I am a software engineer and a researcher working in renewable energy R&D. My passions include reliable and resilient software systems, renewable energy, programming languages, the electric grid, operating systems, large scale data storage systems, distributed systems, and unconventional computing.
Delivering multicast messages in the same order to every process in the group.
The gen_event behavior is still cool, but my approach in the last post was way too complicated.
The gen_event OTP behavior is much more interesting than it first appears. Because supervised handlers can crash without killing their event manager the gen_event behavior provides a kind of “transparent supervision” that makes it easy to depend on failure-prone processes without any extra implementation effort, monitoring, or failure handling complexity.