I'm still confused what would motivate a company to be so secretive about its hardware designs or infrastructure size. I understand that having more efficient

I'm still confused what would motivate a company to be so secretive about its hardware designs or infrastructure size. I understand that having more efficient infrastructure is a competitive advantage, but it seems as though just seeing their machines wouldn't provide enough insight to duplicate them. "Hey, Google's servers have fans on the back and the front" or "Hey, Google's servers are 2U, not 1U" don't seem like overservations that would provide a competitor any insight. Is designing the these type of machines not as straightforward and standardized as it seems? Similarly, say a competitor saw their racks and noticed SSDs instead of physical disks. What information does that really provide? It's reasonable to assume that each company tries to minimize their cost and maximize performance, and that for similar situations they would make similar decisions. Is it for strategic value -- say, knowing where on the price-performance scale Google is choosing to be -- or is it to calculate their operating costs so as to better price their own products? I'm not even sure why quantity of infrastructure is closely guarded. What advantage would a competitor gain knowing that Google has, say, 2M servers instead of 1M?