In comp.os.linux.advocacy, Ramon F Herrera
<ramon@xxxxxxxxxxx>
wrote
on Wed, 14 Nov 2007 19:39:08 -0800 (PST)
<a74f31f4-8467-4cb7-b8e2-b087a05ec639@xxxxxxxxxxxxxxxxxxxxxxxxxxxx>:
> On Nov 11, 4:45 am, Roy Schestowitz <newsgro...@xxxxxxxxxxxxxxx>
> wrote:
>> MIT releases the sources of MULTICS, the father of UNIX!
>>
>
>
> Even the name "Unix" is a play on the word "Multics". The Unix
> creators had an approach of "small is beautiful", and wanted to show
> off things like the new lowercase capable terminals. Multics was the
> first multitasking OS, hence the name. Professor Fernando Corbato who
> later became the head of the CS dept. at MIT, was awarded the Turing
> for his role in Multics ca. 1988.
>
> It took Microsoft 30 years after Multics to achieve a multi-user,
> multi-tasking OS, but they haven't figured it how to do it well to
> this day.
But they do it well enough to make US$billions in profits
per year. That's the maddening thing about all this.
Still, that's the only metric for success that makes any
sense at all, from a business standpoint. At some point
buyers may be required to examine their (our!) own ethos,
to ensure that we only buy operating systems from companies
that are reputable and honorable.
And in any event scheduling is not an easy problem.
Consider, for example, the optimal ordering of read
requests on a drive; this ordering, presumably, would be
an NP-complete problem in the most general sense.
Microsoft's solution works well enough for most -- and that's
about all one can ask for, if one's an Average Joe.
>
> Perhaps now they can take a peek at the source code.
PL/1 is not my strength. :-)
>
> Has the father of Unix been released under GPLv2 or GPLv3?
>
> -Ramon
>
--
#191, ewill3@xxxxxxxxxxxxx
Useless C/C++ Programming Idea #1123133:
void f(FILE * fptr, char *p) { fgets(p, sizeof(p), fptr); }
--
Posted via a free Usenet account from http://www.teranews.com
|
|
