Not sure if anyone has suggested this to the Graylog team but how about making software that just works? Why does one have to jump hoops change config files etc (essentially wasting time) to setup the application???
Guess what? The real purpose of the application is to get insight into logs? Not wasting time trying to get it up and running by fiddling with config files et al.
While it may create some economic benefits to have Linux folks gainfully employed you all are missing what the actual business case of the application is “LOG ANALYSIS”
You application should be transparent.
Click and run people. Abstraction abstraction!!!