a drunk driver caused an accident outside Möhlin
an accident occurred on the main road between the Aargau municipalities of Möhlin and Mumpf
A 66-year-old man left the road on the out-of-town stretch and crashed into a roadworks barrier
The car then drove a good hundred meters through the field before coming to a standstill with total damage
This was reported by the Aargau cantonal police in a press release
The police found the driver uninjured but heavily intoxicated
The cantonal police reported him to the public prosecutor's office and temporarily confiscated his driving license
the Möhlin Table Tennis Club has set itself an ambitious challenge: hosting the Swiss Elite Championships on the weekend of March 22-23
the competition will advance to the round of 16
setting the stage for an action-packed Sunday
and grand final will determine the new Swiss champions
The 64 best male and 32 best female players
along with Switzerland’s top para-athletes
The Möhlin Table Tennis Club and Swiss Table Tennis are eager to welcome both players and fans to the Steinli Sports Center
where everything has been planned to create a warm and vibrant atmosphere
For all information about the Championships
Recap: Swiss Elite & Para Championships 2024 in Rapperswil-Jona
The Swiss Elite & Para Championships 2024 took place on March 23-24 in Rapperswil-Jona
featuring Switzerland’s finest players in thrilling matchups
The final saw Barish MOULLET face Elias HARDMEIER
MOULLET dominated the first three sets (11-3
but HARDMEIER fought back to level the match at 3-3
MOULLET clinched his second national title with a 13-11 victory
Ten-time champion Rachel MORET took on Fanny DOUTAZ in the final
but MORET responded to force a deciding set
ending MORET’s dominance that had lasted since 2016 (except in 2022 due to WTT Smash)
The pair Barish MOULLET & Gaël VENDÉ faced Cédric TSCHANZ & Pedro OSIRO in the final
The duo of Rachel MORET & Fanny DOUTAZ proved unstoppable
securing the title with a 3-1 victory over the TULLII-MAURER pair
Defending champions Rachel MORET & Nicolas CHAMPOD were aiming for their fifth consecutive title
but they were outplayed by Fanny DOUTAZ & Yoan REBETEZ
giving DOUTAZ her third title of the competition
Para: Rolf ACKLIN defended his title with a dominant 3-0 win over Kevin KÖCHLI
Tetra: Dirk KRETZSCHMAR claimed his third consecutive title
Standing: Fabian LICHTIN achieved a historic hat-trick
Para/Tetra Doubles: Silvio KELLER & Kevin KÖCHLI showcased their teamwork
overcoming Ange MONEMOTO & Martin ALTERMATT 3-1 (13-11
The Swiss Elite & Para Championships 2024 delivered incredible moments
setting the stage for another thrilling edition in MÖHLIN 2025
The European Table Tennis Union (ETTU) is the governing body of the sport of table tennis in Europe
and is the only authority recognized for this purpose by the International Table Tennis Federation
The ETTU deals with all matters relating to table tennis at a European level
including the development and promotion of the sport in the territories controlled by its 58 member associations
and the organization of continental table tennis competitions
Senior Director & Distinguished Economist
Kristina Mohlin is a Senior Director and Distinguished Economist, Policy Analysis in EDF's Economics team
where she provides economics expertise to support the organization’s advocacy on energy markets and looks for ways to bridge the gap between the research community and policymakers
Kristina’s area of expertise is climate and energy policy
She leads and coordinates analyses on economics and policy instrument design that supports EDF's advocacy focused on reducing methane emissions from the global oil and gas sector
gas pipeline transportation and wholesale electricity markets and on electricity tariff design
For more information about Kristina’s research, please visit her own website
Kristina was a PhD student at the Department of Economics at the University of Gothenburg
under the supervision of Profs Jessica Coria and Thomas Sterner
she studied engineering and served as a research assistant at the Division of Physical Resource Theory at Chalmers University of Technology
Mohlin, K., M. Lackner, H. Nguyen, A. Wolfe (2022). "Policy instrument options for addressing methane emissions from the oil and gas sector." Environmental Defense Fund Economics Discussion Paper Series
Lackner, M., J.R. Camuzeaux, S. Kerr and K. Mohlin (2021). “Pricing methane emissions from oil and gas production.” Environmental Defense Fund Economics Discussion Paper Series
Mohlin, K., A. Piebalgs and M. Olczak (2021). “Designing an EU methane performance standard for natural gas.” Florence School of Regulation Policy Brief
Robert Schuman Centre for Advanced Studies
Mohlin, K. (2021). “The US gas pipeline transportation market. An introductory guide with research questions for the energy transition.” Environmental Defense Fund Economics Discussion Paper Series
Spiller, E., R. Esparza, K. Mohlin, K. Tapia-Ahumada, B. Ünel (2023). "The role of electricity tariff design in distributed energy resource deployment." Energy Economics 120
Mohlin, K., E. Spiller, K. Tapia-Ahumada, B. Unel (2021). "Cost-Reflective Residential Electricity Tariff Design: Synthesis of research findings on the impacts on customer bills, adoption of distributed energy resources and pollution”." Environmental Defense Fund Economics Discussion Paper Series
Unel, B., R. Esparza, E. Spiller and K. Tapia-Ahumada (2021). “Retail electricity design, distributed energy resources and emissions.” Institute for Policy Integrity at NYU School of Law
Spiller, E., R. Esparza, K. Mohlin, K. Tapia-Ahumada and B. Unel (2020). “The role of electricity tariff design in distributed energy resource deployment.” Environmental Defense Fund Economics Discussion Paper Series
Mohlin, K., A. Bi, S. Brooks, J.R. Camuzeaux and T. Stoerk (2019). "Turning the corner on US power sector CO2 emissions - a 1990-2015 state level analysis." Environmental Research Letters
Mohlin, K., J.R. Camuzeaux, A. Muller, M Schneider and G. Wagner (2018). “Factoring in the forgotten role of renewables in CO2 emission trends using decomposition analysis.” Energy Policy
Coria, J. and K. Mohlin (2017). "On refunding of emission taxes and technology diffusion." Strategic Behavior and the Environment
Convery, F. J., K. Mohlin and E. Spiller (2017). “Policy brief—Designing electric utility rates: insights on achieving efficiency, equity, and environmental goals.” Review of Environmental Economics and Policy
"An introduction to the Green Paradox: The unintended consequences of climate policies." Review of Environmental Economics and Policy
Bonilla, J., J. Coria, K. Mohlin and T. Sterner (2015). "Refunded emission payments and diffusion of NOx abatement technologies in Sweden." Ecological Economics
Badtke-Berkow, M., M. Centore, K. Mohlin and B. Spiller (2015). “ Primer on time-variant electricity pricing,” EDF Whitepaper
Mohlin, K., B. Spiller, M. Badtke-Berkow, J. Fine, G. Donzelli and C. Larose (2014). “SolaROI: estimating returns to residential solar panels from underlying tariff structures and compensation mechanisms," EDF Whitepaper
“Putting a price on the future of our children and grandchildren.” In Livermore
The Globalization of Cost-Benefit Analysis in Environmental Policy
Wirsenius, S., F. Hedenus and K. Mohlin (2011). “Greenhouse gas taxes on animal food products: rationale, tax scheme and climate mitigation effects.” Climatic Change
Copyright © 2025 Environmental Defense Fund. All Rights Reserved. Privacy statement. Terms of Use.
tax-exempt charitable organization under Section 501(c)(3) of the Internal Revenue Code
Donations are tax-deductible as allowed by law
EDF operates with the highest standards of ethics and open, honest communication. Read our whistleblower policy or report a concern through our confidential, third-party compliance site.
Blended modeling aims at boosting the development of complex multi-domain systems by enabling seamless multi-notation modeling. The synchronization mechanisms between notations are embodied in model transformations. Manually defining model transformations requires specific knowledge of transformation languages, and it is a time-consuming and error-prone task. Moreover, whenever any of the synchronized languages or notations evolves, those transformations become obsolete.
In this paper, we propose an automated solution for generating synchronization transformations in an industrial setting.
The approach entails i) the specification of mapping rules between two arbitrary domain-specific modeling languages leveraging a mapping modeling language, appositely defined for this purpose, and ii) the automatic generation of synchronization model transformations driven by the mapping rules.
We validated the proposed approach in two use cases. Although our main goal was to provide a solution for synchronization between graphical and textual notations of UML-RT state machines, the proposed approach is language- and notation-agnostic.
Volume 4 - 2022 | https://doi.org/10.3389/fcomp.2022.1008062
This article is part of the Research TopicModel-Centered Software and System DevelopmentView all 6 articles
Introduction: Blended modeling aims at boosting the development of complex multi-domain systems by enabling seamless multi-notation modeling
The synchronization mechanisms between notations are embodied in model transformations
Manually defining model transformations requires specific knowledge of transformation languages
and it is a time-consuming and error-prone task
whenever any of the synchronized languages or notations evolves
we propose an automated solution for generating synchronization transformations in an industrial setting
Results: The approach entails i) the specification of mapping rules between two arbitrary domain-specific modeling languages leveraging a mapping modeling language
and ii) the automatic generation of synchronization model transformations driven by the mapping rules
Discussion: We validated the proposed approach in two use cases
Although our main goal was to provide a solution for synchronization between graphical and textual notations of UML-RT state machines
the proposed approach is language- and notation-agnostic
by using domain-specific abstractions formalized in domain-specific modeling languages (DSML)
to describe complex functions in a more domain-focused and human-centric way than if using traditional programming languages
DSMLs formalize the communication language of engineers at the level of domain-specific concepts such as an engine and wheels for a car
These concepts may not exist in other domains
DSMLs support more efficient integration of software with designs and implementations of other disciplines
Domain-specific modeling demands a high level of customization of modeling tools
typically involving combinations and extensions of DSMLs and tailoring of the modeling tools for their respective development domains and contexts
tools are expected to provide multiple modeling means
to satisfy the requirements set by different development phases
especially those based on the Unified Modeling Language (UML) and its profiles (as in the industrial tool addressed in this paper)
traditionally focus on one specific notation
A notation that is well understood by one engineering discipline may not be as easily understood by engineers from another discipline
engineers from the same or different disciplines may have different notation preferences; not supporting multiple notations negatively affects the throughput of engineers
choosing one particular notation also limits the pool of available tools to develop and manipulate models that may be needed
choosing a graphical notation limits the usability of text manipulation tools such as text-based diff/merge
This mutual exclusion suffices the needs of developing small-scale applications with only few stakeholder roles
We have defined the notion of blended modeling in a previous work (Ciccozzi et al., 2019) as follows:
Blended modeling is the activity of interacting seamlessly with a single model (i.e.
abstract syntax) through multiple notations (i.e.
allowing a certain degree of temporary inconsistencies
Blended modeling is expected to aid in keeping the cognitive flow of modeling effective and efficient, offering stakeholders a proper set of intertwined formalisms, notations, and supporting computer-aided mechanisms. This is important in the design of modern systems, as their complexity has been increasing exponentially over the past years (Persson et al., 2013)
the focus is not on identifying viewpoints and related views
but rather on providing multiple blended editing and visualizing notations to interact with a set of concepts
The blended modeling paradigm focuses on the provision of multiple concrete syntaxes
for a non-empty set of abstract syntactical concepts
it aims to accommodate different notations
each designed for particular modeling needs
The main implication of this definition of blended modeling is that it assumes a single abstract syntax supported by multiple concrete syntaxes
although this definition is theoretically correct
language-specific modeling frameworks can benefit from multiple abstract syntaxes for the following reasons
the definition and management of a concrete syntax may be simpler if directly related to a dedicated abstract syntax
where we started from a graphical concrete syntax
the first step involved the definition of a textual concrete syntax by first defining a dedicated abstract syntax
This was needed since the two abstract syntaxes were not fully matching
and it generally allows for high syntax-specific customizations
since notations are usually meant to be highly customizable to user needs
and each notation serves a different purpose
it may not be practical to adapt an existing abstract syntax supporting one notation to another
depending on the needs of two different users
there might exist two “different” notations of the same type (e.g.
two different textual notations focusing on different aspects of the same modeling concepts tailored to two different user types)
Having a dedicated abstract syntax per notation alleviates possible syntactical “pollution” caused by reusing and adapting an existing abstract syntax
which was not envisioned for that particular notation
there exist scenarios where different notations are represented by different existing notation-specific DSMLs
formalizing the same underlying language with a significant overlap
to serve the needs of different communities
blended modeling is not always limited to seamless interaction with a single abstract syntax through multiple notations
The representative scenario of our work is a single underlying language (set of concepts) formalized through multiple abstract syntaxes
We define this as extended blended modeling
the abstract syntaxes may represent either two partly overlapping formalizations of the same DSMLs or even two entirely different DSMLs
provided that they are in some way related to each other
The remainder of the paper is organized as follows
The industrial setting and supporting arguments on the motivation behind this work are described in Section 2
We discuss work related to the problem domain in Section 3
The developed approach and implementation are described in detail in Section 4
Experiences from validating the approach are included in Section 5
A discussion is provided in Section 6 and the paper is concluded in Section 7
The research work described in this paper was carried out in cooperation with HCL Technologies, which offers an industrial Eclipse-based modeling tool, RTist, for the development of complex
The tool is designed to support UML and its real-time profile (UML-RT)
the tool provides support for specifying UML-RT architectures and applications by means of graphical composite structure diagrams
the tool provides specific features to complement models with fine-grained algorithmic behaviors by embedding C/C++ action code in state-machines
whitespaces) between two semantic tokens and the provision of synchronization mechanism between textual and graphical notations
the synchronization mechanisms were manually defined in terms of model transformations between two specific DSMLs describing textual and graphical notations
If any of the two concrete syntaxes underwent a change
The specific industrial aim of this work was instead to provide an automated solution for generating synchronization infrastructures between potentially evolving concrete syntaxes of UML-RT
To allow for evolution of the entailed DSMLs and the co-evolution of the synchronization mechanisms
in this work we contributed with the design and implementation of a mechanism for the automatic generation of the infrastructure required for seamless synchronization
between virtually any pair of Ecore-based DSMLs (not only graphical and textual UML-RT state-machines)
that may or may not represent two different concrete syntaxes of the same language
The provision of automatic means for generating model synchronization transformations from two given DSMLs that may or not represent two versions of the same language simplifies remarkably the life of modeling tool developers
but also allows domain experts without particular knowledge in model transformations to practically put in place the infrastructure needed for synchronization purposes
our approach brings the following benefits
• It provides the means for seamless synchronization of the standard graphical and the newly introduced textual concrete syntaxes for UML-RT state-machines
customers require viewing and editing UML-RT models using various specialized notations
These DSMLs are tailored to customer needs and
unlike the underlying UML-RT based language
as soon as any of the notations undergoes a change
the synchronization mechanisms can be regenerated with a minimal mapping effort from the developers
the actual model synchronization transformations would need to be manually edited by the developers
Our solution gives architects and developers the possibility to experiment when extending or evolving either concrete syntax
in case the UML-RT language itself would evolve
without a solution like this based on automatic generation of synchronization
all transformations would need to either be co-evolved manually
or re-written from scratch in case of deep changes to the language and/or the related concrete syntaxes
Our solution eases this process and provides the means for a more flexible and “fast prototyping” kind of modeling language and tool engineering process
Engineers and developers can sketch changes to either of the concrete syntaxes and try them out with automated generated synchronization
• If any other language would be included in the tool ecosystem for
our solution aids in establishing synchronization infrastructures between them
Automation gives flexibility but also the possibility to “try out” alternatives without having to spend much effort and time writing and validating synchronization transformations
since the mapping correspondences are defined between elements of M3 level models
while we target the specification of mapping correspondences between elements of M2 level models
write and read) with the model via multiple notations
but the solution only provides pop-up boxes to textually edit elements of graphical models rather than allowing seamless editing of the entire model
are based on the concept of only one abstract syntax
whereas our focus is on multiple abstract syntaxes that may represent two partially overlapping formalizations of the same DSML or even two completely different DSMLs as long as they are related in some way
These approaches focus on the semi-automatic creation of weaving models and their manual adjustment for semantically and syntactically similar languages
and the manual creation of weaving models for semantically and syntactically different languages for several purposes
relying only on metamodel data to create weaving models (i.e.
mapping models) without considering the developer's intentions does not guarantee the accuracy of the mapping model with respect to the requirements
The use of inaccurate and ambiguous weaving models may result in incorrect model transformations that do not meet the initial requirements
Manual verification and adjustment of an extensive weaving model can be as challenging as finding a needle in the haystack
This might lead to the creation of mapping models being simpler than manually fine-tuning automatically generated ones
while one may argue that the weaving approaches provide a flexible and automated way to derive mapping models
they may not be able to properly deal with semantic differences among the mapped languages (i.e.
semantics often needs human understanding to be correctly managed)
the mapping modeling language is not the main focus
but rather a key enabler for the overall goal being the definition of powerful higher-order transformations (HOTs) for generating synchronization transformations and addressing challenging cases
such as synchronization between different languages
by allowing for more complex unambiguous mappings
we can cater to a wider range of languages
and provide powerful mechanisms to support the translation of these mappings to model transformations
thus the generation of the synchronization infrastructure between languages and notations
we also increase the generalizability of our approach
these solutions tend to provide a tree-based editor only
while providing an additional textual editor can prove useful thanks to its syntax-agnostic editing features
a textual mapping language that defines correspondences between EMF objects and RDF resources
The approach generates ATL transformations that enable the use of RDF resources as EMF objects and the serialization of EMF objects in RDF resources
While the last two solutions provide mapping languages and semi-automatic approaches for the generation of model transformations
they do not provide support for Ecore-based DSMLs
Consider the model of our solution depicted in Figure 1
Depending on what the involved pair of DSMLs represents
DSMLA and DSMLB represent two notations of the same language (e.g.
graphical and textual UML-RT state-machines)
then the generated M2M transformations provide synchronization across different notations of the same language
then the generated M2M transformations provide synchronization across different notations of different languages
Semi-automatic synchronization infrastructure
We refer to a mapping language as a structured and formalized means for the specification of mapping rules between two or more DSMLs. The definition of the mapping language is given in terms of a metamodel; thus, it can also be defined as the correspondence of elements between arbitrary metamodels (Lopes et al., 2006)
A mapping language provides a fundamental input to correctly synchronize models conforming to different DSMLs
as explicit mapping rules link multiple DSMLs deterministically
mapping rules in those models drive the HOTs to properly generate model transformations conforming to QVTo
The mapped DSMLs shall conform to the Ecore meta-metamodel and may represent two different notations of a same language
Although more intuitive and easier to interact with than complex model transformations
MML is still intended for users with meta-modeling knowledge
Understanding of the meta-modeling concepts and structure is essential to properly describe how concepts between DSMLs are intended to be mapped
Definition of mapping rules instead of manually writing model transformations is particularly useful for domain experts with no specific knowledge of model transformation languages
but also for developers who can benefit from a semi-automatic
and less cumbersome approach for establishing synchronization mechanisms
domain knowledge is the only required input
Also maintenance of the generated model transformations in response to evolving DSMLs or requirements can be performed by domain experts
since adjustments only need to be made at the level of the mapping models
while HOTs would use them to regenerate model transformations
That is to say that developers are not expected to manually edit generated model transformations at any point in the process
we conducted an analysis to identify the input required to correctly generate model transformations
Considering that one of the main goals of this study was to minimize the amount of manual input required from the user
we first identified the maximum set of information that could be automatically retrieved from DSMLs (DSMLA and DSMLB)
abstract or non-abstract) is calculated based on whether the source element is abstract or not
disjunctive/disjuncted mapping rules and inheriting/inherited mapping rules are automatically generated by examining the hierarchical structure of the involved metamodels
The user is also relieved from invoking the corresponding mapping rules as they can also be retrieved automatically
The assignment operator is automatically generated in the case of mono-valued attributes or properties (i.e.
while in the case of multi-valued elements the user must manually define it based on whether the goal is to add elements to the collection (i.e.
+=) or to reinitialize the collection with the element
and ->) are also automatically generated by assessing whether the source is a single element or a collection of elements
navigation paths are also automatically generated when they involve containment references or when there is a single non-containment reference between two EClasses
The remaining information would have to be manually provided by the user
we defined the concepts to be included in the MML metamodel
Once identified the concepts that MML should include
focusing on abstracting away the implementation—specific details (e.g.
syntax and semantics) of model transformation languages
and allowing the user to focus exclusively on the specification of mapping rules
we implemented the MML in a blended modeling fashion
allowing the user to interact with MML via both textual and tree-based editors
since it combines the strength of text for syntax-agnostic editing operations
and a good integration with widely used versioning and configuration tools
with clear structural overview features typical of tree-based editors
We assume that MML users have experience with at least one object-oriented programming (OOP) language; therefore, several syntactical features of MML are similar to those of OOP languages. In addition to that, users have an advantage if they have some knowledge of the Object Constraint Language (OCL)
since in certain cases mapping rules require the specification of conditions for the correct navigation of elements and for the expression of the so-called “guards” in the generated transformations
MML is developed using the Xtext language workbench. Xtext relies on EMF and it generates an Ecore model that represents the abstract syntax tree (AST), lexer, parser, and the corresponding Java code. In Figure 2 we describe the MML metamodel defined as an Ecore model
and in the following we detail the metaconcepts of MML
*Unbounded upperbound for cardinalities
The MappingModel serves as the root of the metamodel and is a tuple <name
where name is a unique name for MappingModel
Rules* is a possibly empty set of elements of type MappingRule
SourceMetamodels* and TargetMetamodels* are sets of elements of types SourceMetamodel and TargetMetamodel respectively
MainSourceMetamodel is a single element of type SourceMetamodel that in the case of multiple SourceMetamodels is required to indicate the SourceMetamodel to be used at the entry point of the transformation to be generated
ChildHelpers*> where name is a unique name for MappingRule
and operator represents the type of operator between mappings (i.e.
This is required when it comes to Collections to determine whether the user intends to append an element to the Collection or to reinitialize the Collection by deleting all previous elements and adding the new one
condition supports the definition of a condition that can be interpreted in different ways depending on the type of source and target elements of the mapping rule (i.e.
mapping guard for EClasses and OCL filter for EReferences and EAttributes)
comment supports the definition of comments to the mapping rule which can help the user keep track of the piece of generated code with the corresponding mapping rule
source and target are optional elements of type EObject that represent the source and target elements of the MappingRule
helperLiteral is used for EEnumLiterals and is included since EcoreQualifiedNameProvider does not support EEnumLiterals
we need two references; one to the EEnum and the other to the EEnumLiteral
source or target will be used to reference EEnum and helperLiteral to reference EEnumLiteral
ChildRules* is a possibly empty set of elements of type MappingRule
while ChildHelpers* is a possibly empty set of elements of type HelperStatement
SourceMetamodel and TargetMetamodel represent the DSMLs that will be involved in the transformation and inherit all members of Metamodel
where name is a unique model name and model is the EPackage representing the root element of a particular metamodel involved in the mapping
A HelperStatement is a tuple <statement
ChildHelpers*> where statement is a unique element that allows the user to define statements; for the moment
Operator is an enumeration with two mutually exclusive possible values
used when a single input element in the source model is mapped to a single output element in the target model
or when a non-empty set of input elements in the source model are mapped to a non-empty set of output elements in the target model by re-initializing the set of output elements
used when a non-empty set of input elements in the source model are transformed into a non-empty set of output elements in the target model by adding to the set of output elements
we leverage the features provided by Xtext in combination with EMF to automatically generate textual and tree-based editors
we customize them to provide a more user-friendly and precise scoping as well as more intuitive labeling of the mapped model elements
we specialize the MappingRuleItemProvider class
especially for the source and target plays a significant role in reducing the likelihood of errors on the part of the user
the customization of scoping limits the user into defining child mapping rules (i.e.
and EEnums) only if there exists a navigation path from the source and target element (i.e.
EClass) of the main mapping rule to the source and target of the child mapping rule
This is particularly useful in the tree-based editor for distinguishing between different metaelements that may have the same name
we specialize the Formatter class to customize indentation
to improve the readability of MML textual models
and two sets of high-level mapping rules between them defined in two mapping models
the HOTs generate two unidirectional QVTo transformations that
and transform it into a model instance of the other DSML
Each mapping rule defined in the mapping models is transformed into one or more mapping operations in the generated QVTo transformations
The choice of QVTo as target transformation language was due to its suitability for both in-place and out-of-place
as well as endogenous and exogenous transformations
and for its imperative programming fashion
which is particularly suitable for automatic generation of complex algorithms
we opted for multiple unidirectional transformations rather than bidirectional transformations to simplify the maintainability of the generated transformations and their manageability in the target modeling tool ecosystem
which compiles into readable Java 8 compatible source code and is particularly suitable for the generation of pretty-printed textual artifacts
The remainder of this section is structured in paragraphs corresponding to the different metaconcepts of MML to maximize readability
we describe how the HOTs combine the input specified by the user in the mapping models with the information automatically extracted from the mapped DSMLs in order to generate the model transformations for synchronization
Mapping Model: MappingModel is the root element of the mapping language and represents the starting point for traversing a mapping model
which is then used to generate the name of the model transformation
the user must select the MainSourceMetamodel
which represents the metamodel that is used as the entry point of the model transformation
the latter is automatically selected as MainSourceMetamodel
Metamodels: The information extracted from SourceMetamodel and TargetMetamodel is used to generate the modeltype
and transformation signature of the model transformations
The user loads the DSMLs and selects the EPackages to be mapped that will be used as the source and target of the model transformation
The HOTs automatically retrieve the name and nsURI of the EPackages to generate the modeltype
identify the direction of the model transformation
and generate parts of the transformation signature according to the following pattern:
in «SourcePackageName»Model : «SourcePackageName»
out «TargetPackageName»Model : «TargetPackageName»
Mapping Rule: The mapping rules can be grouped based on the EObject that contains them as follows
MappingRules are contained in MappingModel and we refer to them as immediate mapping rules
source and target of these mappings are objects of type EClass
MappingRules are contained in other MappingRules or HelperStatements and we refer to them as child mapping rules
source and target of these mappings are objects of type EReference
The mapping rules in the first category are used to generate the mapping declaration
whereas those in the second are utilized to generate the body of the mapping operations
we detail the implementation of the features that apply to each category
Mapping operation name: The name of the mapping operation is automatically generated as: «sourceElementName»2«targetElementName»
This not only reduces the amount of manual effort from the user
as the naming follows a specific standard pattern and is rather intuitive
to minimize the risk of errors when mapping elements with the same name
source and target elements are printed using fully qualified names (i.e.
Mapping operation type: The generated mapping operations can be abstract or non-abstract
Abstract mapping operations are used when the target of the mapping operation is abstract
This information is automatically extracted from the target DSML; hence
the HOTs determine whether the target element of the mapping rule is an abstract or non-abstract EClass
the mapping operation is printed as abstract according to the following pattern: Abstract«sourceElementName»2«targetElementName»
Conditions: For immediate mapping operations
the source and target elements are EClasses
therefore conditions that are manually defined by the user are automatically generated as when clauses that are evaluated to determine in which circumstances the mapping operation should be executed
Inheritance: The concept of inheritance allows reuse of mapping operations under the condition that the signature of the inherited mapping conforms to the one of the inheriting mapping
The source and target of any potential inherited mapping rule must be supertypes of
the source or target of the inheriting mapping rule
The HOTs iterate through each of the immediate mapping rules in the mapping model and determine whether the mapping operation under analysis inherits from any of the iterated mapping rules
after the transformation signature and the inherits keyword
the names of the inherited mapping rules are printed (in the case of multiple inherited mapping rules
Disjunction: Invocation of a disjunct mapping operation results in an assessment of disjunct candidate mapping operations
To determine whether a mapping operation is disjunctive and
the HOTs iterate through all the immediate mapping rules of the mapping model and identify those where the source and target are identical or subtypes of the source and target of the potentially disjunctive mapping rule
the analyzed mapping rule is considered disjunctive and is named according to the following pattern: «sourceElementName»2«targetElementName» Disjunct
After printing the signature of the mapping operation and the disjuncts keyword
the HOTs print the identified disjunct candidates
A mapping can be both abstract and disjunctive
The user needs to define the mapping rule only once in the mapping model and the HOTs will generate two rules: one abstract and one disjunctive
since QVTo does not allow to combine them into one
There are three different possible scenarios for child mapping rules
depending on the values of the source and target attributes
In SC1 a non-empty set of input elements in the source model are transformed into a non-empty set of output elements in the target model. In SC2 a non-empty set of output elements are added to the target model. In SC3 a non-empty set of input elements in the source model facilitates the navigation of model elements in the generated transformations. This is used for complex and possibly ambiguous navigation cases, such as the one depicted in Figure 3
where on the left-side is illustrated an excerpt from the source metamodel and on the right side an excerpt from the target metamodel
Consider that the user defines an immediate mapping rule Organization2Company as detailed in Figure 4
The user then wants to map the value of the name attribute of Person in the source metamodel
to the managerName attribute of Company in the target metamodel
by defining the mapping rule name2managerName
The correct navigation path in this case would be self.department.ma-nager.name
also another path; self.department.secretary.name
this navigation path is not considered correct
as it would map the name of a Person that is a secretary in Department to managerName in Company
the information to navigate to name attribute via manager reference is to be decided by the user
as the HOTs cannot automatically determine which path to take
the user needs to define an additional mapping rule manager2null that will guide the HOTs in generating the expected model transformation
Being that manager is not an immediate reference of Organization HOTs must automatically generate the navigation path from Organization to manager
we implement a recursive Depth-First Search (DFS) algorithm
Organization) and explores as far as possible along each EClass
before backtracking (unless it finds the target)
mapping operations are run with an explicit rule-invocation style
which initiates execution from an entry mapping operation generally found in the main function
and invokes the other mapping operations in a nested manner
The entry mapping operation that should be invoked is automatically determined
OCL expressions: For sub-mapping operations
where the source and target elements are EReferences or EAttributes
condition attributes of the mapping rules are used to specify OCL expressions
Navigation operators: The HOTs determine the navigation operator based on whether the source of the mapping rule is a single object or a collection of objects (i.e.
A single object is navigated using the dot (.) operator
whereas collections of objects are navigated using the arrow (->) operator
HelperStatement: It is intended to facilitate the definition of complex mappings requiring the use of for loops
HelperStatements are contained in MappingRules (i.e.
immediate mapping rules) similarly to how they are defined in mapping operations in QVTo transformations
they may contain other HelperStatements and MappingRules that are generated within the loop or statement defined by HelperStatement
The proposed approach and reference implementation have been validated by means of two use cases and model-to-text testing
the implementation of our solution was validated in multiple testing phases and the results of each phase were analyzed and used
In Section 5.1 we provide details on the two use cases in isolation and then conduct a comparison between the two
while in Section 5.2 we provide the details of the model-to-text test cases
in Section 5.3 we provide an example of our industrial use case
The first use case refers to the UML-RT language
more specifically the subset for modeling state-machines
where DSMLA and DSMLB represent the graphical and textual notation of the UML-RT language
The second use case concerns two disjoint DSMLs
one for describing and manipulating calendars
while the other for describing and manipulating organizational structures
Both use cases encompass scenarios entailed by our solution
UML-RT is a real-time profile that aims to simplify the ever increasing complexity of the software architecture specification for real-time embedded systems
UML-RT enables both structure modeling and behavior modeling of real-time systems
This use case focuses on the behavioral part which is represented using state-machine diagrams
Considering that both DSMLA and DSMLB represent two different notations of the UML-RT language
As a result of textual concrete syntax requirements aimed at maximizing usability and reducing learning curves
the different DSMLs for different notations are required
the DSML associated with the textual notation has evolved to fit the needs of various customers
so we have been able to also support the co-evolution of model transformation in response to DSML changes
we provide more details on the mapping models and generated QVTo transformations
The Textual2Graphical mapping contains a total of 71 mapping rules
as they are applied to mapping rules that link two EClasses
This mapping model generates a total of 29 main mapping operations in the output QVTo transformation from a total of 25 mapping rules that link EClasses
The four additional mapping operations are the result of the abstract and disjunctive mappings that are automatically calculated from the HOTs
The Graphical2Textual mapping contains a total of 61 mapping rules
while the other are in the form of OCL filters
This mapping model generates a total of 26 mapping operations in the output QVTo transformation from a total of 22 mapping rules that link EClasses
The same reasoning as in the case of Textual2Graphical mapping applies for the four additional mapping operations
Making a comparison between the two mapping models
we notice that the most significant differences are with regard to SC2 and SC3
While in the Textual2Graphical mapping model only 1.4% of the mapping rules fall under SC2 (i.e
are used for adding a non-empty set of elements in the output model)
in the Graphical2Textual mapping model 8.2% of the mapping rules fall under SC2
This is a consequence of the fact that the DSML representing the textual notation contains more concepts that are either not present in the DSML representing the graphical notation (e.g.
TransitionBody) or are more specialized (e.g.
The high number of mapping rules that contain conditions in the Graphical2Textual mapping model compared to the Textual2Graphical one is another indicator of the specialization of concepts
we notice that while the Graphical2Textual mapping model has no mapping rules falling under this category
in the Textual2Graphical mapping model 5.6% of the mapping rules are used to facilitate the navigation of elements in the textual model that cannot be directly accessed
The second use case relates to two disjoint DSMLs where one is used to describe a meeting calendar for an organization
while the other is used to describe the organization
An organization consists of personnel that can have different availability (e.g.
available or on vacation) and is divided into multiple departments
Each department is responsible for multiple projects
which in turn consist of multiple work packages and external partners
active or non-active) and consists of multiple tasks for which external partners and/or organization personnel are in charge
A calendar can be split into multiple divisions
where each division consists of active meetings
and personnel that is not participating in any meeting
Active and non-active meetings consist of a group of participants composed of internal
These are two semantically and syntactically disjoint DSMLs
thus the definition of mapping links might not be as intuitive as for the first use case
as depending on the status of work packages in the model representing the organization and people in charge
meetings are automatically created in the calendar
There is a similar relation for the reverse transformation
The Calendar2Organization mapping model contains a total of 50 mapping rules
one under SC2 (2%) and four under SC3 (8%)
Eight mapping rules contain conditions and they are all in the form of OCL filters
this mapping model introduces the use of HelperStatements in the form of for loops and if conditional statements
This mapping model generates a total of 12 mapping operations in the output QVTo transformation from a total of 11 mapping rules that link EClasses
The Organization2Calendar mapping model contains a total of 40 mapping rules
of which seven are in the form of guards as they are applied to mapping rules that link two EClasses
while three are in the form of OCL filters
this mapping model introduces the use of HelperStatements in the form of if conditional statements
we notice that the number of mapping rules that fall under SC1 is equal in both
It is important to note that the two DSMLs contain an approximately equal number of elements (i.e.
while Calendar contains 39 elements) and an equal number of EClasses; thus
which deeply affects the distribution of mapping rules with regard to the three scenarios
Assuming that there is no loss of information (typically occurs when not all elements of the involved DSMLs are linked by mapping rules)
after the execution of the forward and backward transformations
in the case of two DSMLs of significantly different sizes
we believe that it is likely to have a higher number of mapping rules associated with SC2 and SC3
Comparing the distribution of the mapping rules between the three scenarios
the number of mapping rules that fall under SC2 and SC3 is mainly due to the specialization of concepts
while in the second use case it is due to semantic and syntactical differences
Despite the fact that there is no significant difference between the number of mapping rules falling under SC1 for the first and the second use case
we still argue that the second use case is more complex than the first
since while in the UML-RT use case there is a string similarity between the mapped elements of the involved DSMLs and similarity in the structure of the DSMLs
in the second use case such similarities cannot be found
while the first use case covers only a subset of the concepts of the MML
the second use case covers all concepts of the MML including the HelperStatement and helperLiteral
which we could not validate in the first use case
What adds to the complexity of the second use case is that
while the mapping models for the UML-RT use case exhibit a flatter hierarchy (a maximum of two-level deep-nested hierarchies)
the mapping models of the second use case exhibit a deeper hierarchy
reaching a maximum of five-level deep-nested hierarchy
This is the case in the Calendar2Organization mapping model
where the Division2Department mapping rule is made up of a mix of two consecutive HelperStatements and three mapping rules that cover the three scenarios
the generation of QVTo transformations for the second use case demonstrating more complex mappings is a stronger indication of the powerful HOTs
Validation was performed based on model-to-text tests classified by Tiso et al. (2013) as i) conformance tests
Taking into account the different types of model-to-text tests
we evaluated whether the generated QVTo transformations matched the expected QVTo transformations
we defined transformation test cases <MappingModel_file
where MappingModel_file represents the mapping model used to determine the links between elements of the source and target metamodels
and Exp_QVTo_file represents the expected QVTo transformations that we manually defined
the output of the HOTs (generated QVTo transformations) must match Exp_QVTo_file
we provide more details on the different types of model-to-text tests
Failed conformance tests can also be the result of incomplete or incorrect mapping models defined by the users
For instance the State2CompositeState mapping rule in Line 11 is invoked automatically
meaning that the user does not need to define the mapping rule that should be invoked when assigning the value of the source element to the target element
The only requirement is that the State2CompositeState mapping rule should be created by the user in the mapping model (a rule that has not been defined in the mapping model cannot be invoked)
The absence of this rule in the mapping model before executing the HOTs would result in an error
and the user would have to revisit the mapping model and define the State2CompositeState mapping rule
navigation between EClass Department and EClass Person can be accomplished through manager or secretary references
They are used to verify whether the textual elements of the generated model transformations have the required format
Errors discovered through these tests are not identified through conformance testing
Examples of textual testing involve checking whether the name of the transformation is the one inputted by the user or whether the names of the mapping operations are defined following the defined template
these tests verify whether the generated transformations adhere to the QVTo formatter (e.g.
On a side note, to increase the reliability of our approach, we complemented the aforementioned tests by manually defining target models that represent the expected outputs of the execution of the generated QVTo transformations. We compared the XML representation of the manually defined target models and the generated target models using the XML compare tool. As part of this process
it is important to leverage extensive input models
depending on the direction of the transformation)
that cover as many variations and combinations of concepts as possible
thus minimizing the likelihood of untested scenarios
While the generated target models were identical to the manually defined target models (with the exception of the line order)
it is still a valid concern whether the target models generated are correct and meet the requirements
In order to generate correct target models
two conditions must be met; i) the user must select “correct” mappings that illustrate the requirements
and ii) the HOTs should generate the expected QVTo transformations based on the input data (i.e.
While we have executed model-to-text test cases to validate the correctness of the HOTs
there is no guarantee that the user will select the “correct” mappings that conform to the requirements
Since mappings reflect the user intentions
we cannot provide guarantees on their appropriateness
meaning their reflection of the user's intentions
such a risk is apparent also on the traditional approach of manually writing model transformations
in addition to the limitations built in the MML by customizing the scope provider
the execution of the HOTs in cases where the mapping model is not correct (e.g.
the user is mapping an EClass to an EReference) will generate an error message
we can argue with certain confidence that MML contains all those concepts needed to specify deterministic unidirectional mappings between two Ecore-based DSMLs
and mapping models can then be effectively used to generate well-formed model transformations
In Figure 6 we provide an example of a mapping model between the excerpts of DSMLA and DSMLB illustrated in Figures 7, 8. In Figure 9 we illustrate the properties view for three mapping rules defined in Figure 6
to provide the reader with a more concrete example of the manual input that is required from the user in different cases
The requirements for transforming from DSMLA to DSMLB are as follows:
• StateMachine element in MDSMLA is transformed to StateMachine element in MDSMLB
being that a StateMachine element in MDSMLB must contain only one direct CompositeState element
if the StateMachine element in MDSMLA contains only one direct State
the latter is transformed to a CompositeState; alternatively if the StateMachine element in MDSMLA contains more than one direct State a new CompositeState element is created in MDSMLB
• A State element in MDSMLA is transformed to a SimpleState element in MDSMLB
if the State element in MDSMLA does not contain any other elements
• A State element in MDSMLA is transformed to a CompositeState element in MDSMLB if the State element in MDSMLA contains at least one element
Figure 9. Properties view for the mapping rules defined in Figure 6
while the operator is set to assign by default
The user would then define a new child mapping rule (i.e.
Both of these mapping rules fall under SC1
An interesting mapping rule falling under SC2
defined as a child mapping rule for the second StateMachine2StateMachine mapping rule
The null2top mapping rule creates a new CompositeState element in MDSMLB
In Figure 10 we present a more extensive excerpt of the Textual2Graphical mapping model for UML-RT state machines on the left-hand side and an excerpt of the generated QVTo transformation on the right-hand side
The generated QVTo transformation is the output of the execution of the HOTs that take as input DSML
There are a few peculiarities to highlight here
Textual to graphical mapping model and generated QVTo transformations for UML-RT state machines
The first StateMachine2StateMachine mapping rule generates Lines 12–13
being that the mapping model contains two mapping rules named StateMachine2StateMachine where sources and targets are identical
the HOTs generate three mapping operations
where one of them (Line 9) is a disjunctive mapping operation that disjuncts the other two (Lines 12 and 19)
The disjunctive mapping operation is invoked on Line 6 and the first matching candidate (i.e.
StateMachine2StateMachine0 or StateMachine2StateMachine1) is selected
HOTs determine the order in which the disjunctive candidates are printed based on whether they have a mapping condition
StateMachine2StateMachine0 on Line 12 has a mapping condition; thus
it is printed as the first disjuncted mapping operation on Line 10
To define the model transformations manually
the users would have to possess a strong understanding of these details
users would need to understand the syntax of the model transformation language (i.e.
the order in which the disjunct candidates appear)
and which rules to invoke in particular situations (e.g.
the user is only required to define two mapping rules
the HOTs would be able to automatically generate model transformations that conform to the QVTo syntax and include concepts that the user is not expected to understand
this reduces the amount of effort and expertise required
Lines 21–25 detail the generation of mappings from child mapping rules according to SC2 described in Section 4.2
thus the latter is added to the target model
Line 16 details the generation of a mapping from a child mapping rule according to SC1
the State2State mapping rule in the mapping model has been generated as two mapping operations: AbstractState2State in Line 47
because the State class in the target metamodel is abstract
as in the mapping model there are two rules (State2SimpleState and State2CompositeState) that fulfill the conditions to be disjuncted mapping operations (sources are identical
since SimpleState and CompositeState extend State in the target metamodel
the mapping operations in Lines 28 and 38 inherit the abstract mapping operation in Line 47
The example above further illustrates the reduced effort and expertise needed
as it eliminates the need for the user to comprehend the concepts of abstract mappings
HOTs reduce the likelihood of human-errors by automatically analyzing the involved DSMLs and mapping models
In this section we reflect on several aspects of our approach
We present the design principles that guided our solution
followed by an analysis of the benefits and an identification of the limitations and possible solutions
Three main principles have guided the design of the proposed solution
Separation of concerns: A strict separation between domain logic and implementation-specific details reduces complexity and allows for increased reusability
maintanability and extensibility of the solution
the definition of domain logic in a separate model (i.e.
using a language that provides a higher level of abstraction than model transformation languages
facilitates the understanding and solving of problems and ensures that software developers are not exposed to unnecessary information
Consideration of user's intentions: A complete and carefully written specification of how the DSMLs are mapped to one another forms the basis for producing complete and accurate model transformations
it is essential to provide the developer with the ability to capture his/her intentions in the form of unambiguous mapping links between elements of two DSMLs as semantics often needs human understanding to be correctly managed
Tooling that seamlessly integrates the target audience's current tool ecosystem: An essential aspect of successful software is the ability to seamlessly integrate with the environment that the target audience already uses
Our target modeling environment is the well-established Eclipse Modeling Framework
thereby our approach focuses on Ecore-based DSMLs and we opted for technologies (Xtend
etc.) that integrate seamlessly with the Eclipse environment
The adoption of a novel approach for the synchronization infrastructure between multiple notations for blended modeling can be challenging due to the customers' uncertainty of whether the benefits outweight the costs
Our proposed approach is therefore subjected to a cost-benefit analysis while simultaneously being compared with traditional approaches
we will discuss the transition costs involved
The transition costs consist of i) implementation costs and ii) training costs
Implementation costs are concerned with the adaptation of an organization's existing systems to integrate the proposed approach
Since our tool seamlessly integrates with EMF tools
a concern that companies with existing synchronization infrastructures would have to create the mapping models from scratch although they already have the synchronization infrastructure in place
the benefits of doing so outweigh the costs in a significant manner due to the following reasons
the co-evolution of the model transformations can be facilitated
as the respective changes can be made at a higher level of abstraction
in case DSMLA and DSMLB would represent two versions of the same language DSML (e.g.
the generated transformations would instead provide model co-evolution
allowing for user feedback prior to releasing a new version of the modeling tool
and verifying that the requirements of the users are understood and met
As part of our strategy to further reduce costs
we intend to use reverse transformations that build mapping models from model transformations
Training costs are instead concerned with the time and resources required to learn to utilize the proposed approach
Many industries are cautious to adopt new technologies that require a considerable amount of training and practice before they can be effectively implemented
when applications are implemented with a focus on user experience
we have minimized the training costs by designing the MML to be as simple and intuitive as possible
MML exhibits the following characteristics:
• Encapsulates the minimum set of concepts necessary for defining deterministic mappings
• Developed with a blended modeling approach to support textual and tree-based notations
which exhibit complementary usability features
• Syntactically similar to object-oriented programming languages
which pushes down the learning curve for the average software developer
• Raises the level of abstraction by allowing the user to focus on the domain's logic instead of dealing with lower-level model transformations
While users must become familiar with MML and while at first glance it may appear to be an additional overhead
it is in fact a one-time effort which proves beneficial in the long run
mapping models require significantly less input
resulting in lowers effort on the part of the developer
MML also enables domain experts without model transformation knowledge to be involved in the definition of the mapping models
since domain logic is presented in a format that is easily understood by all stakeholders rather than embedded in boilerplate model transformations
This could even reduce the time of development and number of errors caused by misunderstandings or miscommunications between domain experts and developers
The approach has also demonstrated acceptance and practical applicability in an industrial setting among HCL developers
who used it i) to define mapping models for generating the synchronization infrastructure between graphical and textual DSMLs
and ii) to co-evolve the mapping models and consequently the model transformation in response to changes in the textual DSML until the latter was refined to its present form
HCL is currently applying this approach for model co-evolution/migration purposes
To further decrease the learning curve and consequently
we have contributed with a tutorial to an established MDE community (ICSA conference)
and we plan on delivering a tutorial at a premier conference for practitioners and researchers interested in software architecture
we plan to contribute more examples and step-by-step tutorials to the online repository
As we presented a summary of potential costs and benefits associated with our approach
we would like to emphasize that the settings in which an organization operates is an instrumental factor in this analysis
The size of the involved DSMLs and the frequency of their evolution
can greatly impact the decision on whether the adoption of the approach is appropriate for the specific organization
Our first recommendation is for interested industrial parties to conduct their own cost-benefit analysis using this example as a guide
we recommend a gradual and step-wise adoption of the approach through the establishment of a multi-functional team staffed with both domain experts and software developers to investigate the integration and usability of the approach in their particular settings through our prototype
we have identified a number of limitations and potential solutions associated with the automatic generation of the synchronization infrastructure for blended modeling
While there could be an opportunity to incorporate the definition of bidirectional mappings
a proper balance must be found also with respect to the usability of the tool and the correctness of the generated model transformation with respect to the requirements
Interoperability. Another interesting point relates to the use of MML as a core artifact for interoperability between model transformation languages (Jouault and Kurtev, 2007)
MML is designed with a focus on generalizability; in our context
this is defined as the ability to use the same mapping model to drive the generation of model transformations in multiple model transformation languages
Generalizability is achieved by ensuring the separation of the domain-logic from implementation-specific details
whereas implementation-specific details are specified in the HOTs
which are specific to a transformation language
By providing an automatic generation of mapping models from existing model transformations
the generated mapping models could be used as input to other HOTs for generating model transformations conforming to other transformation languages
Although we cannot exclude that MML could need to be extended to support specific transformation languages
we are confident that the eventual changes would be relatively minor and only related to language-specific details that would require user input to generate correct transformations
In this work we assume that there exist mapping links between all elements of DSMLA and DSMLB
DSMLA) can have higher expressiveness than its counterpart (e.g.
it should be noted that the existence of mapping links for all elements cannot be guaranteed
when executing the transformation from DSMLA to DSMLB and then executing the reverse transformation
We assume that scenarios with disjoint DSMLs will generally be more affected by this phenomenon compared to simpler scenarios
This can be mitigated by leveraging in-place transformations that can propagate changes to the target model (which can be the same as the source model)
we presented our effort toward an approach for the automatic generation of synchronization model transformations to support blended modeling in an industrial setting
The resulting solution entails the provision of automatic means for the generation of model synchronization transformations across multiple notations of the same or different languages
This was accomplished by first designing and implementing a mapping modeling language (MML) in terms of an Ecore model
MML is instantiated in terms of mapping models
which define mapping relations between elements of two arbitrary Ecore-based DSMLs
Given the two DSMLs in terms of Ecore models and two mapping models (one per direction)
we provide a set of higher-order transformations (HOTs) that generate synchronization model transformations in QVT Operational between the two DSMLs
the generated transformations support two types of synchronization:
The DSMLs represent two notations of the same language (e.g.
then the generated transformations provide synchronization across different notations of the same language
then the generated transformations provide synchronization across different notations of different languages
Validation of the solution was performed by leveraging two use cases that represented the two aforementioned scenarios: synchronization across different notations of one language (UML-RT state-machines)
and synchronization across different languages (calendar and organizational structure use case)
the solution applied to UML-RT was deemed very promising by the engineers and tool architects at HCL
Several directions for future increments of this research and engineering effort have been identified
One direction is to tune the HOTs and leverage the proposed MML for the generation of co-evolution transformations to provide automated support for model co-evolution in response to metamodel evolution
Another interesting direction is related to the use of MML for interoperability between model transformation languages
we plan to provide automatic generation of mapping models from existing synchronization transformations
The generated mapping models can then be used as input to other HOTs for generating synchronization transformations conforming to other transformation languages than QVTo
The datasets presented in this study can be found in online repositories. The names of the repository/repositories and accession number(s) can be found below: https://github.com/MLJworkspace/BlendedModellingSolution
All authors listed have made a substantial
and intellectual contribution to the work and approved it for publication
This work was supported by Vinnova through the ITEA BUMBLE project (rn
MM was employed by the company HCL Technologies
The remaining authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest
All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations
Any product that may be evaluated in this article
or claim that may be made by its manufacturer
is not guaranteed or endorsed by the publisher
1. https://www.iso.org/standard/50508.html
2. https://www.hcltech.com/software/rtist
3. http://lightuml.sourceforge.net
4. https://github.com/ogheorghies/MetaUML
5. https://plantuml.com
6. https://jankoehnlein.github.io/FXDiagram/
7. https://www.jetbrains.com/mps/concepts/
8. https://eclipse.googlesource.com/emf/org.eclipse.emf/+/R2_8_3/plugins/org.eclipse.emf.mapping.ecore2ecore/
9. https://www.eclipse.org/modeling/emf/
10. https://www.eclipse.org/xtend/
11. https://wiki.eclipse.org/QVTo
12. https://github.com/MLJworkspace/BlendedModellingSolution
13. https://wiki.eclipse.org/OCL
14. https://www.eclipse.org/Xtext/
15. https://extendsclass.com/xml-diff.html
Blended graphical and textual modelling for uml profiles: a proof-of-concept implementation and experiment
CrossRef Full Text | Google Scholar
Google Scholar
CrossRef Full Text | Google Scholar
“Malan: a mapping language for the data manipulation,” in Proceedings of the Eighth ACM Symposium on Document Engineering (São Paulo)
Google Scholar
“A hybrid graphical and textual notation and editor for uml actions,” in European Conference on Model Driven Architecture-Foundations and Applications (Eschede: Springer),237–252
Google Scholar
Multi-view approaches for software and system modelling: a systematic literature review
CrossRef Full Text | Google Scholar
why and how,” in MPM4CPS Workshop (Munich)
Google Scholar
Blended modeling in commercial and open-source model-driven software engineering tools: a systematic study
Didonet Del Fabro
Towards the efficient development of model transformations using model weaving and matching transformations
CrossRef Full Text | Google Scholar
“Traceability mappings as a fundamental instrument in model transformations,” in International Conference on Fundamental Approaches to Software Engineering (Uppsala: Springer)
Google Scholar
“Every architecture description needs a framework: expressing architecture frameworks using ISO/IEC 42010,” in 2009 Joint Working IEEE/IFIP Conference on Software Architecture and European Conference on Software Architecture (Cambridge: IEEE)
Google Scholar
“Bridging emf applications and rdf data sources,” in Proceedings of the 4th International Workshop on Semantic Web Enabled Software Engineering SWESE (Karlsruhe)
Google Scholar
“Empirical assessment of mde in industry,” in Proceedings of the 33rd International Conference on Software Engineering (Waikiki
PubMed Abstract | Google Scholar
On the interoperability of model-to-model transformation languages
CrossRef Full Text | Google Scholar
“Mapping-based exchange of models between meta-modeling tools,” in Proceedings of the 14th Workshop on Domain-Specific Modeling (Portland
Google Scholar
“Blended graphical and textual modelling of uml-rt state-machines: an industrial experience,” in Accepted in ECSA Post-Proceedings (Berlin: Springer)
Google Scholar
“Towards automated support for blended modelling of uml-rt embedded software architectures,” in 15th European Conference on Software Architecture ECSA 2021
Virtual (originally Växjö)
Google Scholar
Integrating alf editor with eclipse uml editors
CrossRef Full Text | Google Scholar
Umple: model-driven development for open source and education
“vUML: a tool for verifying uml models,” in 14th IEEE International Conference on Automated Software Engineering (Cocoa Beach
Google Scholar
“Mapping specification in mda: from theory to practice,” in Interoperability of Enterprise Software and Applications (Berlin: Springer)
Google Scholar
“On integrating graphical and textual editors for a uml profile based domain specific language: an industrial experience,” in Proceedings of the 2015 ACM SIGPLAN International Conference on Software Language Engineering (Pittsburg
“A characterization of integrated multi-view modeling in the context of embedded and cyber-physical systems,” in Proceedings of the International Conference on Embedded Software
Synchronization of abstract and concrete syntax in domain-specific modeling languages
CrossRef Full Text | Google Scholar
“Messir: a text-first DSL-based approach for UML requirements engineering (tool demo),” in Proceedings of the 11th ACM SIGPLAN International Conference on Software Language Engineering
Google Scholar
“Textual modelling embedded into graphical modelling,” in European Conference on Model Driven Architecture-Foundations and Applications (Berlin: Springer)
Google Scholar
Google Scholar
Bidirectional model transformations in qvt: semantic issues and open questions
CrossRef Full Text | Google Scholar
“On the use of higher-order model transformations,” in European Conference on Model Driven Architecture-Foundations and Applications (Eschede: Springer)
Google Scholar
“A method for testing model to text transformations,” in AMT© MoDELS (Miami
Google Scholar
Ciccozzi F and Mohlin M (2023) Higher-order transformations for the generation of synchronization infrastructures in blended modeling
Received: 31 July 2022; Accepted: 15 December 2022; Published: 04 January 2023
Copyright © 2023 Latifaj, Ciccozzi and Mohlin. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY)
distribution or reproduction in other forums is permitted
provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited
in accordance with accepted academic practice
distribution or reproduction is permitted which does not comply with these terms
*Correspondence: Malvina Latifaj, bWFsdmluYS5sYXRpZmFqQG1kdS5zZQ==
Disclaimer: All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations
Any product that may be evaluated in this article or claim that may be made by its manufacturer is not guaranteed or endorsed by the publisher
94% of researchers rate our articles as excellent or goodLearn more about the work of our research integrity team to safeguard the quality of each article we publish
The historic "Ryburg" restaurant in Möhlin AG will soon be replaced by modern residential buildings
15 apartments and a commercial space are to be built on the site
known to residents of Möhlin AG as the "Nidli"
Möhliner Anec AG is planning the construction project
The plan is for 2.5 to 4.5-room apartments
Construction work is to begin on the current parking lot and the site of the restaurant
with one of the buildings being given an extension
The planning application has recently gone on public display and the plans have been agreed with the municipality's specialist commission for core zones
as the building is subject to volume protection
construction work could start as early as next winter
The "Ryburg" has an eventful history dating back to 1872
Originally known as the "Restaurant zur Pinte"
it was one of two such establishments in the village
The exact circumstances of the name change are unclear
but the inn remains an important part of local history
even if it will soon disappear from the townscape
This website is using a security service to protect itself from online attacks
The action you just performed triggered the security solution
There are several actions that could trigger this block including submitting a certain word or phrase
You can email the site owner to let them know you were blocked
Please include what you were doing when this page came up and the Cloudflare Ray ID found at the bottom of this page
3-year-old Golden retriever from Bemidji is the new mayor of Dorset
“Bear” Mohlin’s name was randomly plucked from the ballot box on Sunday
The only requirements were a full name and telephone number on the ballot
Candidates could buy their way into office by dropping $1 with their nomination in shoe boxes and mayonnaise jars located throughout the village of Dorset
Kvitne thanked everyone for their participation
“The Dorset Business Association would like to thank all past mayors for their distinguished services
The DBA also wishes great success to the outgoing mayor
Johnson promoted the Children’s Cancer Center
He’s also created a GoFundMe page to build a sandbox in the village
Taste of Dorset is an annual feast in the self-described “Restaurant Capital of the World.” This year’s theme was “Get Your Fix on Minnesota 226.”
I’ve had mahi mahi taco and ice cream and a pork-chop-on-a-stick for dinner to-go tonight because this isn’t going to be enough,” she said
Deanna and Gary Norman live in White Bear Lake
but their Big Sand Lake cabin that they bought in 1980 is their home away from home
They’ve been to quite a few Taste of Dorset festivals
“I can give him a $5 bill and he could make it last.”
Roy Schemmel slurped a towering brownie S’more ice cone
He didn’t know if the generous helping was a triple or double decker
When visitors had their fill of strawberry margaritas
Crayfish or minnow races were offered by the Dorset General Store
They experienced crayfish racing on Sunday
Her parents have a cabin on Little Sand Lake
The Dorset Business Association organizes the festival
He noticed that visitors stuck around after a 20-minute rain shower and even after the event concluded at 3 p.m
majority shareholder of Kuehne + Nagel International AG and Dr
attended the ground-breaking ceremony for the “Pharma Hub Switzerland”
This new facility will strengthen Kuehne + Nagel’s position as one of the leading logistics service providers for Swiss and international customers in the pharma and healthcare industry
The Pharma Hub will be part of the KN PharmaChain network
providing multimodal temperature-controlled door-to-door shipping on a global scale
enhanced visibility and traceability of the deliveries
as well as intensive and proactive risk management
The new Kuehne + Nagel logistics facility in Möhlin will provide 16,300 m2 of logistics space
which will be used to manage approximately 46,000 pallet spaces using a warehouse system that combines manual and automatic operation
Three different temperature zones will be available for the highly sensitive products: -20 degrees
hazardous goods and narcotics can be stored in dedicated areas
The Möhlin Pharma Hub will also focus on providing value-added services
The facility will provide end-to-end shipping and logistics services based on GxP-compliant standard
said: “Switzerland is an important pharma location with global reach
The cutting-edge facility in Möhlin is a strategic investment in this growth sector and supports us to serve our customers even better
especially through integration with the global Kuehne + Nagel pharma network.”
and perhaps we also have goals we want to achieve
Whether or not we succeed depends on how we act
The same applies in interactions between companies
“Game theory is a mathematically formulated theory of human interaction
It attempts to bring clarity to the jumble of interdependences between your actions and mine
and it can be used whether the motivation is altruism
Parts of the theory can even be applied to animal behavior,” Mohlin explains
He works at Lund University Department of Economics
But the theoretical models are often highly simplified
They frequently describe choice situations
with clearly defined alternative courses of action
and roughly how the players involved perceive the likelihood of various consequences
This is something that Mohlin wants to problematize
He wants to develop the theory by incorporating the way players see their surroundings
the way they simplify the world around them
and the social norms they are influenced by
but gives an example: Someone walking along the sidewalk who passes a person begging is expected to consider whether they should give the beggar some money
But the social norm does not require a person passing by on the opposite side of the street to consider giving money
People can be seen to exploit this by crossing the street before they get to the beggar
“They choose a situation where nothing is expected of them
But imagine if you don’t cross the street – you walk toward the beggar
Imagine the beggar gets up and gathers his things before you get there
Game theory needs to be developed so it can explain a wider range of phenomena of this kind
I think that both game theory and economics would be better if they paid more attention to how we simplify and categorize the world around us,” Mohlin says
Mohlin and his colleagues are not the first to research in this field
but their aim is to formulate a more uniform theory than the present one
He describes it as an extension of the change that took place in economics in the 1990s
“when economists realized that people are not only selfish”
“As long as economists confined themselves almost exclusively to markets
it was natural that game theory was mainly about how we reacted to the price of a product
But these days economists do almost anything
When they developed an interest in questions that were traditionally the realm of psychologists and sociologists
game theory needed to be able to describe norms
It may sound as though one needs to be a psychologist to research into social norms and people’s idea of reality
And Mohlin does indeed read literature in the fields of psychology and cognitive science as part of the project
“But you also have to realize that it’s about different things
It’s one thing for a psychologist to map all possible processes that help us to understand the social world around us
It’s another for an economist to see which of the processes play a part in the economic or social outcome.”
Mohlin hopes the methods he is developing will make it easier to communicate between the disciplines
so that economics can derive more benefit from psychology and sociology
He describes it as a kind of translation project:
we will develop analytical tools to make the explanations and predictions of game theory more accurate
it will then be easier to understand the world around us.”
When Mohlin went to university he chose first to study philosophy
After a while he transferred to Stockholm School of Economics
inspired by a friend already studying there
He thought it would be easier to get a job as an economist than as a philosopher
And studying economics also appealed to him because of his interest in politics
when all political discussions ended with ‘we can’t afford it’
I wanted to understand this: is it really true that we can’t afford it
My impression then was that politics was very much a question of economics
becoming a Wallenberg Academy Fellow represents financial security
I can spend up to 80 percent of my time teaching
This makes it very hard to find time for research
The grant enables me to halve my teaching time for five years
so during that period I won’t have to worry about a lack of funding.”
so the possibility of research occurred to him at a young age
Mohlin describes the work he does now as model-building – a creative process in which he pieces together various situations like a Lego house to see how they can be understood
but I personally have a tendency to dwell on the fundamental questions – the underlying principles
And as a researcher I’m entitled to do so.”
Knut and Alice Wallenberg Foundation has since its establishment in 1917 awarded over SEK 39 billion in grants
In 2024 the yearly grants to excellent basic research and education in Sweden was in total almost SEK 2.4 billion
[email protected]
After a norovirus outbreak at a kebab store in Möhlin
over 2,000 francs were collected to help it reopen
Two weeks after the norovirus outbreak in a kebab store in Möhlin AG
over 2,000 francs have been donated for the reopening of the store
The fundraising campaign, which was initiated by a loyal customer, has not only received local support, but has also attracted the attention of a prominent donor. This was reported by the "Aargauer Zeitung" newspaper.
launched it on the "Go Fund Me" platform to help the affected kebab store owner
The store had to close after 60 people complained of nausea and vomiting
which was later traced back to noroviruses in the homemade cocktail sauce
wanted to make it easier for the owner to restart
The fundraising campaign quickly exceeded the original target of 1,000 francs and has already reached 2,300 francs
Zumsteg was particularly surprised by an individual donation of 500 francs from Patrick Liotard-Vogt
Zumsteg thanked him via Instagram and received positive feedback
Despite some criticism on social media, Zumsteg received a lot of encouragement, especially on Tiktok, where a video from "Blick" publicized the campaign
His son was asked about the campaign at school
He plans to end the fundraising campaign soon and hand over the money collected to the kebab store owner
The date of the kebab store's reopening remains unclear as the owner was unavailable
One critical comment criticized the store owner's lack of public apology and suggested that the donations be given to charitable organizations
The editor wrote this article with the help of AI
About Rotary(down arrow opens sub-menu)>
Get Involved(down arrow opens sub-menu)>
Our Causes(down arrow opens sub-menu)>
Our Programs(down arrow opens sub-menu)>
News & Features(down arrow opens sub-menu)>
For Members(down arrow opens sub-menu)>
It’s just before dusk as Marcus Mohlin takes his seat at the rooftop restaurant of the Grand Palais hotel
the streets of the Cambodian capital are a tangled mess of traffic
where Audi and Range Rover SUVs jostle for space with three-wheeled rickshaws
the air is peaceful as people arrive for a meeting of the Rotary Club of Phnom Penh Metro
There’s a visiting Rotarian from Croatia and a representative of an NGO that supports Cambodian schools
A Japanese Rotarian makes a brief appearance and leaves a club banner
and interest in working together on a local project
a Swedish expat who is the club’s current president
says this is not unusual: His club has hosted 34 Rotary Foundation grants with a total budget of $1.2 million since it was chartered in 2004
“There are so many clubs reaching out to us because they want to do a global grant in Cambodia,” he says
“We’re overwhelmed by all the potential projects.”
It makes sense that Rotarians around the world are drawn to Cambodia
a country still recovering from years of civil war and a brutal dictatorship
at least 1.7 million Cambodians perished at the hands of the Khmer Rouge
a regime that forced people to labor in slavelike conditions on collective farms
the country remains one of the least developed in Asia
things are changing fast: Much of Phnom Penh is a giant construction site
with an influx of Chinese capital driving an unprecedented boom in high-rises
Yet the country still faces massive gaps in education
One-third of Cambodian children under five are stunted from a lack of adequate nutrition
and nearly a half-million do not attend school — some because of a lack of funds
others because their parents need them to work in the rice fields
Cambodia has one doctor for every 5,900 people
compared with roughly 1 for every 1,200 in neighboring Vietnam and Thailand
Half of rural households lack access to piped water or a well
Phnom Penh Metro — one of eight Rotary clubs in Cambodia and four in the capital — has worked to combat these challenges
in partnership with Veterans International Cambodia and the Rotary Club of Salisbury
supported Cambodians who have been injured by land mines — relics of the war period that still pose threats — or who have been affected by polio
which was endemic in Cambodia until the late 1990s
The club has hosted Foundation grants and organized other projects that have brought clean water to rural communities
equipped schools and health facilities with modern equipment
and helped restore eyesight for tens of thousands of Cambodians by providing eyeglasses and cataract operations
in partnership with the Rotary Club of Kobe East
it supported the training of Cambodian technicians to screen patients for cervical cancer
“Now it will be up to the same standards as everywhere else in the world,” says Steve Swartz
a retired surgeon and the club’s president-elect
The Phnom Penh Metro club has worked to combat Cambodia’s challenges
Although the Phnom Penh Metro club’s record of service has been consistent
its membership has gone through many iterations
Its origins are local: Founding President Ty Peseth is a prominent Cambodian businessman
most of its members are expatriates; several
are first-time Rotarians who felt a call to service after moving to Cambodia
But the club’s roster also includes Suon Sotheara
who was the charter president of a Rotaract club that the Rotarians sponsor
a Phnom Penh-based physician who was inspired to join because
I have the ability to help.” During his childhood in rural Cambodia
Mohlin says a primary goal of his presidency has been to boost club membership
with a particular focus on bringing in more Cambodian members
he introduced a program where members pay fees on a sliding scale and shifted meetings from weekly lunches to biweekly evenings at the Grand Palais
a boutique hotel that’s a short walk from Phnom Penh’s central market
Mohlin and other club members have traveled to Siem Reap
with members of the Rotary Club of Sentosa
which funded a pilot project there to build wells and equip homes with modern roofs
By purchasing materials from a Siem Reap-based manufacturer
the proj-ect also gave a small boost to the local economy
the group continues to discuss Cambodia’s many development challenges and ideas about how to overcome them
• This story originally appeared in the May 2020 issue of The Rotarian magazine
played with icy charisma by Charlotta Mohlin
Immortality explores the misogynistic underbelly of 20th century cinema
Immortality brings its themes to the fore in a truly captivating finale
Read more: Immortality review | An unforgettable, thought-provoking edit-’em-up
Get the latest issue here →Subscribe here →
Get the latest issue here →Subscribe here →
Apr 08, 2025 | magazine, film stories
Feb 19, 2025 | magazine, film stories
by Maria Lattila | May 05, 2025 | News, Reviews, TV
by Simon Brew | May 03, 2025 | Reviews, TV
by Ryan Lambie | May 02, 2025 | News, TV
by Maria Lattila | Apr 30, 2025 | News, TV
by Simon Brew | May 02, 2025 | Gaming, News
by Dan Cooper | May 02, 2025 | Gaming, News
by Maria Lattila | Apr 25, 2025 | Gaming, News
by Ryan Lambie | Apr 17, 2025 | Gaming, News
by Maria Lattila | Jul 30, 2024 | Feature, News
The Russos are back to direct two more Avengers films
by Simon Brew | Aug 14, 2024 | Feature
The fear of asking a question in case you look stupid
A few thoughts and words on that right here...
by Ryan Lambie | Sep 13, 2024 | Gaming
indie publisher Annapurna Interactive has been rocked by the sudden resignation of its entire..
by James Harvey | Nov 27, 2024 | Feature
production designer and puppet team leader to find out just what makes Aardman's poultry-packed sequel tick...
we and our partners use technologies like cookies to store and/or access device information
Consenting to these technologies will allow us and our partners to process personal data such as browsing behavior or unique IDs on this site and show (non-) personalized ads
may adversely affect certain features and functions
Click below to consent to the above or make granular choices. Your choices will be applied to this site only. You can change your settings at any time
or by clicking on the manage consent button at the bottom of the screen
Please select what you would like included for printing:
Copy the text below and then paste that into your favorite email application
This site is protected by reCAPTCHA and the Google Privacy Policy and Terms of Service apply
Service map data © OpenStreetMap contributors
Ulla Engbergulla.engberg@sverigesradio.se
Daily news podcast Mon - Fri at 4.30pm, plus weekly summary on Fridays at 4.30pm on P2 (P6 89.6FM in Stockholm), repeated on Mondays at 4.30pm on P6.
Kontakta gärna Sveriges Radios forum för teknisk support där vi besvarar dina frågor vardagar kl. 9-17.
Experience of suicide: Teaching people to deal with suicidal individuals
Occupation: National instructor at the Swedish Migration Agency’s Detention Training Unit
first published in the magazine Medicinsk Vetenskap No 3/2020
“We do see some mental illness in the Swedish Migration Agency’s detention centres
Those who have served a prison sentence and are awaiting deportation are used to being locked up and are generally better equipped to deal with their time in detention than those who have been found in the country without a residence permit or who have no identity documents
We still perform suicide screening on all new arrivals
I train case officers and supervisors at Migration Agency detention centres
We try to ensure that we have two instructors when we teach courses in Emergency Care of a Suicidal Person (AOSP)
given that many of the course participants have their own experiences of mental illness.
We discuss how to identify behavioural changes in suicidal individuals
such as changes in appetite and circadian rhythm or that they stop using the internet or don’t want to go outside
Someone who has been very depressed and appears to be recovering may have decided to take their own life and found inner peace
We sometimes have language deficiencies and must therefore be good at reading body language and behaviour
We also teach how to get people to talk about how they are feeling and provide various keys that can help people out of a given state of mind
Participants practice various scenarios with their backs to one another
one playing the suicidal [inmate] and the other the staff member
they need to find a genuine interest in what the person might need or want to share; however
the most important thing is to have the courage to deal with the issue
to approach someone who looks like they are struggling and actually ask how they are feeling.”
Education
Doctoral education
Research
About KI
Student
Calendar
Ladok
Canvas
Schedule
Student e-mail
Course and programme websites
Student at KI
Staff portal
Contact and visit Karolinska Institutet
University Library
Support research and education
Jobs at KI
Contact the press Office
Karolinska Institutet171 77 StockholmPhone: +46-(8)-524 800 00
Org.nr: 202100-2973VAT.nr: SE202100297301About this websiteAccessibility report
© Karolinska Institutet - a medical university
Volvo’s V8 Supercar may not race any more but the S60 fights on
It’s much different to the car we saw in Australia – and can it win the title this year
THE VOLVO S60 Supercar might be gone from Australian racetracks but in the memories of many – particularly someone like Scott McLaughlin – its legend lives on
From the time the car first appeared and made an impression in Adelaide in 2014 (the ‘Jandal’ race) to its final bow at Homebush last December the Swedish missile made more than its presence felt
It looked different to the other Supercars; it sure as heck sounded different; unlike any pushrod V8 motor
and more like two 250cc four-cylinder motorbikes fighting for the lead in a race
That made it a prime candidate to be a car at which AA looked for Under the Skin
which made writing about it a fraction… challenging
But the S60 only stopped racing in Australia
In the World Touring Car Championship the car – or versions of it – now races against the might of
because to anyone used to the rumble and the roar of a V8-powered Supercar
a four-cylinder shopping trolley with wings looks a bit… well
And in spite of its title status the WTCC has just not caught fire the way that the FIA hoped that it might have
But a world title is clearly a lure for makes – well
Volvo announced its WTCC project 30 years to the day after it won the 1985 European Touring Car Championship title with the Volvo 240 Turbo
“We have chosen the FIA WTCC as our global motorsport platform as it combines the cutting edge technology of our Volvo cars with exciting racing all over the world,” said Niels Möller
“The championship enables us to further develop our technology and utilise development from the racing circuits directly to our products for performance oriented Volvo customers.”
Before the company went public work had already started on the Volvo S60 Polestar TC1 race car
Polestar Cyan Racing’s Technical Director
“Volvo started with this project at the beginning of 2014,” he explained
“The target was to have a car ready [to race] by the end of 2015 but that target became 2016
The development was a bit up and down but it was soon understood how big the undertaking was
It was not many months before the debut of the car was postponed.”
It took only two rounds to get an S60 onto the podium
but it was well into the second half of the season that Nicky Catsburg managed to score a maiden win
By year’s end Catsburg was the best-placed Volvo driver
Thed Bjork was 20th and Volvo finished fourth
in the Manufacturers’ Championship – and was the only make to go the whole season without being required to carry ‘success ballast’
under the skin the story is interesting – and so is the series itself
Volvo S60’s four-cylinder B4204T6 motor
The WTCC is based on what was going to be an engine unit used across nearly all categories of motorsport
The initial idea was that the ‘global’ engine would be a 1.6-litre four-cylinder
with the basic unit being used in Formula 1 (with turbocharging and Energy Recovery/Hybrid systems); Sportscar racing (ditto
but in a different specification to differentiate with F1’s IP); Touring Cars and Rallying (with turbocharged units); and even entry-level racing
with performance limited by how much turbocharging
The one-size fits all racing motor was a great idea
save for one small but telling factor; pretty much no one outside of the FIA was vaguely interested in it
which since the middle of the last century has built a business model on charging its customers outrageous amounts of money for road cars powered by V12s
but which had absolutely no interest in making a four-cylinder car
F1’s engineers would eventually settle for a V6 with similar displacement and technical capabilities to the proposed four
to look elsewhere for their forms of motivation
But the ‘global’ engine concept did find a home in the World Rally Championship – and the WTCC
As there is in the British Touring Car Championship
have the option of using a generic motor but Volvo chose not to
“To use the Volvo engine as base and not go for the available generic WTCC race engine for our car was a no-brainer,” explained Mattias Evensson
“Volvo engines are suitable for racing purposes in terms of power
The Drive-E engine is a perfect example with its low friction technology and exotic materials.”
So under the bonnet is a racing unit is based on Volvo’s four-cylinder B4204T6 motor
And it was here that AA struck a little hurdle
The S60 is effectively a works racing car and as such
there is not a huge amount of information available to share with the media…
“It is effectively the four-cylinder standard engine,” says Mohlin
“It is from 1.6 up to 2 litres [in production form] and it has been adapted to run [in the race car] at 1.6-litre size
I don’t know what the piston dimensions are exactly but they are not far away [from standard]
“The injectors are direct injection and that is specific for each engine
The turbocharger is the same one for everybody
and everybody is allowed to run an ALS system.”
The Anti Lag System consists of a valve that opens when the throttle is closing
where is it mixed with the unburnt fuel being expelled
That prompts the turbocharger to maintain its speed and therefore
whether with ‘bespoke’ or generic engines
share a common turbocharger supplied by Garrett
Engine revs are capped to 8500rpm; minimum engine weight is 82kg
pistons and some others – are subject to weight minimums
the engine rules are a carryover from what was Super2000
Formerly the engines were controlled by a 33mm air restrictor; that was changed to 36mm
That does not sound like much but the effect was such that power outputs that had been just over 300hp leapt immediately to 380hp and
And there is one other important change; barring unexpected occurrences like crash damage
nowadays each driver has to do a whole season on a single motor
“You can count on that [distance] being more than the Practice
While we are accustomed to seeing Touring Cars – in our case
Supercars – built around a purpose-built racing chassis that is not the scenario in the WTCC
the bodyshell is mainly that of a normal road car,” says Mohlin
“You have areas that you are allowed to change
and that is mainly because of historic reasons
In the past there was a lot of dependence on the basic layout of the car.”
The S60 is the biggest car on the WTCC grid
So the car is built around the base S60’s steel bodyshell
but with various panels reproduced in carbon fibre – including the roof
But at either end there are major differences
“The wheelbase can be slightly modified but the rest of the car’s dimensions are standard,” says Mohlin
“There is no limit per se for the overall size
though the maximum width allowed is 2m – and that is no problem for any of the cars
“The S60 is a large car but is was thoroughly assessed in the beginning
It was quickly decided that we wanted to race the sedan [version] and the same is there with the Citroen C4.”
The racing version of the S60 is the same length as the road car – but that does not dictate that the body shape is absolutely identical
“In the regulations you have the possibility to move the axis of the wheels so that you can elongate the wheelbase a little bit,” Mohlin explains
“You can extend the body front and rear – larger bumpers – and you can move the wheel axes out from the middle.”
To limit the amount of money that teams can go through refining suspension design
all cars must run a simple and similar system
“With the new regulations it was decided that all the cars should have MacPherson struts
regardless of what the car’s original configuration was
You are allowed to have the ‘standard’ shell but you can modify the front mounts at the top and the bottom
It is not the same for gearboxes – and the gearbox is not a control item
though the teams have met common ground with Xtrac’s P1046 unit
“You can run whatever gearbox you want,” says Mohlin
“In reality everyone is running the same Xtrac gearbox as used in the BTCC
The only limit on the gearbox is that you are allowed to have two different gearsets to change during the season
You can only use a mechanical differential and no ramps
The tyre size is 250/660/18; Supercars tyres are wider and higher
“The cars are a bit different to some; they are quite heavy and they have quite a bit of power so the tyres need to be quite robust
and I am not certain where that is by choice or circumstances
They fall off quite a lot; they are really quick in Qualifying but they drop off quite a lot.”
Supercars have a control wheel supplied by Rimstock; in the WTCC the sizes are identical but you can source them from anywhere
Mohlin explains; “There is a minimum weight for the wheels and they have ideas that the wheels will be a way of controlling cost – for some reason
You are allowed to use as many [studs] as you want
but I think that you can use as many as you want
The minimum weight for the wheels is 10 kilos
and you are not allowed to have any variation
The brand and design are free – but the reality is that everyone has the same wheels…
but what drives the cost is you need to have a lot of them in circulation
We would have five sets per car per weekend
you can have eight or nine sets for weekend.”
On top of the wheels is the suspension and here
Teams and manufacturers work together to finalise the specifications for a three-year period – and that three-year window is the same for everyone
That has an impact on suspension design – though it does not have to be the same for everyone
“Every third year you homologate the car,” says Mohlin
the subframe and the suspension pickup points
As long as you are able to use the same uprights you can alter the design of the damper
but I think that all of the competitors use the same design for the three years
“MacPherson struts are tricky and we are quite keen to have an integrated design
That limits what you can do to the damper anyway
because you always have the same upright.”
That means Ohlins are in use across most of the teams – and very understandably so in the case of Volvo
since they are both based in Sweden…
I have to ask… Volvo is clearly keen on the WTCC as a showcase for its cars – and it continues to compete
as a man who is clearly enthusiastic about all things racing
“I think the Supercar looks nicer because there is something in the regulations that makes the WTCC look not as good
The shape of the fenders are dictated by the rules and when you homologate the car
you are fixing in the things like the floor
“The cars do not look as good as they might look
and that is the difference between the Supercar and the WTCC car.”
but the competition it provided went down a treat with many of the fans
As this edition went to press Catsburg and his Cyan Polestar Volvo were leading the WTCC
How would people cope with the notion of Volvo
Type: Four-door sedan front-wheel drive [rear-wheel drive]
modified by Cyan Polestar Racing [Control spaceframe chassis by Pace Innovations
Weight: 1100kg including driver (+ success ballast) [1410kg including driver]
Engine: 4-cylinder Volvo B4204T6 turbocharged [60-degree V8 Volvo B4204T6 normally aspirated]
Max Power: 390hp @ 7000rpm [645hp @ 7200rpm]
Limited Slip Diff [Albins ST6 6-speed transaxle
380mm rotor [Control AP Racing 6-Piston Caliper
Rear: AP Racing 2-Piston Caliper 280mm rotor [Control AP Racing 4-Piston Caliper 355mm rotor]
design homologated for three seasons [F and R: Double-wishbone
suspension geometry designed by teams/suppliers]
five-stud [Control Rimstock 18” x 11” forged aluminium
Yokohama Slicks/Wets 250/660/18 [Dunlop Slicks/Wets 290/675/18]
Get the latest updates delivered directly to your inbox
You have successfully joined our subscriber list
Caroline Garde[email protected]
Where can I find Auto Action Premium? Click here
In Sweden there are four adoption agencies
they are approved by the state and are responsible for all foreign adoptions to the country
Through the years they have placed thousands of children to heterosexual couples in Sweden
However when it comes to gay couples it’s a rather different story
QX has been in contact with all of Sweden’s adoption agencies and none of them have ever passed a child to a gay couple
– We do not find countries or organizations to suit certain groups in society
We focus our partnerships in places where there is a need for international adoption
Here in Sweden we might think that some countries have too many and strict requirements
but that’s just something that we’ll have to accept
With strict requirements Adoption Center is referring to the fact that several countries do not approve of gay adoptions
However it is Adoption Center that determines which countries they will cooperate with
and today they do not cooperate with any contacts that accept gay adoptions
The result is that gay couples are unable to adopt children despite the Swedish legislation
When QX ask Adoption Center what they think about the fact that no international gay adoptions have ever been made
According to Helene Mohlin their foreign contacts can take offense if they take a stand on this particular issue
do not think that this will come true through international adoptions
instead many gay couples are trying surrogacy
An orphaned child shall first be given a new mom and dad
Otherwise Sweden’s credibility as an adoption country could be questioned
However according to Adoption Center this is not their view anymore
Since the law on gay adoptions was passed in 2002 Adoption Center has changed their values
– We changed our values on this subject a long time ago
Today we welcome same-sex couples just like everyone else to apply for adoption
regnbågs-communityts egen röst med månadstidningen QX och nyhetstidningen qx.se som bevakar det samhälle vi lever i och den kultur och de människor vi bryr oss om
Qruiser.com är den digitala mötesplatsen och i QX Shop finns en mängd identitetsstärkande varor
Vi arrangerar i samarbete med andra aktörer regelbundet event där QX Gaygala utgör kronan på verket
In the new logistics facility to be built in Möhlin
46,000 pallet spaces on a usable area of approx
16,300 square meters with a combination of automatic and manually operated storage systems
Three different temperature ranges are available for the sensitive pharmaceutical products: for -20 degrees
Hazardous goods and narcotics can also be stored in separate areas
Kuehne + Nagel wants to underpin its position as one of the leading providers of logistics services for customers from the Swiss and international pharmaceutical and healthcare industry
which enables global multimodal temperature-controlled door-to-door transportation
delivery traceability and intensive risk management
The location will also be designed to provide value-added services
With around 70,000 employees at more than 1,300 locations in over 100 countries
Kuehne + Nagel claims to be one of the world's leading logistics service providers
contract logistics and overland transportation
Luke Eisner won people's hearts after his excellent portrayal of Stig Mohlin in Tall Girl
He is a charming young man who also sings and is a model
Fans hope to see him in more television shows and movies
PAY ATTENTION: Click “See First” under the “Following” tab to see Legit.ng News on your Facebook News Feed!
The actor attends the premiere of Netflix's 'Tall Girl' at Netflix Home Theater in Los Angeles, California. Photo: @JC OliveraSource: Getty ImagesLuke Eisner shot into fame after featuring in Tall Girl
a 2019 American teen romantic comedy film directed by Nzingha Stewart
Luke Eisner is a multi-talented young man. He is a model, band musician, and model with a bright future ahead
PAY ATTENTION: Share your outstanding story with our editors
Please reach us through info@corp.legit.ng
Luke Eisner's age is 26 years as of 2022
The actor attends the premiere of Netflix's "Klaus" at Regency Village Theatre in Westwood, California. Photo: @Jon KopaloffSource: Getty ImagesWhere is Luke Eisner from?The actor is from Greendale
The actor was raised in a middle-class family in Wisconsin, United States of America. As a child, he wanted to become an author because books fascinated him
His father, Bill, was a branding strategist, advertising and marketing executive, and consultant. Bill lost his six-year battle with cancer in 2019
his paternal grandmother's name is Elaine
The actor went to Greendale High School. He graduated from the institution in 2014. He then moved to Los Angeles to study at the University of Southern California's Thornton School of Music
The young man started his career in early teenagehood
his cousins ran a big wedding DJ company named Sound by Design in Wisconsin
He then registered his own company named Luke Eisner DJ and Lighting LLC
This bold move saw him being named the Wisconsin Young Entrepreneur of the Year by Ernst & Young in 2013
he represented Teens Against Tobacco Use (TATU)
He also provided free musical performances for homebound soldiers
His modelling career started after a modelling scout spotted him in his home area
He is represented by renowned agencies, including IMG Models, Vision Models, Elite Model Management, and Modelwerk Agency. He has worked with reputable brands
whom he first met at the University of Southern California
Voilá has opened concerts for artists and bands
In 2019, Luke started his acting career. He has seven acting credits, as listed below. His role in Tall Girl catapulted him into the limelight. He coached for months with renowned accent coach Erik Singer on the Swedish accent to prepare for this role
He performed I've Never Been in Love Before
and Figure You Out in Tall Girl and Tall Girl 2
The model's girlfriend is Kirby Johnson. Luke Eisner and Kirby Johnson have been together since 2018. They often post pictures together on various social media platforms. They also have a joint YouTube channel, Kirby & Luke Diaries
Luke Eisner's height is 6’ 3” or 191 centimetres
He has a slim frame and weighs about 172 pounds or 78 kilograms
He wears shoe size 12 (UK)/ 13 (US)/ 46 (EU) and has blond hair and blue eyes
Luke Eisner is a man of many talents. He can act, sing, and model. The young man has achieved a lot in the last few years and has a promising future in the entertainment industry
READ ALSO: Tarayummy’s biography: age, height, real name, ethnicity, boyfriend
Legit.ng recently published Tarayummy’s biography
and social media personality from the United States of America
Tarayummy came into the limelight after she started posting her modelling pictures on Instagram and uploading videos on YouTube