Slashdot Mirror
MS-DOS Paternity Dispute Goes to Court
theodp writes "Might be more interesting as a Who's-My-Baby's-Daddy? segment on Maury, but a Court has been asked to decide the parentage of MS-DOS. Tim Paterson, whose operating system 86-DOS (aka QDOS) was sold to Microsoft in 1980, is suing author Harold Evans and Time Warner for defamation. In his book They Made America, Evans devoted a chapter to the late, great Gary Kildall, founder of Digital Research, describing Paterson's software as a 'rip-off' and 'a slapdash clone' of Kildall's CP/M."
It's main purpose was to be as compatible as possible to CP/M to faciliate fast porting of CP/M applications to QDOS.
- jeps
I would be proud to have MSDOS on my resume, as would most serious software architects. MSDOS was used by millions of users, it was a true groundbreaker. MSDOS does not do much compared to VMS or VM/CMS but what it does it does on an 8/16 bit processor running at a few MHz. The original Microsoft Basic was not exactly extensive but most people would agree that it was a cool piece of coding.
But you miss the point in any case. This guy has MSDOS on his resume, what he is objecting to is the claim that he stole it.
Looking for an Information Security student project suggestion?
Try http://dotcrimeManifesto.com/
From: korpela@albert.ssl.berkeley.edu (Eric J. Korpela)
Newsgroups: alt.folklore.computers
Subject: Re: filename separator change in CP/M and MS-DOS
Date: 7 Jul 1998 01:47:52 GMT
>The legend runs something like this:
> 1. The first version of MS-DOS was actually QDOS from Seattle Computer Works
There is much ongoing discussion as to whether it was ever called QDOS.
There is a general consensus that at various times it was called 86-DOS
and SCP-DOS. I belive the real name of the company whas Seattle Computer
Products.
> 2. QDOS ("Quick & Dirty OS") was an unauthorized port of CP/M to x86.
> CP/M ran on Z-80's.
There is little doubt that it was an unauthorized port. (In the US, at least)
No authorization is required to reverse engineer a product. There is much
debate about whether an of the "port" was accomplished by running a disassembly
of CP/M through Intel's 8080->8086 assembly code converter. (This would
be illegal in the US).
The typical (apocryphal) story is one of special key sequences that would
bring up a Digital Research Incorporated copyright notice in early versions
of DOS. (At this point, I've never seen a special key sequence that would
bring up such a notice in any real CP/M version.)
BTW, the CP/M version in question was written to run on the Intel 8080
chip. The ability to run it on the Z-80 was a consequence of the Z-80
design, not vice versa.
> 3a. CP/M used "/" as the separator between components in pathnames
False
> 3b. alternative version: CP/M did not have directories, so did not need or
> use any kind of slash as a pathname piece separator.
The alternative version (3b) is correct here. CP/M did not have directories
other than numbered user areas. In CP/M the '/' character is for command
switches, a trait it inherited from Digital Equipment Corp operating systems
on which it was patterned.
> 4a. QDOS and hence MS-DOS used "\" as the pathname separator to disguise
> the origin of the ripped-off software (unauthorized port from CP/M).
False, this is far too little to disguise the nearly identical APIs of
CP/M and early versions of DOS.
> 4b. alternative version: CP/M and hence QDOS and MSDOS used "/" as an
> option separator to commands, hence it was not available for use
> as pathname separator.
Correct.
Eric
Sadly, he's the exception. The entire computing business (and engineering business, and any other business involving creativity and intelligence) is replete with stories like this. Kildall is just an unusually extreme example.
There's an interesting History of MS-DOS By: Leven Antov at http://www.digitalresearch.biz/HISZMSD.HTM
Um, you don't have the vaguest idea of what you are talking about. Windows NT 3.5 right up to Windows XP are not built on top of DOS. They do not require any DOS commands or interrupts to work. In fact, their support of DOS is totally shitty as compared to the OS/2 VDM or the Linux DOSEMU system. I have a number of DOS programs that work great both in OS/2 and DOSEMU but don't function at all in NT's DOS subsystem. Whatever XP's flaws may be, it is not at all built on DOS, and neither are its predecessors right back to Windows NT.
The world's burning. Moped Jesus spotted on I50. Details at 11.
You think you're joking, but DR-DOS, née CP/M-86, was indeed owned by Canopy/SCO/Caldera for a time. They purchased it, like Unix, from Novell, who had previously bought Digital Research.
echo 33676832766569823265328479713269.8639857989Pq | dc
CP/M-86 wasn't available until after IBM committed to shipping MS-DOS licensed from Microsoft.
MS-DOS dominated the market for one reason and for one reason only -- IBM chose it as the main OS for the PC
You make it sound as if customers dind't have a choice. IBM announced and made available three operating systems - PC-DOS, CP/M-86 and UCSD P-System.
Because Microsoft delivered a working product a year in advance, IBM wrote it's own programs around it. Also, DR charged a much higher licensing fee for CP/M-86, which IBM sold for $240. But there were no programming languages available for it yet and very little software had been ported over from CP/M to the CP/M-86.
If IBM made PC-DOS as "the main OS" for the PC, it was because it was available earlier and had lots of programming languages available. Customers also liked it because it was cheaper.
since the lawsuit is over whether QDOS was a "slapdash clone" of CP/M. Which, in point of fact, it was.
A clone with a completely different file system? There were plenty of CP/M clones in those days, QDOS, later 86-DOS, later MS-DOS wasn't really a clone. It just offered a familar API set for programs porting from CP/M.
the biggest of which was using one of the worst OSs ever made. Not by today's standards, but by the standards then.
Okay, well, what would have been better then for a macine with a 16-bit processor with a 8-bit bus and 16K of memory? Microsoft originally wanted to license XENIX to IBM, but it would never work on that type of machine.
In no way did Tim Patterson rip off CP/M. It is exceedingly clear from several respecible published sources that DR shot themselves in the foot time and time again, while Microsoft delievered not only a operating system, but the programming languages for it - which was the real draw.