[comp.sys.ibm.pc] MKS Idea

john@stiatl.UUCP (John DeArmond) (11/11/88)

I've been watching this discussion on the various incompatibilities with
DOS and MKS and the thought just struck me.  Why does not Mortise Kern
go ahead and turn MKS into a standalone OS?  Hmmmmm...  You guys already
have about 80% of it done with the shell and utilities.  Why not put together
a nice little multitasking, single user kernel to run below the shell.

Maybe we could get a discussion going here and maybe if there's enough
interest, Gerry could make it happen.  How 'bout it guys.. interested in
getting away from Microsloth pseudo-dos?  Lets make some noise.



John De Armond
Sales Technologies, Inc.    Atlanta, GA
...!gatech!stiatl!john

pavlov@hscfvax.harvard.edu (G.Pavlov) (11/13/88)

In article <1396@stiatl.UUCP>, john@stiatl.UUCP (John DeArmond) writes:
> ...........................................  Why does not Mortise Kern
> go ahead and turn MKS into a standalone OS?  Hmmmmm...  You guys already
> have about 80% of it done with the shell and utilities.  Why not put together
> a nice little multitasking, single user kernel to run below the shell.
> 
  I don't think that what has been done is anywhere close to "80%" of what would
  be needed.  The shell and utilities are the least of it.

  greg pavlov, fstrf, amherst, ny

alex@mks.UUCP (Alex White) (11/13/88)

In article <1396@stiatl.UUCP>, john@stiatl.UUCP (John DeArmond) writes:
> I've been watching this discussion on the various incompatibilities with
> DOS and MKS and the thought just struck me.  Why does not Mortise Kern
> go ahead and turn MKS into a standalone OS?  Hmmmmm...  You guys already
	i) You have to be kidding yourself if you think this would ever be
	   commercially viable.
	ii) Any idea of how many different inconsistent incompatible
	    illegal stupid etc. way various programs use DOS?
	iii) Its been done. (e.g. Wendin).  We haven't noticed any particular
	     success by them.
	iv) Its already being done by microsoft.  Its called OS/2.
	    Now OS/2 is pretty bad, but a lot of the work that went into it went
	    into the silly compatability box, which is just what we'd have to
	    do as well.

wheels@mks.UUCP (Gerry Wheeler) (11/14/88)

In article <1396@stiatl.UUCP>, john@stiatl.UUCP (John DeArmond) writes:
> Why does not Mortise Kern
> go ahead and turn MKS into a standalone OS?  You guys already
> have about 80% of it done with the shell and utilities.  Why not put together
> a nice little multitasking, single user kernel to run below the shell.

Well, it's nice to know that people think we could whip up a new OS in a
week or two :-), but I think it would really take a lot of work.  You
may not be pleased with MS-DOS, but it does represent many man-years
(people-years?) of effort.

So far, we've resisted all temptations to make any changes to the DOS
kernel -- all our programs run on stock versions of MS-DOS.  I think
we'll continue to work in that direction.  And, the same will likely be
true for OS/2 -- it may be big, but if we run on the stock version
everyone will be able to use our utilities.  Making changes to the OS
would likely create a lot of problems. 

-- 
     Gerry Wheeler                           Phone: (519)884-2251
Mortice Kern Systems Inc.               UUCP: uunet!watmath!mks!wheels
   35 King St. North                             BIX: join mks
Waterloo, Ontario  N2J 2W9                  CompuServe: 73260,1043

kiy@pte.UUCP (Kevin Young) (11/15/88)

In article <1396@stiatl.UUCP> john@stiatl.UUCP (John DeArmond) writes:
>I've been watching this discussion on the various incompatibilities with
>DOS and MKS and the thought just struck me.  Why does not Mortise Kern
>go ahead and turn MKS into a standalone OS?  Hmmmmm...  You guys already
>have about 80% of it done with the shell and utilities.  Why not put together
>a nice little multitasking, single user kernel to run below the shell.
>

Hear, hear.  I'll take a dozen right now.  It could even be called MkS-DOS.

Kevin

-- 
>|< Kevin Young                                 uunet!edsews!rphroy!pte!kiy  
    Precise Technology & Electronics, Inc.
    (The opinions expressed here are true, the names have been changed to
     protect the idiots.)

keithe@tekgvs.GVS.TEK.COM (Keith Ericson) (11/15/88)

In article <1396@stiatl.UUCP> john@stiatl.UUCP (John DeArmond) writes:
>I've been watching this discussion on the various incompatibilities with
>DOS and MKS and the thought just struck me.  Why does not Mortise Kern
>go ahead and turn MKS into a standalone OS?  Hmmmmm...  You guys already
>have about 80% of it done with the shell and utilities.  Why not put together
>a nice little multitasking, single user kernel to run below the shell.
>
>Maybe we could get a discussion going here and maybe if there's enough
>interest, Gerry could make it happen.  How 'bout it guys.. interested in
>getting away from Microsloth pseudo-dos?  Lets make some noise.
>
>

Without a doubt one of the best ideas ever to come down usenet!  Go for it,
Gerry.  Do a '286 version separate from the '386 version, though, so you
can do the '386 version right and keep the '286 cretinisms out of an
otherwise potentially fantastic product.

I think the key is Multitasking SINGLE USER. I don't WANT other users on my
machine! It's a waste of resources (your programming time and my machine)
to include that capability in my desktop computing environment. If I want
multi-user I'll go with multi-machines.

Oh, yeah: include SCSI and RLL capability, too.  Ethernet if you can.  (Or
contract it out if you can't.) Print spooler(s), too.

And do a "tar" while you're at it, eh?

keith

jes@eniac.seas.upenn.edu (Joe Smith) (11/15/88)

In article <4255@tekgvs.GVS.TEK.COM> keithe@tekgvs.GVS.TEK.COM (Keith Ericson) writes:
>In article <1396@stiatl.UUCP> john@stiatl.UUCP (John DeArmond) writes:
>>I've been watching this discussion on the various incompatibilities with
>>DOS and MKS and the thought just struck me.  Why does not Mortise Kern
>>go ahead and turn MKS into a standalone OS?  Hmmmmm...  You guys already
>>have about 80% of it done with the shell and utilities.  Why not put together
>>a nice little multitasking, single user kernel to run below the shell.
>>
>Without a doubt one of the best ideas ever to come down usenet!
                            ^^^^ - read worst

>And do a "tar" while you're at it, eh?
           ^^^ - get John Gilmore's Gnu tar, it runs will on MS-DOS
                 and with the MKS shell you can do things like

	tar cvf - dirname | compress -c > filename

		 in a more elegant way.

The reason why the MKS-DOS is a bad idea is:

	1) The reason most of us like MKS is, it allows us
	to pretend that we have a rational machine under
	our fingertips. 

	2) The reason MKS is successful is in part because
	they have brought us a large number of Un*x tools
	and a shell that works. If you want more why not
	get the real thing. It is available and even runs
	with fair performance on 286s and 386s.

	3) It has taken Microsoft years of fiddling to get
	OS2 to the point where it is slow and cumbersome.
	Oh, and not just years, but many many man hours.
	So, why take away from the standardization which
	may eventually come from sysV Un*x by inventing yet
	another multitasking os.

	4) Besides, you can remove all the passwds from a personal
	un*x machine except your own, and no one will accuse
	you of having a multiuser-multitasking system. It would
	only be multiuser if you wanted it to be.


So, I have an asbestos suit ready for anyone who wishs to argue this
point more. But lets give a good thing (unix) a fair chance. It is
a lot better than MS-DOS can ever by as far as productivity tools 
are considered, and graphical interfaces are going to blow the 
OS/2 presentation manager out of the water.

I have been a dedicated MKS user for several years, and this
product has been a god send to MS-DOS computing, but the answer
is not to rebuild un*x, but to buy the un*x that exsists and put
it on hardware that will run it.


Frank Kolakowski 
____________________________________________________________________________
|c/o jes@eniac.seas.upenn.edu			||	Univ. of Penna.     |
|kolakowski%c.chem.upenn.edu@relay.upenn.edu	||	Dept of Chemistry   |
|bcooperman.kolakowski@bionet-20.arpa		||	231 South 34th St.  |
|AT&T:	1-215-898-2927				||	Phila, PA 19104     |
=============================================================================



Joe Smith
jes@eniac.seas.upenn.edu

University of Pennsylvania
Department of Chemistry
231 S. 34th Street
Philadelphia, PA 19104
(215) 898-4797