I have worked with old-school journalists, and have done my share of freelancing between ~1993 and 2005. Even reasonably well done journalism is like debugging

I have worked with old-school journalists, and have done my share of freelancing between ~1993 and 2005. Even reasonably well done journalism is like debugging complex technical problems, but instead of working with systems that behave in a well-defined manner, you are dealing with people and complex wetware systems. They do not behave well at all. Journalist hunting for confirmation or corroboration of a good lead is like a software engineer doing root-cause analysis in a byzantine maze, where reliability of information is suspect and all interactions with various interfaces must be assumed adversarial. Or at best to provide wildly incomplete data. And the final output, the article that eventually gets published? Ask yourselves: have you ever written postmortems that must make sense to an audience who do not have the technical background to understand the nuances of the problem domain? Audience who will actively try to misunderstand what you have written, and extract soundbites out of context? Could you include enough information in the postmortem to pre-emptively defuse such attempts, so that you could trivially quote a section to point out the strawman attempt? That's what proper journalism is like. And I consider myself privileged to have worked with people that; they taught me skills I could not have picked up in the university or most workplaces.