Forum Discussion
PeteGomersall
May 07, 2021Iron Contributor
A seeming illogical design\test strategy?
This post is looking for answers from people in the Microsoft Edge design\testing team. I have been in IT system administration and software development for over 30 years but I just don't get your d...
josh_bodner
Microsoft
The main reason we don't set experiments per user is because signing into the browser isn't required. Whatever system we use needs to work regardless of a user's identity, so in order for experiment results to be statistically valid, we can't change the way we set experiment populations for any reason, including whether or not the user has an identity that is known across other devices.
For what it's worth, the best way around this is to use flags for any feature that you want to explicitly turn on, although that does require that we actually create said flag.
For what it's worth, the best way around this is to use flags for any feature that you want to explicitly turn on, although that does require that we actually create said flag.
PeteGomersall
May 22, 2021Iron Contributor
josh_bodner,
Thanks for the info I appreciate the reply. I can totally understand the approach you detail for what one could call "global features" but not for features that provide/require an account to actually work, like those related to sync.
Thanks for the info I appreciate the reply. I can totally understand the approach you detail for what one could call "global features" but not for features that provide/require an account to actually work, like those related to sync.