...

• Agenda discussion - anything that needs to be added?
• Prioritized Topics:
• Decision on old open PRs
• Open Pull Requests VSS
• Open Pull Requests VSS-Tools
• Issues VSS
• Issues VSS-Tools
• Discussions VSS
• Discussions VSS-tools
• Prioritized topics for next meeting

Meeting notes  2024-03-25:

• "Inactive" PRs
  • Do we want to close them after a certain time of inactivity, if no-one seems to be interested in driving them forward?

Meeting notes  2024-03-18:

• AOSP
  • Merge ok
• VSS vs. VSS-tools compatibility
  • Shall VSS-tools vX.y be required to support anything else than VSS X.Y?
  • https://github.com/COVESA/vehicle_signal_specification/issues/734
  • https://github.com/COVESA/vehicle_signal_specification/discussions/732
  • Adnan: Nice to support, but how big is the effort, not easy if requires too much work
  • Daniel: What about packing the same version of the tool.
  • Erik: Advantage for Eclipse Velocitas if the vss-tools "engine" can support also older versions.
  • Erik: Would be quite easy to at least check which old VSS versions that are supported for each release
  • Erik: Depends on if resources are available.
• Overlay and remove
  • Shall new keyword "state" be considered to be part of "std VSS syntax", i.e. described in VSS doc
    • Alternative is to consider it just as a VSS-tools-specific feature
  • https://github.com/nwesem/vss-tools/pulls
  • Adnan: Does not really understand the use of active/inactive
  • Daniel: Is it intended only for exporter, or also other function
  • Adnan: Inactive/active fills a different purpose, not needed for vss-tools for purpose of removing signals of tree
• Terminology:
  • https://github.com/COVESA/vehicle_signal_specification/issues/733
• AMM

  • Topics for discussion:

  • Raw vs. derived signals; does VSS need to define or state how signals are derived?
    • Is this useful to clients?
  • For signals derived through ML; should confidence level, variance, precision, error, etc. be reported?
    • Is this useful to clients?
  • How should model-generated derived signals such as "driver attentiveness" be represented (in the absence of a standard)?
    • Confidence percentage? Enum?
  • To what level of specificity should VSS describe signals?
    • Leave semantics to OEMs (flexibility for OEMs but affects portability) or define them precisely?

...