3 ms·
So Google can manage hundreds of feature/experiment flags on accounts to control which shade of blue you see in an A/B test, but silo-ing which products an acco
by monkeybutton 5y ago
So Google can manage hundreds of feature/experiment flags on accounts to control which shade of blue you see in an A/B test, but silo-ing which products an account has access to just isn't possible?
- CrendKing 5y agoI think it's more like they don't want to do it. If they consider someone a "fraud", why allow this "bad actor" to be still in the system? It's like "if someone steals, put them in the jail altogether, instead of just ban him from grocery stores".
- monkeybutton 5y agoIt still seems overly harsh at the personal/individual level. Should lose your photos because you make a ghastly comment on YouTube, or lose your email access because you app. got pulled? A Google account has incredible breadth. I understand they have to deal with an incredible scale of fraud but surely they can differentiate between a recently created account using a few of their products for fraud vs. an account that has human-looking activity over a longer period of time.
- CrendKing 5y agoI totally agree of differentiating punishment from behaviors. A bad comment shouldn't cause all the pics from Google Photos to be inaccessible. But like you said, Google probably has to deal with these issues in incredible scale, so unless some high up personally deal with the individual cases, almost all of the bans are automated. Also, considering all the intervention would incur some cost, especially with high ups involved, unless unbanning the account has greater benefit to Google, why would they waste time doing it?
- daniellarusso 5y agoI think it is more analogous to “if someone steals, they are no longer allowed to vote.”
- sp332 5y agoIt's the same account though right?