MakieCon feedback

[lazarusA]

Maybe include here all the feedback provided during makiecon, so we can have a clear idea of what is already possible and what is not.

@piever ?

[piever]

Hey, yes, that's a good idea, I plan to go through my notes and upload them here next week.

[piever]

So, with some order, here is the actionable MakieCon feedback.

Usability issues

• Legends and labelling.
  • Provide helpers for customizability (mix and match with existing Makie legends)
  • Legends does not adjust completely to style used (cf. color legend entry #228)
  • Easier renaming (when using wide data or pregrouped data things are clumsy, it may be tricky to remove some names as well)
• Performance.
  • Figure out precompile times (see Improve precompile times #459)

Data processing

• Provide extensibility mechanism for custom analyses (worked at during hackathon, but still wip). At the moment only possible by "copying" the source code for existing analyses.

Documentation issues

• Philosophy first approach. Iinsist more on philosophy on the documentation, it seems easy to end up "using AoG the wrong way", which can be frustrating.
• Insist on possibility of creating first "draft" of plot with AoG and then tweaking with Makie (documentation for that is hidden here).
• Document different data formats (pregrouped data documentation is hidden here, but it can be very useful for plots of 2D data such as heatmaps).

[jkrumbiegel]

I'm going to close this as the recent refactors touched a lot of these points, and if any of these points reappear it would be better to have targeted issues for them.