1 Jul
2016
1 Jul
'16
8:37 p.m.
On 2016-07-01 15:43, William Cheswick <ches(a)cheswick.com> wrote:
Strangely enough, the definition I have of a runtime system is very
different than yours. Languages had/have runtime systems. Some
environments had runtime systems, but they have a somewhat different
scope than what MS-DOS is.
I'd call MS-DOS a program loader and a file system.
> >>...why didn't they have a more
capable kernel than MS-DOS?
>I don't think they cared. or felt it was
needed at the time (I disagreed then and still do).
MS-DOS was a better choice at the time than Unix. It had to fit on floppies, and was very
simple.
“Unix is a system administrations nightmare” — dmr
Actually, MS-DOS was a runtime system, not an operating system, despite the last two
letters of its name.
This is a term of art lost to antiquity. Run time systems offered a minimum of features: a
loader, a file system, a crappy, built-in shell,
I/O for keyboards, tape, screens, crude memory management, etc. No multiuser, no network
stacks, no separate processes (mostly). DEC had several (RT11, RSTS, RSX) and the line is
perhaps a little fuzzy: they were getting operating-ish.
Uh? RSX and RSTS/E are full fledged operating systems with multiuser
proteciton, time sharing, virtual memory, and all bells and whistles you
could ever ask for... Including networking... DECnet was born on RSX.
And RSTS/E offered several runtime systems, it had an RT-11 runtime
system, an RSX runtime system, you also had a TECO runtime system, and
the BASIC+ runtime system, and you could have others. You could
definitely have had a Unix runtime system in RSTS/E as well, but I don't
know if anyone ever wrote one.
In RSX, compilers/languages have runtime systems, which you linked with
your object files for that language, in order to get a complete runnable
binary.
Johnny