20181017 Dev Meeting

Agenda

General:

Update from last call 20180918 (and prev ones 20180822, 20180808)
- AI @ALL - review RFC in #861 "Local and body frames" for next meeting.
Tagged: issues, PR

HamishW:

Test suite - what are constraints (e.g. language)? "Shape" of proposed design - unit tests, clients vs server, what user stories test will meet, how individual libraries should validate.
- FYI https://github.com/ArduPilot/pymavlink/pull/192
Lorenz - per-message versioning follow up.
Standardised altitude naming
Getting stuff into ArduPilot/pymavlink

Thomas/Lorenz:

@Lorenz proposed MAVLink microservices versioning as means to better control change. Team to review/discuss for next meeting.
Deferred until next meeting: AI @ALL - review RFC in #861 "Local and body frames" for next meeting.
@Hamish raised issue of inconsistent naming on co-ordinate systems in docs, mavlink and QGC to avoid confusing.
- In particular Abs/Absolute should never be used for WGS84 as it is a synonym for AGL.
- @James - some docs may not have caught up to recent fixes in code that previously wrong frames.
- @Hamish proposed MSL to refer to AMSL.
Thomas raised https://github.com/mavlink/rfcs/pull/10 for review. Discussion to continue next week. James to see if there are additional stakeholders.
@Hamish What are rules for merging of PRs in MAVLink? Lorenz can't be on critical path for everything. Specific case discussed https://github.com/mavlink/mavlink/pull/1003
@Hamish Can we get common stuff that is eligible to move into ArduPilot/mavlink please. James to see.

AI @HW Follow up with Lorenz on his steps for assessing PRs. Perhaps seeded with checklist.
AI @All review https://github.com/mavlink/rfcs/pull/10 and #861 "Local and body frames" for next meeting.
AI @JamesP See if can find more stakeholders to review UTM_GLOBAL_POSITION
AI @JamesP Check problem with pushing mavlink common messages into ardupilot fork.

Attendees: Julian Oes, Daniel Agar, Ramón Roche