Is a machine WORD always the same or does it depend on the machine architecture? And is the meaning of the word WORD context sensitive or generally applicable?
A word is typically the "native" data size of the CPU. That is, on a 16-bit CPU, a word is 16 bits, on a 32-bit CPU, it's 32 and so on.
And the exception, of course, is x86, where a word is 16 bit wide (because x86 was originally a 16-bit CPU), a DWORD is 32-bit (because it became a 32-bit CPU), and a QWORD is 64-bit (because it now has 64-bit extensions bolted on)