Stefan Naujokat

Semantics-based composition of EMBOSS services

BackgroundMore than in other domains the heterogeneous services world in bioinformatics demands for a methodology to classify and relate resources in a both human and machine accessible manner. The Semantic Web, which is meant to address exactly this challenge, is currently one of the most ambitious projects in computer science. Collective efforts within the community have already led to a basis of standards for semantic service descriptions and meta-information. In combination with process synthesis and planning methods, such knowledge about types and services can facilitate the automatic composition of workflows for particular research questions.ResultsIn this study we apply the synthesis methodology that is available in the Bio-jETI workflow management framework for the semantics-based composition of EMBOSS services. EMBOSS (European Molecular Biology Open Software Suite) is a collection of 350 tools (March 2010) for various sequence analysis tasks, and thus a rich source of services and types that imply comprehensive domain models for planning and synthesis approaches. We use and compare two different setups of our EMBOSS synthesis domain: 1) a manually defined domain setup where an intuitive, high-level, semantically meaningful nomenclature is applied to describe the input/output behavior of the single EMBOSS tools and their classifications, and 2) a domain setup where this information has been automatically derived from the EMBOSS Ajax Command Definition (ACD) files and the EMBRACE Data and Methods ontology (EDAM). Our experiments demonstrate that these domain models in combination with our synthesis methodology greatly simplify working with the large, heterogeneous, and hence manually intractable EMBOSS collection. However, they also show that with the information that can be derived from the (current) ACD files and EDAM ontology alone, some essential connections between services can not be recognized.ConclusionsOur results show that adequate domain modeling requires to incorporate as much domain knowledge as possible, far beyond the mere technical aspects of the different types and services. Finding or defining semantically appropriate service and type descriptions is a difficult task, but the bioinformatics community appears to be on the right track towards a Life Science Semantic Web, which will eventually allow automatic service composition methods to unfold their full potential.

Loose programming with PROPHETS

Loose programming is an extension to graphical process modeling that is tailored to automatically complete underspecified (loose) models using a combination of data-flow analysis and LTL synthesis. In this tool demonstration we present PROPHETS, our current implementation of the loose programming concept. The first part of the demonstration focuses on the preparative domain modeling, where a domain expert annotates the available services with semantic (ontological) information. The second part is then concerned with the actual loose programming, where a process modeler orchestrates the services without having to care about technical details like correct typing, interface compatibility, or platform-specific details. The orchestrated process skeletons are treated as loose service orchestrations that are automatically completed to running applications.

CINCO: a simplicity-driven approach to full generation of domain-specific graphical modeling tools

Even with the help of powerful metamodeling frameworks, the development of domain-specific graphical modeling tools is usually a complex, repetitive, and tedious task, which introduces substantial upfront costs often prohibiting such approaches in practice. In order to reduce these costs, the presented Cinco meta tooling suite is designed to provide a holistic approach that greatly simplifies the development of such domain-specific tools. Our solution is based on the idea to apply the concept of domain specialization also to the (meta-)domain of “domain-specific modeling tools”. Important here is our focus on complex graph-based models, comprising various kinds of nodes and edges together with their individual representation, correlations, and interpretation. This focus allows for high-level specifications of the model structures and functionalities as the prerequisite for push-button tool generation.

Property-driven benchmark generation: synthesizing programs of realistic structure

We present a systematic approach to the automatic generation of platform-independent benchmarks of realistic structure and tailored complexity for evaluating verification tools for reactive systems. The idea is to mimic a systematic constraint-driven software development process by automatically transforming randomly generated temporal-logic-based requirement specifications on the basis of a sequence of property-preserving, randomly generated structural design decisions into executable source code of a chosen target language or platform. Our automated transformation process steps through dedicated representations in terms of Büchi automata, Mealy machines, decision diagram models, and code models. It comprises LTL synthesis, model checking, property-oriented expansion, path condition extraction, theorem proving, SAT solving, and code motion. This setup allows us to address different communities via a growing set of programming languages, tailored sets of programming constructs, different notions of observation, and the full variety of LTL properties—ranging from mere reachability over general safety properties to arbitrary liveness properties. The paper illustrates the corresponding tool chain along accompanying examples, emphasizes the current state of development, and sketches the envisioned potential and impact of our approach.

Service-Oriented Mediation with jETI/jABC: Verification and Export

The paper presents how we solved the mediation challenge in a model driven, service oriented fashion, how we verify properties of the mediator via model checking in the jABC, and how to systematically export jABC/jETI orchestrated services as Web services. Due to the lack of maturity of the involved environments and external components, the latter task is less easy and the solutions possible today are less stable than one would expect from these technologies.

Simplicity-first model-based plug-in development

In this article, we present our experience with over a decade of strict simplicity orientation in the development and evolution of plug‐ins. The point of our approach is to enable our graphical modeling framework jABC to capture plug‐in development in a domain‐specific setting. The typically quite tedious and technical plug‐in development is shifted this way from a programming task to the modeling level, where it can be mastered also by application experts without programming expertise. We show how the classical plug‐in development profits from a systematic domain‐specific API design and how the level of abstraction achieved this way can be further enhanced by defining adequate building blocks for high‐level plug‐in modeling. As the resulting plug‐in models can be compiled and deployed automatically, our approach decomposes plug‐in development into three phases where only the realization phase requires plug‐in‐specific effort. By using our modeling framework jABC, this effort boils down to graphical, tool‐supported process modeling. Furthermore, we support the automatic completion of process sketches for executability. All this will be illustrated along the most recent plug‐in‐based evolution of the jABC framework, which witnessed quite some bootstrapping effects. Copyright

DIME: A Programming-Less Modeling Environment for Web Applications

We present DIME, an integrated solution for the rigorous model-driven development of sophisticated web applications based on the Dynamic Web Application (DyWA) framework, that is designed to accelerate the realization of requirements in agile development environments. DIME provides a family of Graphical Domain-Specific Languages (GDSLs), each of which is tailored towards a specific aspect of typical web applications, including persistent entities (i.e., a data model), business logic in form of various types of process models, the structure of the user interface, and access control. They are modeled on a high level of abstraction in a simplicity-driven fashion that focuses on describing what application is sought, instead of how the application is realized. The choice of platform, programming language, and frameworks is moved to the corresponding (full) code generator.

Archimedean Points: The Essence for Mastering Change

Explicit Archimedean Point-driven (software) system development aims at maintaining as much control as possible via ‘things’ that do not change, and may radically alter the role of modeling and development tools. The idea is to incorporate as much knowledge as possible into the tools themselves. This way they become domain-specific, problem-specific, or even specific to a particular new requirement for a system already in operation. Key to the practicality of this approach is a much increased ease of tool development: it must be economic to alter the modeling tool as part of specific development tasks. The Cinco framework aims at exactly this kind of ease: once the intended change is specified, generating a new tool is essential a push button activity. This philosophy and tool chain are illustrated along the stepwise construction of a BPMN tool via a chain of increasingly expressive Petri net tools. By construction, the resulting BPMN tool has a conceptually very clean semantic foundation, which enables tool features like various consistency checks, type-controlled activity integration, and true full code generation.

Property-Driven Benchmark Generation

We present a systematic approach to the automatic generation of platform-independent benchmarks of tailored complexity for evaluating verification tools for reactive systems. Key to this approach is a tool chain that essentially transforms a set of automatically generated LTL properties into source code for various formats, platforms, and competition scenarios via a sequence of property-preserving steps. These steps go through dedicated representations in terms of Buchi Automata, Mealy machines, Decision Diagram Models, Code Models, and finally the source code of the chosen scenario. The required transformations comprise LTL synthesis, model checking, property-oriented expansion, path condition extraction, theorem proving, SAT solving, and code motion. This combination allows us to address different communities via a growing set of programming languages, tailored sets of programming constructs, different notions of observation, and the full variety of LTL properties – ranging from mere reachability over general safety properties to arbitrary liveness properties. The paper illustrates the whole tool chain along accompanying examples, emphasizes the current state of development, and sketches the envisioned potential and impact of our approach.

Domain-Specific Code Generator Modeling: A Case Study for Multi-faceted Concurrent Systems

In this paper we discuss an elaborate case study utilizing the domain-specific development of code generators within the Cinco meta tooling suite. Cinco is a framework that allows for the automatic generation of a wide range of graphical modeling tools from an abstract high-level specification. The presented case study makes use of Cinco to rapidly construct custom graphical interfaces for multi-faceted, concurrent systems, comprising non-functional properties like time, probability, data, and costs. The point of this approach is to provide user communities and their favorite tools with graphical interfaces tailored to their specific needs. This will be illustrated by generating graphical interfaces for timed automata TA, probabilistic timed automata PTA, Markov decision processes MDP and simple labeled transition systems LTS. The main contribution of the presented work, however, is the metamodel-based domain-specific construction of the corresponding code generators for the verification tools Uppaal, Spin, Plasma-lab, and Prism.