I believe one piece of missing context is that this is part of a debate in the U.S. about how/whether transit planners should serve commuters to non-downtown jo

I believe one piece of missing context is that this is part of a debate in the U.S. about how/whether transit planners should serve commuters to non-downtown jobs, which are a large and growing percentage in some areas. U.S. commuter rail is traditionally entirely oriented towards bringing people from suburbs downtown, but some American cities are getting less downtown-centric in their employment patterns, with various secondary job centers in the suburbs. Some cities' transit planners are trying to find ways to serve those commute patterns. A few European cities are able to do that successfully, and the article looks at whether their approaches are applicable to the U.S. The answer is mostly no, due to the American suburban jobs being more sprawling rather than built on top of / near train stations. People are willing to drive from their house to a park-and-ride for a train to downtown, but doing that in reverse is inconvenient and unpopular. But it might work in a few cases where the suburban job centers are in a compact cluster, like a few examples the article gives in the Boston area. To the extent the article has a take-away recommendation, it's basically, "planners should focus on better serving existing compact suburban job centers like those". One problem mentioned is that even where new transit-oriented development is built in the U.S., there is usually more economic demand for it to be residential rather than commercial, so it generally doesn't make economic sense to put offices on top of suburban train stations.