To all,
Thanks for participating to the meeting today. Here are my notes from the meeting. Feel free to add to it.
Also, thanks to everybody for the great collaboration since we started working on the creation of the Papyrus IC in 2015!
Proposed agenda- Current status Polarsys and ICs
- Current status of Papyrus
- Papyrus UMLLight
- Current projects using Papyrus -- academic and industrial
- Discussion on the need/purpose of an industry consortium
- What went well, what could have been done better
- Where should we go from here?
Participants - Gaël Blondelle (Eclipse Foundation)
- Francis Bordeleau (ETS, Cmind)
- Dimitris Kolovos (University of York)
- Florian Noyrit (CEA)
- Ernesto Posse (Zeligsoft)
- Philip Langer (EclipseSource)
- Quentin Le Menez (CEA)
- Ricardo Sanz (Universidad Politecnica de Madrid)
- Charles Rivet (Zeligsoft)
- Cort Starrett (OneFact)
- Patrick Tessier (CEA)
- Johan Van Noten (Flanders Make)
- Thomas Wiman (Adocus)
Current status Polarsys and ICs - Polarsys officially terminated at the end of 2018
- But projects are still being maintained
- The Eclipse Foundation is still interested in supporting the creation of WGs for the ICs
- But currently don’t see the motivation/interest from the IC side
- We should only create WG if there is a real needs for it
- Otherwise work can be done in the context of the open source Papyrus project
Current status of Papyrus - Organisation
- Official leader of the Papyrus project: Sebastien Gérard
- Inside CEA: the Papyrus development work is led by a committee
- Patrick Tessier is the technical leader of Papyrus inside CEA
- Main evolutions of Papyrus
- See the Papyrus wiki — the wiki is up-to-date
- SysML 1.6 — implementation of the SysML 1.6 profile
- Papyrus has now moved to the new Eclipse release process
- Plan to make a new major release in 2020.3
- Current contributors
- CEA
- EclipseSource
- Open to new contributors
- Ongoing work
- Continue to discuss Sequence Diagrams with Ericsson (Antonio CR)
Current status of Papyrus-RT - Some work done by the group of Juergen Dingel at Queen’s U
- One click generate, compile, and run
- Supported for distributed applications
- Work on debugging for models
- Take an IoT direction
Papyrus UMLLight - Tool — Philip L
- Final release candidate has been released at 2018
- Bugs have been fixed during 2019Q1
- Documented the customization techniques used
- Proposing to move it in the Papyrus project
- Currently leaving on Github
- Could also create an official download for Papyrus UMLLight with Components and Deployment
- Documentation — Charles R
- Done and posted on the wiki
- People can review and comment
- Published on the blog
Current status of Papyrus Compare - Improvement on performance in the last months (major improvement)
- Continue to build at every release
Current projects using Papyrus -- academic and industrial - Number of downloads of Papyrus:
- CEA
- Some European projects: CPS for ???, PDP4E, Robmosys, ...
- Industrial partnerships: Egis (civil engineering for transportation), Renault, plus others not public
- EclipseSource:
- Research project: HybridDLUX (project with AVL in automotive), plus small support contracts
- OpenADX meeting last week: Clear interest in using Papyrus, but too early to say where it will be going
- The goal of the project is to create an end-to-end tool chain for autonomous driving
- Adocus: Swedish gov projects
- Ericsson: still using in some projects
- Cmind: Plastic Omnium
- ONAP
- Actions: participants to list the projects that they have using Papyrus wiki
Discussion on the need/purpose of an industry consortium - To justify the existence of an IC, there needs to be a strong group of committed (end-user) companies
What went well, what could have been done better - Charles — We didn’t go wide enough regarding the development of the community
- We should have considered different application domains and different types of companies
- The product needs to be easy to use and modify — we have been (and still are) too much expert-centric
- Charles — Should look at the way Sirius has done things
- Francis — Too much dependent on a major end-user contributor — bus factor at the end-user level
- Cort — Similar issues in the context of xt-UML
- Not relying on a single end-user, but still facing similar issues regarding the development of the community
- The Papyrus IC core group has a very good WoW and collaborations
- Philip — Should have put more focus on specific products rather than trying to address everything at the same time
- More focus on specific products
- E.g. Papyrus IM, Papyrus-RT, Papyrus UMLLight
Where should we go from here? - Cort — Proposal to officially terminate the Papyrus IC
- No need to do that as the Papyrus IC was officially terminated with the termination of Polarsys at the end of December 2018
- We would like to keep a mailing to keep the communication going
- Blog — Since nobody except Charles has been contributing to the blog, Charles to terminate the blog
Regards,
Francis
|
