still worth pointing own that library-izing chrome for Electron &c would fix 60% of memory complaint issues, and the remaining 40% issues are just fantastically bad web app design.
I'm responsible for mobile crash reporting at my company. Normally we see about 10-15 crash reports/second. We've been seeing over 1500 reports/second since about 2:30 PM US/Eastern.
Across dozens of apps installed on millions of devices, some with year’s old OS and application versions that can’t or won’t update, and includes both foreground and background crashes, yes.
The rate of crashes per application session is (exclusive of this event) very low.
In my anecdotal experience, this is also happening with Chrome webview. My mother's Galaxy S8+ had app crash issues that didn't resolve until I both disabled Chrome and pinned Android System Webview to the factory version.
I know you are joking, but a lot of Google development is data driven.
This means that a lot of decisions are done by A/B testing instead of proper analysis and design.
This is most visible in UI stuff. If you wonder why Gmail or YouTube look or behabe differently every few months, that's the reason.
Suffice to say, this confuses the hell out of less tech savvy users and the main reason some people feel uncomfortable with Android but can't really point on any particular issues.
17 comments
[ 3.2 ms ] story [ 56.7 ms ] threadGood times.
The rate of crashes per application session is (exclusive of this event) very low.
This means that a lot of decisions are done by A/B testing instead of proper analysis and design.
This is most visible in UI stuff. If you wonder why Gmail or YouTube look or behabe differently every few months, that's the reason.
Suffice to say, this confuses the hell out of less tech savvy users and the main reason some people feel uncomfortable with Android but can't really point on any particular issues.
So the faulty version was up for a good few hours?