In my second job, in the mid to late 80s one of the things our software ran on was SAGE IV computers. These had a 68000 processor and ran a variant of CP/M for the 68000, the wittily named CP/M 68K. Now, because they ran CP/M they didn't have directories, just hard disks partitioned into drives of 2MB each (*). The drives could be named A to P (16 letters!). As the different machines we had had different capacity hard disks, for consistency's sake they were arranged so that the floppy drive was always P, the last permitted letter, so that 10 MB machines would have hard disk drives A to E, the 20 MB ones would have A to J and the 30MB ones A to O (if we actually had any, I forget). As I recall the A drive held the OS, the FORTRAN compiler and our source code, later letters could hold customer data, one set per drive.
Now, at this time, we were also developing the software for PCs under DOS which, of course, had floppy drive A and hard disk C, with directories, this being the days of DOS 2.0, though my boss for one was not having any truck with these heathen 'directory' abominations and kept everything in the root of his C drive.
Therefore, people who were working on both could easily get confused and say FORMAT A: on the SAGE forgetting that they should be saying FORMAT P: This, of course, wiped out the contents of the A drive which included the operating system, the FORTRAN compiler and our source code.
How we laughed. Particularly as getting the OS back on was a painful process involving formatting it using the UCSD-P System and then putting CP/M back on.
Ah dear dead days.
(*) I have no idea why they were limited to 2MB - a CP/M limitation or how we wanted to do things.
Looking on Wikipedia, I see that one of the founders of SAGE has written his recollections on his blog http://suddendisruption.blogspot.com/search/label/Stride%20Micro
I recall we had some TDI Pinnacle machines, essentially a knock off of the SAGE as described in the article, though I can't remember what they were used for.