However, you still have a point, I submit transmission losses (and costs) would probably rise. But the issue is uneven production, which is worth a separate treatment.

The first point about uneven production should be that it is not a problem for today. Due to the small size of an individual plant, short-term fluctuations are levelled with multiple parks on the local grid, while for slower changes due to weather, there is plenty of spare capacity to compensate - the reserve capacity already kept for peak hours or to jump in for a major, say a nuclear power plant that malfunctions().

But should wind achieve a larger segment of the electricity production, something more is needed. One possibility is to organise the power system on very large scales, so that entire weather zones (say from front to front) are covered, and it is always possible for electricity from wind parks that happen to have strong winds to be transferred to areas with low winds.

Another possibility is energy storage. The most obvious way (and one applied on a smaller scale by Denmark in cooperation with Sweden and Norway) is to use hydroelectric power passively - i.e., regulate power (and thus water) output in opposition to wind power and in line with demand. There is also pumped storage: excess electricity from wind power used to pump water from a lower reservoir to a higher one, and let it back through turbines when wind is low. There are experiments with another kind of pumped storage: air into disused mines. Maybe others read of further technologies and can supplement.

Finally, since we are speaking about something whose need arises over decades, there is the possibility of combination with photovoltaics. The typical daily output curves of wind power and PV can be combined to roughly match the double-peaked curve of human consumption, and there is also a rough compensation for weather dependencies. So if non-crystalline, non-rare-metals PV cells (especially film cells that can be applied to a great variety of surfaces and can have different colours to solve problems of aesthetics) can be made cheaper (something researchers promise), the age of truly decentralised energy production could come.

() OK, recent events suggest that in some US regions or Italy, this may not be the case - but then, this is a problem to be solved anyway.

_{*Lunatic*, n.
One whose delusions are out of fashion.}