Complaint about my code-generation strategy to implement many-to-many associations in relational databases
More specifically, the complaint was that in the translation of this UML class diagram:

MoDisco featured at InfoQ
EU’s SPEEDS project comes to a successful conclusion
TextUML Toolkit 1.6 RC1 is now available
TextUML Toolkit 1.6 RC1 is now available! You can install it using the Marketplace Client or by pointing Eclipse to the update site:
http://abstratt.com/update
If you find any problems installing this build, please let us know asap so it can be addressed before the final release.
New features
Much of the work in this release went into improving the model building infrastructure to be even more notation agnostic. That work is still ongoing and should be completed in 1.7. But there were plenty of user-facing feature additions as well:
- preconditions on operations (2986923 and 3002571)
- support for a default notation (so file extensions can be optional) (2995372)
- support for implicitly applying profiles/stereotypes (so models are less verbose) (2981580)
- support for derived properties (2928428)
- support for initial values in properties (2115439)
- advanced features (closures, constraints) are now implemented using profiles instead of metamodel extensions (2933692)
In other news
The reason it took so long for a new TextUML Toolkit release to come about was that I have been busy working on AlphaSimple, which went on public beta today. AlphaSimple is an online tool for domain-driven prototyping that currently uses TextUML as modeling notation. Thus, AlphaSimple is also the driving force behind most of the changes that happened in the 1.6 cycle, and you can try them right away by starting a guest session and studying the example projects.
Agile teams are more likely to model than traditional teams
Modeling Safe Interface Interactions in Web Applications
UML class diagrams easier to understand than ER diagrams
The paper presents the results of three sets of controlled experiments aimed at analyzing whether UML class diagrams are more comprehensible than ER diagrams during data models maintenance. -->
Metastorm Partner iOctane Signs License Agreement with the New Zealand Department of Internal Affairs
Rule Interchange Format is now a W3C Standard
W3C announces the publication of a new standard for building rule systems on the Web: Rule Interchange Format (RIF).
The goal of RIF is to facilitate the interoperation and interchange between various rule languages and rule engines. RIF defines a common XML serialization language for several declarative rule languages (like OMG SBVR, OMG PRR, SWRL, a subfamily of RuleML, etc)
-->Short introduction to the MoDisco Model-Driven Software Modernization Framework
Quality of ATL model transformations (help needed)
Marcel van Amstel and Mark van den Brand are working on measuring the quality of ATL model transformations.
-->Implications of limited understanding in code..
Looking for new members for the AtlanMod team
AtlanMod research team at École des Mines de Nantes is opening new positions for postdoctoral fellows or research engineers in the field of Model-Driven engineering (including topics such as: model transformation, metamodeling, DSLs, UML, SBVR, business rules, model weaving, model management, reverse engineering and any other model-related aspects)
-->90% of the people using MDD will/would repeat
Re: [UML Forum] Difference between InstanceSpecification and LiteralSpecification
This mechanism is the one sued to describe ADTs (see my response to your
other message) when they are complex data structures with discrete,
different, identifiable members. Thus an ADT of ComplexNumber actually
represents a data structure of {real, imaginary} where 'real' and
Re: [UML Forum] Primitive Type musn't be a DataType
What OMG is trying to describe is what everyone else calls and Abstract
Data Type (ADT). That is, a data structure that can be hidden behind a
scalar knowledge attribute. Thus the details of the data
implementation's complexity are not of interest at the level of
abstraction of the containing subsystem so they are hidden. The use of
RE: [UML Forum] Primitive Type musn't be a DataType
Can anyone provide a tutorial that explains the classes and diagrams
step by step?
Regards,
Sanjay Amin
Behalf Of Timothy Marc
Sent: Friday, July 09, 2010 1:40 PM
To: umlforum@googlegroups.com
Hi again,
another question regarding PrimitiveTypes. With respect to the UML2
spec, PrimitiveTypes may not have any logical structure. That means,
[UML Forum] Primitive Type musn't be a DataType
another question regarding PrimitiveTypes. With respect to the UML2 spec,
PrimitiveTypes may not have any logical structure. That means, they won't
have operations and attributes within UML (the spec mentiones clearly that
operations and structures may be defined outside of the scope of UML). But










