🚀 go-pugleaf

On Thu, 27 Jun 2002 13:12:10 -0700, Jeffrey Siegal <jbs@quiotix.com> wrote:
> Christopher Browne wrote:
>> Oops! Jeffrey Siegal <jbs@quiotix.com> was seen spray-painting on a wall:
>> 
>>>Kermit Tensmeyer wrote:
>>>
>>>>   Linux is the OS.
>>>
>>>Actually, Linux is a kernel, not an OS.
>> 
>> It's all a question of definitions.
> 
> The relevant definition is that used by the people who created (and 
> those continuing to create) Linux.  They consider it to be a kernel.

I don't think those people have a monopoly on defining what is and
is not an operating system.  The term precedes linux.  Perhaps they
have the ability to define what the "Linux Operating System" is.

The definition I recall from CSC-501 was that that an operating system 
was the software layered over top of the computer hardware that provided 
services making it easier to write programs.  Things typically provided 
by such a system are IO services, support for, multi-tasking and process 
scheduling, management of storage devices, control of peripherals 
among other things. 

I would consider that to be at least one conventional definition.  It's the
one that was in most OS textbooks in the early nineties.  The linux kernel 
certainly fits that definition.   But of course so might software layered
over or used in conjunction with the kernel if people actually write
their programs to that layer rather than to the kernel.  

I don't know what's in text books these days. 

That makes everyone right when the say that linux is an operating system
and also right when they include all the stuff you suggest.  But when
they deny each others definition, they are both completely wrong.

> There is a well-recognized consensus definition of an operating system, 
> and that includes at least such things as basic run-time libraries ("C 
> library" on a Unix-like system), login/logout utilties, print spoolers, 
> system loggers, shells, basic networking daemons, etc.  It probably 
> doesn't include things like Emacs and TeX, but that's not really 
> relevant to this discussion.

And maybe it includes X and maybe it doesn't.

I don't dispute that this is yet another viable definition.

People arguing about whether the linux kernel is or is not an operating 
system are IMO not disagreeing so much as talking past each other.  
Sometimes it seems that the miscommunication is deliberate.  

Isaac