It is absolutely not true that a sparsely populated area would not be interesting to provide with optical fiber.

Different country, but tell that to Comcast, for example. It is an inescapable economic fact that the population density of an area has a significant impact on expected turnover, because that puts a cap on what you as an operator can extract from that area, and in the meantime pay the same fixed amount for infrastructure construction. This means that there is a tipping point in terms of population density where the cost of construction simply costs more than can be recovered in a reasonable time.

I don’t know if that’s the case here. But even if that’s not the case, it’s clearly profitable for Delta to get rid of those 200,000 customers.

If you build a network correctly and don’t miss a single farm, you will have a network in the countryside that will fully pay for itself within 15 years, including maintenance, etc.

As my grandmother once said, ash, ash, ash burns peat.
Plus, 15 years is a long time in which a lot can happen. As a payback period, it is not completely risk-free. But it’s up to Delta to decide whether it wants to accept that or not.

Each cable in the Netherlands (including coaxial cable) has a revenue of between 15 and 20 euros per home/customer per month.

This is very strange, because I’m paying $17.50 for 1 Gb/s symmetric. Does that mean the connection itself only costs $2.50 per month? Because that seems very strong to me.