Those are not really remotely the same. One is about a stupid architectural design (combining 16 bit segment and offset to get only 20 bits of address space), limiting not just a given computer, but any future computer with the same architecture, and doing so at a time when other CPU vendors like Motorola were switching to full 32 bit addressing (even though they didn't actually have 32 bits of address space). The other is about a lack of demand for extremely large hard drives today.
It also seems to be true. Some people may need more space, but 99% of desktop/laptop computers are well served by a single SSD or SSD + <= 6 TB hard drive for mass storage. Even "small" servers that need 10s of TB would usually prefer to spread that over multiple spindles for performance and fault tolerance. The people who might be a current market for a hypothetical 50 TB drive are cloud storage providers but they need petabyes of storage anyway, so really they care about TCO/TB more than TB/spindle.
Maybe that will change in the future, but it certainly seems to be the case today.