While I agree that the cost of operation and yield are a valid concern, the same argument could have been used against renewable energies like wind and solar only 30 to 40 years ago.

The price of these energie sources has come down a lot since, for a large part thanks to the modern day widespread use. We have a lot of experience generating power this way which drives down cost, and increases yield.

Novel techniques like the one described in the article don’t yet benefit from that experience and scale. And if we don’t try new things every now and then they never will.

That is not to say all novel techniques will be equally fruitful, but if you don’t occasionally try new things you will never learn.