Nilearn: Operationalize and apply triaging issue labels

[tsalo]

We have nine new issue labels: priority: low, priority: medium, priority: high, impact: low, impact: medium, impact: high, effort: low, effort: medium, and effort: high. These labels, along with the preexisting good first issue label, are meant to help triage issues, which the new @nilearn/triaging team will help with.

However, in order to best use these labels, we need to (1) do our best to operationalize them and (2) apply them to existing issues.

Toward the first aim, we need to define what "low", "medium", and "high" effort, priority, and impact should mean. I'll start with the definitions used in ME-ICA/tedana, and we can adjust from there:

• effort
  
  
  
  • good first issue: Equivalent to "very low effort"
  
  
  • low: Estimated at less than a day's work
  
  
  • medium: Estimated at <40h total work
  
  
  • high: Estimated at more than 40h total work
  
  
• impact
  
  
  
  • low: Improving code/documentation cleanliness/clarity, not function
  
  
  • medium: Improves code/documentation functionality for some users
  
  
  • high: Enhancement or functionality improvment that will affect most users
  
  
• priority
  
  
  
  • low: Issues that are not urgent
  
  
  • medium: Should be addressed soon
  
  
  • high: Issues that would be really helpful if they were fixed already
  
  

[bthirion]

Thx for the suggestion.
In the context of the sprint, I believe that the effort task is the most useful. Given the limited time available, I would actually bet that the most important is still to identify effort:low / good first issues.

[tsalo]

Absolutely. The immediate concern should be the low-hanging fruit for new contributors. How do folks feel about the definitions though? Are they clear enough? Are there any rules that would make them more useful?

EDIT: Also, to mention an idea @PeerHerholz brought up in Discord, would level: <beginner/medium/advanced> labels be more useful than effort: <low/medium/high>?

[bthirion]

No strong opinion on the alternative. I find effort: <low/medium/high> good enough.