Langer, Werner Retschitzegger, Wieland Schwinger, and Konrad Wieland * Eclipse Modeling Framework (EMF) 7
Introduction Adaptable Model Versioning  AMOR merge process Vr1 Conflict Detection Conflict Detection Conflict-tolerant Operation-based State-based Resolution Vm Operation Detection Vo Merge Vr2 P. Brosch, G. Kappel, M. Seidl, K. Wieland, M. Wimmer, H. Kargl, and P. Langer. “Adaptable Model Versioning in 8 Action”. In Modellierung’2010, volume 161 of LNI, pages 221–236. GI, 2010.
Introduction Vo Vr2 Vr1  Focus of this dissertation Operation Detection Operation-based Conflict Detection Conflict-tolerant Merge State-based Conflict Detection Resolution Adaptable Model Versioning Vm 9
Motivating Examples Violated Preconditions of Composite Operations Add state “Active”, Move states Introduce into “Active”, Composite fold transition State “hangup” dial Rename Rename “hangup” “hangup” 10
Motivating Examples Violated Preconditions of Composite Operations Introduce Composite State The preconditions of the refactoring “Introduce Composite State” are violated! dial Rename “hangup” 11
Contributions at a Glance Adaptable Model Versioning using Model Transformation By Demonstration Adaptation Artifacts Match Operation Signifier Validation Rules Specifications Specifications Rules C1 Adaptable Model Versioning Framework C2 C4 C2 Composite Operation Specification “Model Transformation By Demonstration” Vr1 Conflict Detection Conflict Detection Conflict-tolerant Operation-based C3 Operation Detection State-based Resolution Operation Vm Detection Vo Merge C4 Conflict and Warning Detection C3 C4 C4 Vr2 C1 Adaptable Model Versioning Framework Existing technology has been reused. 12
C2 Model Transformation By Demonstration  Composite operation specification 1 Initial 2 Working Create initial model Copy initial model Modeling model model Revised 3 model Perform updates 5 4 State-based Imply conditions Configuration & Generation comparison Conditions [implied] Diff model 6 Edit conditions Conditions 7 Generate Operation Operation [revised] Specification Specification Legend automatic manual P. Brosch, P. Langer, M. Seidl, K. Wieland, M. Wimmer, G. Kappel, W. Retschitzegger, and W. Schwinger. “An Example 13 Is Worth a Thousand Words: Composite Operation Modeling By-Example”. In Proceedings of the International Conference on Model Driven Engineering Languages and Systems (MoDELS’09), LNCS. Springer, 2009.
C2 Model Transformation By Demonstration  Composite operation specification 1 Initial 2 Working Create initial model Copy initial model Modeling model model Revised 3 model Perform updates 5 4 State-based Imply conditions Configuration & Generation comparison Conditions [implied] Diff model 6 Edit conditions Conditions 7 Generate Operation Operation [revised] Specification Specification Legend automatic manual 14
C2 Model Transformation By Demonstration  Composite operation specification 1 Initial 2 Working Create initial model Copy initial model Modeling model model Revised 3 model Perform updates 5 4 State-based Imply conditions Configuration & Generation comparison Conditions [implied] Diff model 6 Edit conditions Conditions 7 Generate Operation Operation [revised] Specification Specification Legend automatic manual 15
C2 Model Transformation By Demonstration  Composite operation specification 1 Initial 2 Working Create initial model Copy initial model Modeling model model Revised 3 model Perform updates 5 4 State-based Imply conditions Configuration & Generation comparison Conditions [implied] Diff model 6 Edit conditions Conditions 7 Generate Operation Operation [revised] Specification Specification Legend automatic manual 16
C2 Model Transformation By Demonstration  Composite operation specification 1 Initial 2 Working Create initial model Copy initial model Modeling model model Revised 3 model Perform updates 5 4 State-based Imply conditions Configuration & Generation comparison Conditions [implied] Diff model 6 Edit conditions Conditions 7 Generate Operation Operation [revised] Specification Specification Legend automatic manual 17
C2 Model Transformation By Demonstration  Composite operation specification 1 Initial 2 Working Create initial model Copy initial model Modeling model model Revised 3 model Perform updates 5 4 State-based Imply conditions Configuration & Generation comparison Conditions [implied] Diff model 6 Edit conditions Conditions 7 Generate Operation Operation [revised] Specification Specification Legend automatic manual 18
C2 Model Transformation By Demonstration  Composite operation specification 1 Initial 2 Working Create initial model Copy initial model Modeling model model Revised 3 model Perform updates 5 4 State-based Imply conditions Configuration & Generation comparison Conditions [implied] Diff model 6 Edit conditions Conditions 7 Generate Operation Operation [revised] Specification Specification Legend automatic manual 19
C2 Model Transformation By Demonstration  Besides being executable… Adaptation Artifacts Match Operation Signifier Validation Rules Specifications Specifications Rules C2 C4 Vr1 Conflict Detection Conflict Detection Conflict-tolerant Operation-based State-based Resolution Operation Vm Detection Vo Merge C3 C4 C4 Vr2 Adaptable Model Versioning Framework 20
C3 Operation Detection  Operation detection process 1. UUID-based matching 2. Rule-based matching 3. Atomic operation detection 4. Composite operation detection 21
C3 Operation Detection  Operation detection process 1. UUID-based matching 2. Rule-based matching 3. Atomic operation detection 4. Composite operation detection Vo Vr1 lift lift Idle Idle DialTone Active hangup hangup DialTone dial dial Dialing hangup Dialing 22
C3 Operation Detection  Operation detection process 1. UUID-based matching 2. Rule-based matching 3. Atomic operation detection 4. Composite operation detection Vo Vr1 lift lift Idle Idle DialTone Active hangup hangup DialTone dial dial Dialing hangup Dialing 23
C3 Operation Detection  Operation detection process 1. UUID-based matching 2. Rule-based matching 3. Atomic operation detection 4. Composite operation detection Vo Vr1 lift lift Idle Idle DialTone Active hangup hangup DialTone dial dial Dialing hangup Dialing 24
C3 Operation Detection  Operation detection process Diff 1. UUID-based matching Model 2. Rule-based matching 3. Atomic operation detection 4. Composite operation detection No difference Vo Vr1 lift lift Idle Idle DialTone Active hangup hangup DialTone dial dial Dialing hangup Dialing 25
C3 Operation Detection  Operation detection process Diff 1. UUID-based matching Model ! 2. Rule-based matching 3. Atomic operation detection 4. Composite operation detection ! Transition target changed! Vo Vr1 lift lift Idle Idle DialTone Active hangup hangup DialTone dial dial Dialing hangup Dialing 26
C3 Operation Detection  Operation detection process Diff 1. UUID-based matching Model !! 2. Rule-based matching 3. Atomic operation detection 4. Composite operation detection ! Transition source changed! Vo Vr1 lift lift Idle Idle DialTone Active hangup hangup DialTone dial dial Dialing hangup Dialing 27
C3 Operation Detection  Operation detection process Diff 1. UUID-based matching Model !! + 2. Rule-based matching 3. Atomic operation detection 4. Composite operation detection + State Active has been added! Vo Vr1 lift lift Idle Idle DialTone Active hangup hangup DialTone dial dial Dialing hangup Dialing 28
C3 Operation Detection  Operation detection process Diff 1. UUID-based matching Model !! + + + 2. Rule-based matching 3. Atomic operation detection 4. Composite operation detection Vo Vr1 lift lift Idle Idle DialTone Active hangup hangup DialTone dial dial Dialing hangup Dialing 29
C3 Operation Detection  Operation detection process Diff 1. UUID-based matching Model !! + - + + 2. Rule-based matching 3. Atomic operation detection 4. Composite operation detection - Transition hangup has been deleted! Vo Vr1 lift lift Idle Idle DialTone Active hangup hangup DialTone dial dial Dialing hangup Dialing 30
C3 Operation Detection  Composite operation detection process Diff Input Operation Operation Operation Diff Model Model Specification Specification Operation Specification Specifications !! + - ++ «foreach» «foreach» Diff Model Preprocessing Derive Precondition Derive Postcondition Binding Binding Input Operation Operation Operation Signature Specification Specification Signature Precondition Postcondition Binding Binding Preselection Evaluate Binding Evaluate Binding [no diff match] [diff matches] [invalid] [valid] [invalid] [valid] Valid Operation Operation Potential Operation Precondition Operation Operation Specification Specification Occurrence Operation Specification Specification Binding Occurrence 1 2 3 Refactoring Diff Refactoring Operation Occurrence Occurrence Occurrence Model P. Langer, K. Wieland, and P. Brosch. Specification, Execution, and Detection of !! + 31 + - + Refactorings for Software Models. In Proceedings of the WiP Session @ PPPJ’10, 2010.
C4 Conflict Detection Input Adaptation Artifacts Diff Match Operation Signifier Validation Model1 Rules Specifications Specifications Rules C4 !! + - + + Vr1 Conflict Detection Conflict Detection Diff Conflict-tolerant Operation-based State-based Resolution Operation Vm Detection Model2 Vo Merge ! + + C4 C4 Vr2 Adaptable Model Versioning Framework 32
C4 Conflict Detection  Operation-based conflict detection 1. Atomic operation conflict detection 2. Composite operation conflict detection 3. Signifier warning detection 33
C4 Conflict Detection  Operation-based conflict detection 1. Atomic operation conflict detection 2. Composite operation conflict detection 3. Signifier warning detection 34
C4 Conflict Detection  Atomic operation conflict detection  Nine conflict patterns for EMF-based models  Delete-use conflict du : DeleteUse delete use do : DeleteObject o : Object fo : FeatureOperation object value {self.oclIsTypeOf(InsertFeatureValue) or self.oclIsTypeOf(FeatureUpdate)}  Update-update conflict uu : UpdateUpdate update1 affected feature Object fo1 : FeatureOperation f : Feature o : Object {upperBound = 1 and fo1.value <> fo2.value} fo2 : FeatureOperation feature affected Object update2  … G. Taentzer, C. Ermel, P. Langer, and M. Wimmer. “A Fundamental Approach to Model Versioning Based on Graph Modifications”. Submitted to Software and System Modeling. Springer, 2011. 35
C4 Conflict Detection  Operation-based conflict detection 1. Atomic operation conflict detection 2. Composite operation conflict detection 3. Signifier warning detection 36
C4 Conflict Detection  Composite operation conflict detection For each composite operation application, check whether it is applicable at the opposite side Vo Diff Model1 !! + - + + Vr1 Vr2 !! + + + 37
C4 Conflict Detection  Composite operation conflict detection  Revisiting the motivating example Vr1 lift Idle Active Vo Introduce hangup Composite DialTone lift Idle State dial DialTone dial hangup Dialing dial Dialing hangup Vr2 lift Idle DialTone abort dial dial Dialing hangup 38
C4 Conflict Detection  Composite operation conflict detection Diff  Revisiting the motivating example Vr1 Model1 lift Idle !! + - + + Active Vo Introduce hangup Original Composite DialTone match Idle lift State dial SingleState_0 DialTone dial Transition_0 hangup Dialing SingleState_1 dial Transition_1 Precondition of Dialing SingleState_2 hangup composite Transition_2 Vr2 operation is lift violated! Idle SingleState_0 DialTone Transition_0 abort SingleState_1 dial Transition_1.name = Transition_2.name dial Transition_1 SingleState_2 Dialing hangup Transition_2 39
C4 Conflict Detection  Composite operation warning detection For each composite operation application, check whether its match size increased at the opposite side Vo Diff Model1 !! + - + + Vr1 Vr2 !! + + + 40
C4 Conflict Detection  Composite operation warning detection Add state “Active”, Move states into “Active”, Introduce fold transition Composite “hangup” State dial Add state Add state “Connecting” “Connecting” 41
C4 Conflict Detection Your composite  Composite operation warning detection operation should incorporate opposite operations! Introduce Introduce Composite Composite State State dial Add state Add state “Connecting” “Connecting” 42
C4 Conflict Detection  Composite operation warning detection Vr1 lift Idle Active Vo Introduce hangup Composite DialTone lift Idle State dial DialTone dial hangup Dialing dial Dialing hangup Vr2 lift Idle DialTone hangup dial dial Dialing hangup end_dialing 43 hangup Connecting
C4 Conflict Detection  Composite operation warning detection Diff Vr1 Model1 lift Idle !! + - + + Active Vo Introduce hangup Original Composite DialTone match Idle lift State dial SingleState_0 DialTone dial Transition_0 hangup Dialing SingleState_1 dial Transition_1 Composite Dialing SingleState_2 hangup operation Transition_2 Vr2 is applicable! lift Idle SingleState_0 DialTone Transition_0 hangup SingleState_1 dial dial Transition_1 SingleState_2 Dialing hangup Transition_2 end_dialing 44 hangup Connecting
C4 Conflict Detection  Composite operation warning detection Diff Vr1 Model1 lift Idle !! + - + + Active Vo Introduce hangup Original Composite DialTone match Idle lift State dial SingleState_0 DialTone dial Transition_0 hangup Dialing SingleState_1 dial Transition_1 Your composite SingleState_2 hangup Dialing operation should Vr2 include the state Transition_2 “Connecting”! lift Idle SingleState_0 DialTone Transition_0 hangup SingleState_1 dial dial Transition_1 SingleState_2 Dialing hangup Transition_2 end_dialing 45 hangup Connecting
C4 Conflict Detection  Operation-based conflict detection 1. Atomic operation conflict detection 2. Composite operation conflict detection 3. Signifier warning detection 46
C4 Conflict Detection  Signifier warning detection Change target Change target of r1 of r1 Change source Change source of r1 of r1 47
C4 Conflict Detection  Signifier warning detection The superior meaning of r1 has been con- currently changed! Change target Redefinition of r1 of r1 Change source Redefinition of r1 of r1 48
C4 Conflict Detection  Signifier warning detection  Signifier  Combination of properties of a model element type  Convey the “superior meaning” of a model element  e.g., the meaning of a reference is mainly conveyed by its name, source, and target  Signifier specifications (using ECL*) rule ReferenceSignifier match l : Left!EReference with r : Right!EReference { compare : l.name = r.name and l.source.isMatched(r.source) and l.target.isMatched(r.target) } * Epsilon Comparison Language (ECL), www.eclipse.org/gmt/epsilon 49
C4 Conflict Detection  Signifier warning detection rule ReferenceSignifier match l : Left!EReference with r : Right!EReference { compare : l.name = r.name and l.source.isMatched(r.source) and l.target.isMatched(r.target) } Vr1 Match Trigger Vo Vr2 Match Trigger 50
C4 Conflict Detection  Signifier warning detection rule ReferenceSignifier match l : Left!EReference with r : Right!EReference { compare : l.name = r.name and l.source.isMatched(r.source) and l.target.isMatched(r.target) } Vr1 Match Trigger Vo Concurrent Compare signifier of r1::Vo with r1::Vr1 signifier change signifier change of reference “r1”! Compare signifier of r1::Vo with r1::Vr2  signifier change You might want to duplicate “r1”? Vr2 Match Trigger 51
C4 Conflict Detection Adaptation Artifacts Match Operation Signifier Validation Rules Specifications Specifications Rules C4 Vr1 Conflict Detection Conflict Detection Conflict-tolerant Operation-based State-based Resolution Operation Vm Vo Detection Merge C4 Vr2 Adaptable Model Versioning Framework cf. PhD thesis by Konrad Wieland cf. PhD thesis by Petra Brosch 52
Evaluation  Adaptable Model Versioning Framework C1  Model Transformation By Demonstration C2  Operation Detection C3  Conflict Detection C4 53
Evaluation  Model Transformation By Demonstration (MTBD) C2  Overall research question  Is it easier to use our MTBD tool than EMF Tiger*?  Study design  57 students (who have no experience in model transformation)  Specification of two composite operations with both tools  Questionnaire  Results  88 % preferred using our MTBD tool  74 % indicated that our MTBD tool is easy to use  Students needed nearly twice as much time using EMF Tiger  However, 53 % stated that configuring the conditions is hard * EMF Tiger: EMF-based graph transformation tool, user.cs.tu-berlin.de/~emftrans 54
Evaluation  Operation Detection C3  Overall research question  How accurate is our tool for detecting applications of composite operations?  Study design  Applying our tool to the evolution of GMF*  141 composite operation applications  342 atomic operation applications  Results  98 % precision  “How many detected composite operations have actually been correct?”  70 % recall  “How many applied composite operations have been correctly detected?”  Recall decreases with a growing number of concurrent composite operations  Composite operation sequences * Graphical Modeling Framework (GMF), www.eclipse.org/modeling/gmf 55
Evaluation  Conflict Detection C4  Overall research questions  How accurate is our tool for detecting atomic conflicts?  Can the same accuracy be achieved as operation tracking approaches?  Study design  Automatically executable benchmark  23 carefully selected versioning scenarios  12 of these scenarios include conflicting atomic operations  Benchmark applied to our tool, EMF Compare*, and EMFStore**  Results  Our tool: 100 % precision and recall  EMF Compare: 67 % precision and 17 % recall  EMFStore: 100 % precision and recall * EMF Compare, http://www.eclipse.org/emf/compare ** EMFStore, www.eclipse.org/emfstore/ 56
Limitations and Future Work  Adaptable Model Versioning Framework C1  Are the provided adaptation points adequate and sufficient?  Empirical case study required!  Model Transformation By Demonstration C2  Configuring conditions still very challenging  Further automation of the specification process  Multiple demonstrations  Operation Detection C3  Overlapping sequences of operations  Conflict Detection C4  Deriving the conflicting opposite operation that caused a (composite operation’s pre-)condition to be invalid  Semantic inconsistencies 57
Thank you for your attention!  Dissertation available  www.big.tuwien.ac.at/teaching/theses/70  Open source implementations  EMF Modeling Operations  www.modelversioning.org/emf-modeling-operations  AMOR Conflict Detection  code.google.com/a/eclipselabs.org/p/amor-conflict-detection 58

More Related Content

PDF
Sop - Packaging
Terry Koch
 
PPT
SDLC
Chandan Pandit
 
PPTX
Technology powerpoint
kbarbin
 
PPTX
Technology powerpoint
kbarbin
 
PPTX
Tech blog
kbarbin
 
PDF
Formacion investigativa-e-investigacion-formativa
marcocanales
 
PPTX
Njdgarcscmr
marcocanales
 
PDF
Process Mining - Chapter 1 - Introduction
Wil van der Aalst
 
Sop - Packaging
Terry Koch
 
SDLC
Chandan Pandit
 
Technology powerpoint
kbarbin
 
Technology powerpoint
kbarbin
 
Tech blog
kbarbin
 
Formacion investigativa-e-investigacion-formativa
marcocanales
 
Njdgarcscmr
marcocanales
 
Process Mining - Chapter 1 - Introduction
Wil van der Aalst
 

Similar to Adaptable Model Versioning using Model Transformation By Demonstration (20)

PDF
Process mining chapter_01_introduction
Muhammad Ajmal
 
PDF
Business Process Configuration in the Cloud: How to Support and Analyze Multi...
Wil van der Aalst
 
PDF
Postdoc Symposium - Abram Hindle
ICSM 2011
 
PDF
Iswim for testing
ClarkTony
 
PDF
Colex: A Web-based Collaborative Conflict Lexicon
Philip Langer
 
PDF
Deploying Functional Qualification at STMicroelectronics
DVClub
 
PDF
Benjamin q4 2008_bristol
Obsidian Software
 
PDF
Version control thesis
Waleed Mohamed
 
PDF
S-CUBE LP: Service Versioning, Compatibility and Evolution
virtual-campus
 
PDF
Discovering Concurrency: Learning (Business) Process Models from Examples
Wil van der Aalst
 
PDF
Semantic Enterprise Architecture
Michael zur Muehlen
 
PDF
Web2MexADL - CSMR Presentation
jccastrejon
 
PDF
ICSM08a.ppt
Ptidej Team
 
PDF
Aras Vision and Roadmap with Aras Innovator PLM Software
Aras
 
PDF
Data Migration In An Agile Open Source World
Craig Smith
 
PDF
M3 Modernization Case Study
ADC Austin Tech
 
PDF
Cardamom Omg
Emmanuel Fuchs
 
PDF
Cardamom Omg
Emmanuel Fuchs
 
PPTX
Lets focus on business value
Einar Ingebrigtsen
 
PPTX
Lets focus on business value
Einar Ingebrigtsen
 
Process mining chapter_01_introduction
Muhammad Ajmal
 
Business Process Configuration in the Cloud: How to Support and Analyze Multi...
Wil van der Aalst
 
Postdoc Symposium - Abram Hindle
ICSM 2011
 
Iswim for testing
ClarkTony
 
Colex: A Web-based Collaborative Conflict Lexicon
Philip Langer
 
Deploying Functional Qualification at STMicroelectronics
DVClub
 
Benjamin q4 2008_bristol
Obsidian Software
 
Version control thesis
Waleed Mohamed
 
S-CUBE LP: Service Versioning, Compatibility and Evolution
virtual-campus
 
Discovering Concurrency: Learning (Business) Process Models from Examples
Wil van der Aalst
 
Semantic Enterprise Architecture
Michael zur Muehlen
 
Web2MexADL - CSMR Presentation
jccastrejon
 
ICSM08a.ppt
Ptidej Team
 
Aras Vision and Roadmap with Aras Innovator PLM Software
Aras
 
Data Migration In An Agile Open Source World
Craig Smith
 
M3 Modernization Case Study
ADC Austin Tech
 
Cardamom Omg
Emmanuel Fuchs
 
Cardamom Omg
Emmanuel Fuchs
 
Lets focus on business value
Einar Ingebrigtsen
 
Lets focus on business value
Einar Ingebrigtsen
 
Ad

More from Philip Langer (6)

PDF
Tailor made model comparison: How to customize EMF Compare for your modeling ...
Philip Langer
 
PDF
What every Eclipse developer should know about EMF
Philip Langer
 
PDF
Play Framework: The Basics
Philip Langer
 
PPTX
A Brief Introduction to Working with Git
Philip Langer
 
PDF
You need to extend your models? EMF Facet vs. EMF Profiles
Philip Langer
 
PDF
From UML Profiles to EMF Profiles and Beyond (TOOLS'11)
Philip Langer
 
Tailor made model comparison: How to customize EMF Compare for your modeling ...
Philip Langer
 
What every Eclipse developer should know about EMF
Philip Langer
 
Play Framework: The Basics
Philip Langer
 
A Brief Introduction to Working with Git
Philip Langer
 
You need to extend your models? EMF Facet vs. EMF Profiles
Philip Langer
 
From UML Profiles to EMF Profiles and Beyond (TOOLS'11)
Philip Langer
 
Ad

Recently uploaded (20)

PPTX
MYSQL Presentation for SQL database connectivity
Swati270511
 
PPTX
Digital-Transformation-Roadmap-for-Companies.pptx
David Malick Dieng
 
PDF
Build a system with the filesystem maintained by OSTree @ COSCUP 2025
Jian-Hong Pan
 
PDF
Reach Out and Touch Someone: Haptics and Empathic Computing
Mark Billinghurst
 
PDF
Profit Center Accounting in SAP S/4HANA, S4F28 Col11
Libreria ERP
 
PDF
Blue Purple Modern Animated Computer Science Presentation.pdf.pdf
Akar16
 
PDF
Machine learning based COVID-19 study performance prediction
IAESIJAI
 
PDF
Network Security Unit 5.pdf for BCA BBA.
Serpent6
 
PDF
Empathic Computing: Creating Shared Understanding
Mark Billinghurst
 
PDF
Review of recent advances in non-invasive hemoglobin estimation
IAESIJAI
 
PDF
Approach and Philosophy of On baking technology
30anthuong17
 
PDF
NewMind AI Weekly Chronicles - August'25-Week II
NewMind AI
 
PPTX
ACSFv1EN-58255 AWS Academy Cloud Security Foundations.pptx
23bcla24
 
PDF
The Rise and Fall of 3GPP – Time for a Sabbatical?
3G4G
 
PDF
Unlocking AI with Model Context Protocol (MCP)
Brian McKeiver
 
PDF
MIND Revenue Release Quarter 2 2025 Press Release
MIND CTI
 
PPT
“AI and Expert System Decision Support & Business Intelligence Systems”
ZubinRadhakrishnan
 
PDF
Dropbox Q2 2025 Financial Results & Investor Presentation
Dropbox
 
PDF
TokAI - TikTok AI Agent : The First AI Application That Analyzes 10,000+ Vira...
SOFTTECHHUB
 
PDF
Electronic commerce courselecture one. Pdf
OmerMohamed64
 
MYSQL Presentation for SQL database connectivity
Swati270511
 
Digital-Transformation-Roadmap-for-Companies.pptx
David Malick Dieng
 
Build a system with the filesystem maintained by OSTree @ COSCUP 2025
Jian-Hong Pan
 
Reach Out and Touch Someone: Haptics and Empathic Computing
Mark Billinghurst
 
Profit Center Accounting in SAP S/4HANA, S4F28 Col11
Libreria ERP
 
Blue Purple Modern Animated Computer Science Presentation.pdf.pdf
Akar16
 
Machine learning based COVID-19 study performance prediction
IAESIJAI
 
Network Security Unit 5.pdf for BCA BBA.
Serpent6
 
Empathic Computing: Creating Shared Understanding
Mark Billinghurst
 
Review of recent advances in non-invasive hemoglobin estimation
IAESIJAI
 
Approach and Philosophy of On baking technology
30anthuong17
 
NewMind AI Weekly Chronicles - August'25-Week II
NewMind AI
 
ACSFv1EN-58255 AWS Academy Cloud Security Foundations.pptx
23bcla24
 
The Rise and Fall of 3GPP – Time for a Sabbatical?
3G4G
 
Unlocking AI with Model Context Protocol (MCP)
Brian McKeiver
 
MIND Revenue Release Quarter 2 2025 Press Release
MIND CTI
 
“AI and Expert System Decision Support & Business Intelligence Systems”
ZubinRadhakrishnan
 
Dropbox Q2 2025 Financial Results & Investor Presentation
Dropbox
 
TokAI - TikTok AI Agent : The First AI Application That Analyzes 10,000+ Vira...
SOFTTECHHUB
 
Electronic commerce courselecture one. Pdf
OmerMohamed64
 

Adaptable Model Versioning using Model Transformation By Demonstration

  • 1. Adaptable Model Versioning using Model Transformation By Demonstration PhD Defense Vienna, 21st of December 2011 Philip Langer Business Informatics Group Institute of Software Technology and Interactive Systems Vienna University of Technology Favoritenstraße 9-11/188-3, 1040 Vienna, Austria phone: +43 (1) 58801-18804 (secretary), fax: +43 (1) 58801-18896 office@big.tuwien.ac.at, www.big.tuwien.ac.at
  • 2. Introduction Context Adaptable Model Versioning 2
  • 3. Introduction Adaptable Model Versioning Managing the evolution of software artifacts performed by a distributed team. Vo Vr1 Merge Process Developer 1 Check Out Check In Repository Vm Check Out Check In Developer 2 Vo Vr2 t0 t1 t2 t3 3
  • 4. Introduction Adaptable Model Versioning Software models in the context of model-driven engineering. Vo Vr1 Merge Process Developer 1 Check Out Check In Repository Vm Check Out Check In Developer 2 Vo Vr2 t0 t1 t2 t3 4
  • 5. Introduction Adaptable Model Versioning Impedance Mismatch 5
  • 6. Introduction Adaptable Model Versioning  Deficiencies of existing model versioning systems  Modeling editor/language-specific vs. imprecise versioning  Tightly bound to a specific modeling editor or modeling language  Generic and unaware of the specifics of modeling languages  Specification of composite operations (e.g., refactorings)  Currently very challenging task  Requires specific knowledge which untrained users don’t have  Absence of information on applied composite operations  Operation recording  inherently puts restrictions on the modeling editor  Model differencing  no support for detecting composite operations  Insufficient conflict detection  Composite operations are not considered  Important types of conflicts and warnings are neglected P. Brosch, G. Kappel, P. Langer, M. Seidl, K. Wieland, and M. Wimmer. “The Past, Present, and Future of Model 6 Versioning”, in Emerging Technologies for the Evolution and Maintenance of Software Models. IGI Global, 2011.
  • 7. Introduction Adaptable Model Versioning  Generic but adaptable model versioning  Supporting all EMF*-based modeling languages  Adaptation points to incorporate language specifics  Advantages of generic and language-specific versioning systems  AMOR project  Funded by FFG FIT-IT Semantic Systems  Gerti Kappel, Martina Seidl, Manuel Wimmer, Petra Brosch, Horst Kargl, Philip Langer, Werner Retschitzegger, Wieland Schwinger, and Konrad Wieland * Eclipse Modeling Framework (EMF) 7
  • 8. Introduction Adaptable Model Versioning  AMOR merge process Vr1 Conflict Detection Conflict Detection Conflict-tolerant Operation-based State-based Resolution Vm Operation Detection Vo Merge Vr2 P. Brosch, G. Kappel, M. Seidl, K. Wieland, M. Wimmer, H. Kargl, and P. Langer. “Adaptable Model Versioning in 8 Action”. In Modellierung’2010, volume 161 of LNI, pages 221–236. GI, 2010.
  • 9. Introduction Vo Vr2 Vr1  Focus of this dissertation Operation Detection Operation-based Conflict Detection Conflict-tolerant Merge State-based Conflict Detection Resolution Adaptable Model Versioning Vm 9
  • 10. Motivating Examples Violated Preconditions of Composite Operations Add state “Active”, Move states Introduce into “Active”, Composite fold transition State “hangup” dial Rename Rename “hangup” “hangup” 10
  • 11. Motivating Examples Violated Preconditions of Composite Operations Introduce Composite State The preconditions of the refactoring “Introduce Composite State” are violated! dial Rename “hangup” 11
  • 12. Contributions at a Glance Adaptable Model Versioning using Model Transformation By Demonstration Adaptation Artifacts Match Operation Signifier Validation Rules Specifications Specifications Rules C1 Adaptable Model Versioning Framework C2 C4 C2 Composite Operation Specification “Model Transformation By Demonstration” Vr1 Conflict Detection Conflict Detection Conflict-tolerant Operation-based C3 Operation Detection State-based Resolution Operation Vm Detection Vo Merge C4 Conflict and Warning Detection C3 C4 C4 Vr2 C1 Adaptable Model Versioning Framework Existing technology has been reused. 12
  • 13. C2 Model Transformation By Demonstration  Composite operation specification 1 Initial 2 Working Create initial model Copy initial model Modeling model model Revised 3 model Perform updates 5 4 State-based Imply conditions Configuration & Generation comparison Conditions [implied] Diff model 6 Edit conditions Conditions 7 Generate Operation Operation [revised] Specification Specification Legend automatic manual P. Brosch, P. Langer, M. Seidl, K. Wieland, M. Wimmer, G. Kappel, W. Retschitzegger, and W. Schwinger. “An Example 13 Is Worth a Thousand Words: Composite Operation Modeling By-Example”. In Proceedings of the International Conference on Model Driven Engineering Languages and Systems (MoDELS’09), LNCS. Springer, 2009.
  • 14. C2 Model Transformation By Demonstration  Composite operation specification 1 Initial 2 Working Create initial model Copy initial model Modeling model model Revised 3 model Perform updates 5 4 State-based Imply conditions Configuration & Generation comparison Conditions [implied] Diff model 6 Edit conditions Conditions 7 Generate Operation Operation [revised] Specification Specification Legend automatic manual 14
  • 15. C2 Model Transformation By Demonstration  Composite operation specification 1 Initial 2 Working Create initial model Copy initial model Modeling model model Revised 3 model Perform updates 5 4 State-based Imply conditions Configuration & Generation comparison Conditions [implied] Diff model 6 Edit conditions Conditions 7 Generate Operation Operation [revised] Specification Specification Legend automatic manual 15
  • 16. C2 Model Transformation By Demonstration  Composite operation specification 1 Initial 2 Working Create initial model Copy initial model Modeling model model Revised 3 model Perform updates 5 4 State-based Imply conditions Configuration & Generation comparison Conditions [implied] Diff model 6 Edit conditions Conditions 7 Generate Operation Operation [revised] Specification Specification Legend automatic manual 16
  • 17. C2 Model Transformation By Demonstration  Composite operation specification 1 Initial 2 Working Create initial model Copy initial model Modeling model model Revised 3 model Perform updates 5 4 State-based Imply conditions Configuration & Generation comparison Conditions [implied] Diff model 6 Edit conditions Conditions 7 Generate Operation Operation [revised] Specification Specification Legend automatic manual 17
  • 18. C2 Model Transformation By Demonstration  Composite operation specification 1 Initial 2 Working Create initial model Copy initial model Modeling model model Revised 3 model Perform updates 5 4 State-based Imply conditions Configuration & Generation comparison Conditions [implied] Diff model 6 Edit conditions Conditions 7 Generate Operation Operation [revised] Specification Specification Legend automatic manual 18
  • 19. C2 Model Transformation By Demonstration  Composite operation specification 1 Initial 2 Working Create initial model Copy initial model Modeling model model Revised 3 model Perform updates 5 4 State-based Imply conditions Configuration & Generation comparison Conditions [implied] Diff model 6 Edit conditions Conditions 7 Generate Operation Operation [revised] Specification Specification Legend automatic manual 19
  • 20. C2 Model Transformation By Demonstration  Besides being executable… Adaptation Artifacts Match Operation Signifier Validation Rules Specifications Specifications Rules C2 C4 Vr1 Conflict Detection Conflict Detection Conflict-tolerant Operation-based State-based Resolution Operation Vm Detection Vo Merge C3 C4 C4 Vr2 Adaptable Model Versioning Framework 20
  • 21. C3 Operation Detection  Operation detection process 1. UUID-based matching 2. Rule-based matching 3. Atomic operation detection 4. Composite operation detection 21
  • 22. C3 Operation Detection  Operation detection process 1. UUID-based matching 2. Rule-based matching 3. Atomic operation detection 4. Composite operation detection Vo Vr1 lift lift Idle Idle DialTone Active hangup hangup DialTone dial dial Dialing hangup Dialing 22
  • 23. C3 Operation Detection  Operation detection process 1. UUID-based matching 2. Rule-based matching 3. Atomic operation detection 4. Composite operation detection Vo Vr1 lift lift Idle Idle DialTone Active hangup hangup DialTone dial dial Dialing hangup Dialing 23
  • 24. C3 Operation Detection  Operation detection process 1. UUID-based matching 2. Rule-based matching 3. Atomic operation detection 4. Composite operation detection Vo Vr1 lift lift Idle Idle DialTone Active hangup hangup DialTone dial dial Dialing hangup Dialing 24
  • 25. C3 Operation Detection  Operation detection process Diff 1. UUID-based matching Model 2. Rule-based matching 3. Atomic operation detection 4. Composite operation detection No difference Vo Vr1 lift lift Idle Idle DialTone Active hangup hangup DialTone dial dial Dialing hangup Dialing 25
  • 26. C3 Operation Detection  Operation detection process Diff 1. UUID-based matching Model ! 2. Rule-based matching 3. Atomic operation detection 4. Composite operation detection ! Transition target changed! Vo Vr1 lift lift Idle Idle DialTone Active hangup hangup DialTone dial dial Dialing hangup Dialing 26
  • 27. C3 Operation Detection  Operation detection process Diff 1. UUID-based matching Model !! 2. Rule-based matching 3. Atomic operation detection 4. Composite operation detection ! Transition source changed! Vo Vr1 lift lift Idle Idle DialTone Active hangup hangup DialTone dial dial Dialing hangup Dialing 27
  • 28. C3 Operation Detection  Operation detection process Diff 1. UUID-based matching Model !! + 2. Rule-based matching 3. Atomic operation detection 4. Composite operation detection + State Active has been added! Vo Vr1 lift lift Idle Idle DialTone Active hangup hangup DialTone dial dial Dialing hangup Dialing 28
  • 29. C3 Operation Detection  Operation detection process Diff 1. UUID-based matching Model !! + + + 2. Rule-based matching 3. Atomic operation detection 4. Composite operation detection Vo Vr1 lift lift Idle Idle DialTone Active hangup hangup DialTone dial dial Dialing hangup Dialing 29
  • 30. C3 Operation Detection  Operation detection process Diff 1. UUID-based matching Model !! + - + + 2. Rule-based matching 3. Atomic operation detection 4. Composite operation detection - Transition hangup has been deleted! Vo Vr1 lift lift Idle Idle DialTone Active hangup hangup DialTone dial dial Dialing hangup Dialing 30
  • 31. C3 Operation Detection  Composite operation detection process Diff Input Operation Operation Operation Diff Model Model Specification Specification Operation Specification Specifications !! + - ++ «foreach» «foreach» Diff Model Preprocessing Derive Precondition Derive Postcondition Binding Binding Input Operation Operation Operation Signature Specification Specification Signature Precondition Postcondition Binding Binding Preselection Evaluate Binding Evaluate Binding [no diff match] [diff matches] [invalid] [valid] [invalid] [valid] Valid Operation Operation Potential Operation Precondition Operation Operation Specification Specification Occurrence Operation Specification Specification Binding Occurrence 1 2 3 Refactoring Diff Refactoring Operation Occurrence Occurrence Occurrence Model P. Langer, K. Wieland, and P. Brosch. Specification, Execution, and Detection of !! + 31 + - + Refactorings for Software Models. In Proceedings of the WiP Session @ PPPJ’10, 2010.
  • 32. C4 Conflict Detection Input Adaptation Artifacts Diff Match Operation Signifier Validation Model1 Rules Specifications Specifications Rules C4 !! + - + + Vr1 Conflict Detection Conflict Detection Diff Conflict-tolerant Operation-based State-based Resolution Operation Vm Detection Model2 Vo Merge ! + + C4 C4 Vr2 Adaptable Model Versioning Framework 32
  • 33. C4 Conflict Detection  Operation-based conflict detection 1. Atomic operation conflict detection 2. Composite operation conflict detection 3. Signifier warning detection 33
  • 34. C4 Conflict Detection  Operation-based conflict detection 1. Atomic operation conflict detection 2. Composite operation conflict detection 3. Signifier warning detection 34
  • 35. C4 Conflict Detection  Atomic operation conflict detection  Nine conflict patterns for EMF-based models  Delete-use conflict du : DeleteUse delete use do : DeleteObject o : Object fo : FeatureOperation object value {self.oclIsTypeOf(InsertFeatureValue) or self.oclIsTypeOf(FeatureUpdate)}  Update-update conflict uu : UpdateUpdate update1 affected feature Object fo1 : FeatureOperation f : Feature o : Object {upperBound = 1 and fo1.value <> fo2.value} fo2 : FeatureOperation feature affected Object update2  … G. Taentzer, C. Ermel, P. Langer, and M. Wimmer. “A Fundamental Approach to Model Versioning Based on Graph Modifications”. Submitted to Software and System Modeling. Springer, 2011. 35
  • 36. C4 Conflict Detection  Operation-based conflict detection 1. Atomic operation conflict detection 2. Composite operation conflict detection 3. Signifier warning detection 36
  • 37. C4 Conflict Detection  Composite operation conflict detection For each composite operation application, check whether it is applicable at the opposite side Vo Diff Model1 !! + - + + Vr1 Vr2 !! + + + 37
  • 38. C4 Conflict Detection  Composite operation conflict detection  Revisiting the motivating example Vr1 lift Idle Active Vo Introduce hangup Composite DialTone lift Idle State dial DialTone dial hangup Dialing dial Dialing hangup Vr2 lift Idle DialTone abort dial dial Dialing hangup 38
  • 39. C4 Conflict Detection  Composite operation conflict detection Diff  Revisiting the motivating example Vr1 Model1 lift Idle !! + - + + Active Vo Introduce hangup Original Composite DialTone match Idle lift State dial SingleState_0 DialTone dial Transition_0 hangup Dialing SingleState_1 dial Transition_1 Precondition of Dialing SingleState_2 hangup composite Transition_2 Vr2 operation is lift violated! Idle SingleState_0 DialTone Transition_0 abort SingleState_1 dial Transition_1.name = Transition_2.name dial Transition_1 SingleState_2 Dialing hangup Transition_2 39
  • 40. C4 Conflict Detection  Composite operation warning detection For each composite operation application, check whether its match size increased at the opposite side Vo Diff Model1 !! + - + + Vr1 Vr2 !! + + + 40
  • 41. C4 Conflict Detection  Composite operation warning detection Add state “Active”, Move states into “Active”, Introduce fold transition Composite “hangup” State dial Add state Add state “Connecting” “Connecting” 41
  • 42. C4 Conflict Detection Your composite  Composite operation warning detection operation should incorporate opposite operations! Introduce Introduce Composite Composite State State dial Add state Add state “Connecting” “Connecting” 42
  • 43. C4 Conflict Detection  Composite operation warning detection Vr1 lift Idle Active Vo Introduce hangup Composite DialTone lift Idle State dial DialTone dial hangup Dialing dial Dialing hangup Vr2 lift Idle DialTone hangup dial dial Dialing hangup end_dialing 43 hangup Connecting
  • 44. C4 Conflict Detection  Composite operation warning detection Diff Vr1 Model1 lift Idle !! + - + + Active Vo Introduce hangup Original Composite DialTone match Idle lift State dial SingleState_0 DialTone dial Transition_0 hangup Dialing SingleState_1 dial Transition_1 Composite Dialing SingleState_2 hangup operation Transition_2 Vr2 is applicable! lift Idle SingleState_0 DialTone Transition_0 hangup SingleState_1 dial dial Transition_1 SingleState_2 Dialing hangup Transition_2 end_dialing 44 hangup Connecting
  • 45. C4 Conflict Detection  Composite operation warning detection Diff Vr1 Model1 lift Idle !! + - + + Active Vo Introduce hangup Original Composite DialTone match Idle lift State dial SingleState_0 DialTone dial Transition_0 hangup Dialing SingleState_1 dial Transition_1 Your composite SingleState_2 hangup Dialing operation should Vr2 include the state Transition_2 “Connecting”! lift Idle SingleState_0 DialTone Transition_0 hangup SingleState_1 dial dial Transition_1 SingleState_2 Dialing hangup Transition_2 end_dialing 45 hangup Connecting
  • 46. C4 Conflict Detection  Operation-based conflict detection 1. Atomic operation conflict detection 2. Composite operation conflict detection 3. Signifier warning detection 46
  • 47. C4 Conflict Detection  Signifier warning detection Change target Change target of r1 of r1 Change source Change source of r1 of r1 47
  • 48. C4 Conflict Detection  Signifier warning detection The superior meaning of r1 has been con- currently changed! Change target Redefinition of r1 of r1 Change source Redefinition of r1 of r1 48
  • 49. C4 Conflict Detection  Signifier warning detection  Signifier  Combination of properties of a model element type  Convey the “superior meaning” of a model element  e.g., the meaning of a reference is mainly conveyed by its name, source, and target  Signifier specifications (using ECL*) rule ReferenceSignifier match l : Left!EReference with r : Right!EReference { compare : l.name = r.name and l.source.isMatched(r.source) and l.target.isMatched(r.target) } * Epsilon Comparison Language (ECL), www.eclipse.org/gmt/epsilon 49
  • 50. C4 Conflict Detection  Signifier warning detection rule ReferenceSignifier match l : Left!EReference with r : Right!EReference { compare : l.name = r.name and l.source.isMatched(r.source) and l.target.isMatched(r.target) } Vr1 Match Trigger Vo Vr2 Match Trigger 50
  • 51. C4 Conflict Detection  Signifier warning detection rule ReferenceSignifier match l : Left!EReference with r : Right!EReference { compare : l.name = r.name and l.source.isMatched(r.source) and l.target.isMatched(r.target) } Vr1 Match Trigger Vo Concurrent Compare signifier of r1::Vo with r1::Vr1 signifier change signifier change of reference “r1”! Compare signifier of r1::Vo with r1::Vr2  signifier change You might want to duplicate “r1”? Vr2 Match Trigger 51
  • 52. C4 Conflict Detection Adaptation Artifacts Match Operation Signifier Validation Rules Specifications Specifications Rules C4 Vr1 Conflict Detection Conflict Detection Conflict-tolerant Operation-based State-based Resolution Operation Vm Vo Detection Merge C4 Vr2 Adaptable Model Versioning Framework cf. PhD thesis by Konrad Wieland cf. PhD thesis by Petra Brosch 52
  • 53. Evaluation  Adaptable Model Versioning Framework C1  Model Transformation By Demonstration C2  Operation Detection C3  Conflict Detection C4 53
  • 54. Evaluation  Model Transformation By Demonstration (MTBD) C2  Overall research question  Is it easier to use our MTBD tool than EMF Tiger*?  Study design  57 students (who have no experience in model transformation)  Specification of two composite operations with both tools  Questionnaire  Results  88 % preferred using our MTBD tool  74 % indicated that our MTBD tool is easy to use  Students needed nearly twice as much time using EMF Tiger  However, 53 % stated that configuring the conditions is hard * EMF Tiger: EMF-based graph transformation tool, user.cs.tu-berlin.de/~emftrans 54
  • 55. Evaluation  Operation Detection C3  Overall research question  How accurate is our tool for detecting applications of composite operations?  Study design  Applying our tool to the evolution of GMF*  141 composite operation applications  342 atomic operation applications  Results  98 % precision  “How many detected composite operations have actually been correct?”  70 % recall  “How many applied composite operations have been correctly detected?”  Recall decreases with a growing number of concurrent composite operations  Composite operation sequences * Graphical Modeling Framework (GMF), www.eclipse.org/modeling/gmf 55
  • 56. Evaluation  Conflict Detection C4  Overall research questions  How accurate is our tool for detecting atomic conflicts?  Can the same accuracy be achieved as operation tracking approaches?  Study design  Automatically executable benchmark  23 carefully selected versioning scenarios  12 of these scenarios include conflicting atomic operations  Benchmark applied to our tool, EMF Compare*, and EMFStore**  Results  Our tool: 100 % precision and recall  EMF Compare: 67 % precision and 17 % recall  EMFStore: 100 % precision and recall * EMF Compare, http://www.eclipse.org/emf/compare ** EMFStore, www.eclipse.org/emfstore/ 56
  • 57. Limitations and Future Work  Adaptable Model Versioning Framework C1  Are the provided adaptation points adequate and sufficient?  Empirical case study required!  Model Transformation By Demonstration C2  Configuring conditions still very challenging  Further automation of the specification process  Multiple demonstrations  Operation Detection C3  Overlapping sequences of operations  Conflict Detection C4  Deriving the conflicting opposite operation that caused a (composite operation’s pre-)condition to be invalid  Semantic inconsistencies 57
  • 58. Thank you for your attention!  Dissertation available  www.big.tuwien.ac.at/teaching/theses/70  Open source implementations  EMF Modeling Operations  www.modelversioning.org/emf-modeling-operations  AMOR Conflict Detection  code.google.com/a/eclipselabs.org/p/amor-conflict-detection 58