Wp7 workshop summer 2013 minutes

• Tasks (T) - which must have an Owner and a deadline
• Decisions (D)
• Information (I) - everything else of interest.

Type	Description	Owner	Deadline
I	Already dropped out: CORE, Why3, GOPPRR, Color Petri Nets
D1	We will use Papyrus for SysML modeling (Enterprise Architect will be a fallback in case we encounter a showstopper with Papyrus)		Vote: 0 in favor of Enterprise Architect
I	Closed Source: Closed source on the critical path is okay if the inputs and outputs are documented. David Mentré and Jonas Helming did not agree with this statement.
T1	Is SysML formal, and can breaking curves be modelled with it? Jan Peleska will provide material next month	Jan Peleska	1-Aug-2013
D2	SysML will be the starting point for a high-level design. All other means of description will we consider for continuing the tool chain further down.		No vote against
T2	Over the next weeks, everybody is invited to propose constellations of tools to complete the primary tool chain. Marielle will distribute further information about this process, and we will discuss the status of this at next week's telco.	Marielle / All	12-Jul-2013
I	**Note: Will be rephrased by Alstom and ERTMS Solutions: ** Alstom supports the choice of SysML, as their internal evaluation concluded that the SysML-Scade tool chain was a viable option. ERTMS Formal Specs is their backup solution.
T3	As Scade is still in the evaluation, its underlying platform architecture should be considered and evaluated as part of the platform evaluation	Uwe	12-Jul-2013
T4	Uwe will lead the initiative that is analyzing how SysML and Scade can be integrated, and how this fits into the OpenETCS process. Marielle will help him, due to the link with D2.4 and D7.1.	Uwe	12-Jul-2013
D3	Eclipse with the modeling tools will be the development platform.		No vote against.
D4	If Frameworks are added to Eclipse, the justification should be documented.		No vote against.
I	Uni Bremen: Interface with RT Tester
I	Siemens: Review & Test
I	ERTMS Solutions: Integration of ERTMS Formal Specs, according to the project's needs
I	TWT: Eclipse Integration Work (ideally with System C)
I	Uni Rostock: Support executable Models (also with System C)
I	Uni Braunschweig: Feedback, Petri-net based safety modeling
I	DLR: Produce common test database to reduce redundancy in testing, make fit for laboratory use
I	CEA: Interface incoming and outgoing flows (SysML), possibly with Petri-Nets
I	All4Tec: development of specific user layer (EMF based or similar) Scade and petri-net based testing, many more ideas.
I	DB: Early user of tool chain, review, feedback, testing
I	LAAS: Activity-diagram and state-machine-based model checking, Petri-Net based black-box testing possible as well
I	Mitsubishi Electric: Enduser of modeling tools, experiments with code generation
I	Fraunhofer: Executable models in conjunction with Papyrus.
I	Fraunhofer: End-user, code verification, abstract interpretation, using Frama-C tools
I	CEA: support Fraunhofer and ERSA with Frama-C.
I	Institute Telecom: End user of secondary tool chain
I	SNCF: Observer / collect WP2 feedback for requirements
I	EclipseSource: Support implementation, contribute EMF expertise, distribution, QA
I	Alstom: Enduser
I	Systerel: End-user, V&V-partner, participate a little to integration (Rodin/Eclipse), analyze tools.
T5	Aggregate the previous items for T7.3, and broadcast the request for support on the mailing list, to make sure that those partners not present have a chance to define their contributions	Cecille	12-Jul-2013
T6	Clarification needed on how to deal with closed-source solutions. Do we allow closed-source tools to be on the critical part without alternatives? Clear no from Klaus-Rüdiger Hase.
I	Hase: Any closed-source tool can only be a plan B, for every closed-source tool there must be a feasible open source alternative. Steinke: No, this will undo today's work
I	WP4 will use SysML (Papyrus) and Scade for their first round in V&V
I	Jonas Helming raised the concern, that once we have chosen a closed source alternative first and start to work with it, that there will be ever a valuable open source alternative
T7	Who volunteers in secondary tool criteria definition? Who is willing to assess? Please contact Marielle	All, Marielle	12-Jul-2013
T8	Bernd will assemble a small team to capture requirements for SSRS tool functionality and data model.	Bernd	9-Jul-2013
I	Highest priority with the SSRS is the data model.
D5	First tool to be used for SSRS will be excel
I	Decision on SSRS tooling will be made by the end of September
T9	Volunteers wanted for SSRS tools and SSRS support. Please respond asap.	All	12-Jul-2013
I	Stan disagrees with the "First proposal" picture that outlines a SysML - SCADE toolchain.
I	We have to find out at what point ("going down") SysML has to stop. Stan states that for im SysML is not usable below "System analysis" stage.
T10	We are open to proposals for alternative tool chains (compared to SysML/SCADE) from system analysis to code generation.
I	We'd like to state explicitly that the tools outside Papyrus and Scade are still considered for WP7, including ERTMS Formal Specs, B/Event-B, System C.