I think it is a fascinating way to use energy right in the desert, at the source, without transporting it to some other destination.
At the time, I said that the only problem with that is that there is no sea water available in the Gobi desert, the one I am most interested in.
Since then, I have reconsidered.
Actually, it would be a trivial task to build a pipeline and just pump sea water from the nearest ocean. Build one way to bring the sea water to the desalination capacity in the desert, and another one right next to it to export fresh water to China. And then use some of that water for agriculture right in the desert.
The distance is only about 1000 kilometers from Tianjin to the Gobi. Researching this a bit I find that there already is a proposal to build a water pipeline from the Bohai Sea to Xilinhot in Inner Mongolia, which would be about 600 kilometers and cost only $6 billion. Ironically, in the proposal that water is supposed to be used for mining coal in Inner Mongolia, which is of course the last thing the planet needs right now, but that proposal is of interest to get a first impression of the very small costs involved to get sea water to the Gobi desert.
Of course this is not something one will do in the first two decades of such a project, but it is of interest to note for the very long term as one of the ways one would want to use basically zero cost solar energy in unlimited quantity, once the project gets running at a scale relevant to counter global warming.
For the time being, these first steps in agriculture using only sun and sea water will need to be located close to the ocean, as the first Sundrop Farms greenhouse in Australia.