I'm surprised by the ridiculously broad conclusion (headline) based on the narrow conditions of the experiment and how little emphasis was placed on the factors

I'm surprised by the ridiculously broad conclusion (headline) based on the narrow conditions of the experiment and how little emphasis was placed on the factors that lead to success. It's no surprise that work that's extremely limited in scope and may actually benefit from limited collaboration could benefit from remote work. As a simple example, working on the wrong thing for a few hours can easily have a negative effect on real productivity over the long term. "Amount of stuff produced" is therefore a terrible measure for that type of work, where collaborating continuously can help to cut off unproductive paths at their root. I don't think there's anything here to suggest that similar results would transfer to more creative work such as software development.