Yes, We Can Melt It.
You can use mirrors or lasers to melt sand. Doing so from orbit is a bit over-the-top and impractical. Not only do you need to deliver the energy to melt the glass, but you first need to get the energy past the atmosphere (which would involve heating that up to or above the target's melting temperature), and then you need to deliver that energy and move the soil out of the way!
Let's do Some Physics!
We need to know a few things first. I'm assuming the sand is made of mostly quartz, or $SiO_2$, even though real glass has a bunch of other things in it.
How Much Sand?
I'm going to do my calculations for 1 kg of sand. You'll need to figure out how much glass you're making on your own.
Using a simple $Q=mC_p \Delta T$, and thinking this sand is already in a hot desert (~40C), the sand requires 1.378 MJ just to get to melting. To completely melt, the sand requires another 156 kJ, so it takes 1.534 MJ to melt a kilo of sand into glass. This calculation does not include penetrating the atmosphere, or any cooling effects at the target.
Here is a nifty crater calculator. Modeling this weapon as an explosion above the surface (with the equivalent energy of 0.0003667 Tons of TNT), this calculator says the surface crater is a measly half a meter, and only 15.6 cm at maximum depth! Even so, any energy we use to make the crater is less energy we use to make the glass.
In any case, this calculator can be used to estimate the energy needed to form a crater. It should be noted that the Trinity test did melt sand into a green glass, and it had a payload of 84 TJ. Its crater was only 1.5 m deep and 9.1 m wide.
Given all this, it seems an orbital glass maker may be able to make glass, but it will not form a crater without ridiculous amounts of energy!