84

So, although I think that is beneficial to conceptual integrity of a software system if it is the product of a single mind, I do understand certain aspects that incentivise organizations against such setups. The risk when that single mind leaves usually outweight the concerns about internal qualities of a system.

Therefore organizations assign multiple employees to a product. But when the division of labor is along the layers of the stack (e.g. a front-end and a back-end dev), the risk is similar as in the situation where only one person knows the system. So, on the one hand the orgs push division of labor as a strategy to derisk products/projects, but that division of labor also pushes individuals into technical specialization, which in this case acts a opposing force to the idea of risk reduction. To have low risk you also want the people to be able to cover many technical layers with reasonable depth. Of course, business always wants to have the cake and eat it as well.

In a talk I attended recently, I've heard a pithy statement that resonated a lot with me: for a full stack engineer nothing in a system is somebody else's problem. But in my experience, that is a mindset, and as such can not be bought.