I've been looking this challenge over and thinking about it for a while now. I think the intent is to simulate the kind of circumstances Linus Torvalds had when he went to University and wanted a better OS for his 386. In Torvalds' case, that led him to produce under the GPL. But here you are a believer in
the Public Domain Project and want to produce a better than MS-DOS product that will become the new base line OS all "future" (i.e. post 1983) computers are loaded with.
In essence, the challenge is to give PDOS the shot in the arm that GNU got from Linux. So, there is no implied class where you are given this as an assignment to do, no time limit the project is expected to be completed by, and no clearly-defined goals as to what the OS is supposed to be like other than that you are familiar with DOS and want an improved version of it. For example, you feel that all the power of the 386 is ignored/wasted by MS-DOS and you want a full 32-bit version of DOS, so you start writing a improving PDOS for this goal.
Am I right about this
Kerravon/Paul?