You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Currently our participation metrics are boolean gauges per peer per duty type. This is visualised by red crosses and green checks for the current "participation state" per peer per duty. Red crosses are not good UX in general. It could also be misleading since the peer might have only missed one duty out of many. This is compounded by rare duties like proposals. If a peer doesn't participate in a proposal, then it has a red cross until the next proposal, which might be weeks.
Example with peer "Bar" sometimes not participating.
Peer
Attester
Proposer
Foo
✅
✅
Bar
✅
❌
Proposed solution
Change the boolean gauges into counters.
Change visualisation to number of participations per peer per duty per time window.
This removes the "stuck red cross" issue.
It also relativises participation as a more accurate ratio or sliding scale as opposed to binary greed/red.
Example with peer "Bar" sometimes not participating.
Peer
Attestations
Proposals
Foo
97
3
Bar
96
2
The text was updated successfully, but these errors were encountered:
Problem to be solved
Currently our participation metrics are boolean gauges per peer per duty type. This is visualised by red crosses and green checks for the current "participation state" per peer per duty. Red crosses are not good UX in general. It could also be misleading since the peer might have only missed one duty out of many. This is compounded by rare duties like proposals. If a peer doesn't participate in a proposal, then it has a red cross until the next proposal, which might be weeks.
Proposed solution
The text was updated successfully, but these errors were encountered: