The author does cover this: "It can be argued that firmwares are indistinguishable from OSes, and that firmware is the only "true" bare metal programming one can do."
Maybe it's just my limited knowledge on the subject, but it seems to me it's a bit too much to consider the BIOS an operating system. At that point, what isn't an operating system? You're still writing code against a microcode interpreter even once you remove the BIOS. Is that an OS too?
I would agree. The BIOS/UEFI POSTs the system, and then passes execution to another program (usually a bootloader). At that point, the BIOS is no longer performing the tasks that an OS does. It's not scheduling tasks for the processor or handling I/O at all. Instead, it switches to what's basically an ACPI interface with functions that can be called by the OS. It's like an old TSR program.
UEFI is the successor to BIOS on modern machines, and it stands for "Universal Extensible Firmware Interface". Is it also an operating system (it doesn't have "system" in the name)?
(I don't disagree with you -- UEFI provides BootServices which arguably are more numerous than the number of syscalls on most operating systems. Just that "system" being in the name isn't really the best argument. :P)
It was thought to be like that - back in the day when everything used interrupts for syscalls, hence 'system' was mentioned. (Apple II had fixed address calls, though - no interrupts)
It's quite the same as DOS being 'disk operating system' - it was able to drive the floppy and provide a file system.
lts not talk about acpi :D ... or system management mode which i think you refer to as BIOS, since bios really hands off control. but has set up SMM to do stuff outside of the os.
