The post says "Unfortunately, in introductory texts, their operation is often explained in confusing ways." and this is 100% true. I remember being baffled by them in high school (when I was taking an online college-level EE course). The blog is a remarkably good explanation.

I was wondering about this as well. Now I would like about the performance of that multiplier and what kind of multiplication to expect from it (4-quadrant?), but the summing, subtraction and integrator circuits are bog-standard and the first present in most analog signal mixing.