Gold sponsors

Intel logo

IBM Corporation

Cisco

JBoss

Sonatype

SAP

Oracle

Silver sponsors

Blackberry

agitar

bsi logo

Microsoft

Google

Instantiations

Soyatec

Xored

amazon

Actuate

Bronze sponsors

Purple Scout

froglogic

Paremus

objectivity

Sopera

Genuitec

itemis

excelsior

Activity sponsor

eclipsesource

Media sponsors

Methods & Tools

SD Times logo

TSSJS

Be a Sponsor

The Twenty Modeling Things

Stephan Eberle (itemis AG ), Kenn Hussey

Making With Eclipse · Standard (25 mins)
Monday, 13:30, 25 minutes | Cypress

Tags: Modeling
7
·
8
·
9
·
10
·
11
·
12
·
13
·
14
·
15
·
16
·
17
·
18

The demand for integrated environments to support model-driven development based on standardized or domain-specific modeling languages is on the rise. When setting out to develop such environments, toolsmiths must frequently use, buy, or build the same handful of common building blocks to support things like model exploration, property description, resource management, validation, compare/merge, search, index, etc.. While they can certainly benefit from the wide range of existing modeling frameworks and tools at Eclipse, experience has shown that there is still a significant gap between these existing components and what users expect to see in an industrial strength environment. This has resulted in a growing number of increasingly redundant modeling technologies and overlapping consumer-driven modeling components at Eclipse. We owe it to our communities to do better than this.

Just as the Eclipse e4 project has a set of composeable application services, also referred to as the "twenty things", the Modeling project should have its own list of essential services - the "twenty modeling things". Efforts are already underway to put some of these things into practice by creating higher level modeling tool platforms. Among these are the Papyrus project, where reusable parts that comprise UML tooling are being refactored into a generic backbone, and the Artop project (www.artop.org), where an AUTOSAR modeling tool platform has been built atop an AUTOSAR-independent layer of generic services. This talk will propose a core set of modeling services and APIs (based on industry experience and community input), discuss whether/how existing efforts support each service, and speculate as to how any remaining gaps might be filled by the upcoming Sphinx project, where Papyrus and Artop committers will join forces to create an integrated modeling tool platform at Eclipse.

Stephan Eberle works as product development manager at Geensys in France. He leads the development of the core parts in the Artop open source project (www.artop.org) which provides reusable building blocks for Eclipse-based AUTOSAR design tools used in automotive software engineering. He is also committer for the Teneo component in EMF and will be co-lead of the recently announced Sphinx project under MDT. Stephan holds a Ph.D. in Engineering and a Diploma degree in Electrical Engineering, both from the University of Stuttgart (Germany).

Kenn Hussey is an independent software developer, consultant, and blogger. A strong advocate of open specifications and open source, he is the leader of the Model Development Tools (MDT) project, a committer on the Eclipse Modeling Framework (EMF) project (among others), and representative of the Eclipse Foundation at the Object Management Group (OMG). Kenn holds Master of Science (Computer Science) and Bachelor of Computer Science (Honors) degrees from Acadia University. His personal interests include singing, sailing, and yoga.

Slides