I'm a bit confused about the use of 'noobs' in the options. I thought it was important for suppressing the observation numbers, but I can't recall if it's necessary here.
I think the best approach here is to go with option B. Deploying the update as a new version and splitting traffic between the new and current versions seems like the most straightforward way to test the update with production traffic before fully replacing the current version.
Carri
4 months agoJenise
4 months agoKayleigh
4 months agoHuey
4 months agoQuentin
5 months agoRonny
5 months agoAleisha
5 months agoEstrella
5 months agoAlonso
5 months agoFanny
5 months agoWilda
5 months agoKarl
5 months agoDanica
5 months agoKassandra
5 months ago