those sentences aren't logically connected - the outage this post is about is mostly confined to `us-east-1`, anyone who wanted to build a reliable system on top of AWS would do so across multiple regions, including Amazon itself.
`us-east-1` is unfortunately special in some ways but not in ways that should affect well-designed serving systems in other regions.
`us-east-1` is unfortunately special in some ways but not in ways that should affect well-designed serving systems in other regions.