Phillip Remaker
user of dozens of common and uncommon operating systems.Updated 8y
Microsoft did, in fact, have a Unix based operating system called Xenix in the late-DOS early-Windows timeframe. I had a PC/AT (80286 based) in 1989 running Xenix with about 1MB (yes, MB) of memory. It was never more than a niche product for Microsoft.
You have to remember that the Microsoft of the 1980s was writing software for the personal computers of the time, which were significantly underpowered relative to the so-called minicomputers running Unix at the time. Multiuser was not an issue for Microsoft (it was, after all, a “personal” computer) and even running multiple apps at the same time was not a priority.
The first class of OSes that Microsoft made were really nothing more than primitive program loaders. Although Windows added a GUI, it only added multi-application capability later in its evolution. Prior to Windows NT, Windows never had serious really multi-user capability because the underlying filesystem had no sense of multiple users or anything more than primitive file security. Windows for Workgroups added multi-user, but only network based files had any file security.
Windows NT (the ancestor of modern Windows) was Microsoft’s first serious multi-user OS. Windows NT attempted to leapfrog Unix with a more sophisticated sense of system and user identity as well as much more fine grained access control in the file system, including a well defined distributed directory (Active Directory). A lot of what Windows did was, in fact, superior to the ancestral models of file access, distributed directory and user identity in Unix. Their development team came out of DEC, so the roots of the OS have more in common with VMS than Unix.
Consider also that the Unix licensing was a wild mess in the Windows NT era. Lawyers in 1993 might be nervous about AT&T asserting their rights to Unix. The USL v. BSDi lawsuit cast a long shadow over the future of Unix. Most “free” variants of Unix descended from BSD, which was on shaky ground as a product that could be commercially sold. If Microsoft wanted to make a commercial Unix, they would either have to risk being sued or pay royalties to AT&T. Coming off a troubled and tumultuous relationship with IBM, Microsoft elected to forge a path ahead where they controlled their own destiny.
16.7K views
View 154 upvotes
Jim Dawkins
· 7y
Because Bill Gates and Steve Jobs wanted to do the exact opposite. They wanted to build a computer that was intuitive and useful for the layperson. I would say they were pretty successful.
Jerry Coffin
· Jul 20
Although Windows added a GUI, it only added multi-application capability later in its evolution.
Not so. Although it lacked the memory to do so for any more than minuscule programs, Windows 1.0 had all the mechanism in place to run multiple programs concurrently.
Phillip Remaker
· Jul 20
Only if they were specifically written for Windows. It wasn’t until the later versions that multiple DOS applications could run.
Jerry Coffin
· Jul 20
Right—and admittedly, when Windows 1.0 first came out there (obviously) weren’t many native Windows applications for it to run. From a practical viewpoint, the difficulty was less with getting Windows to run multiple programs, and more with finding even one program you cared about that ran in Windows at all.
Lee Hauser
·
Follow
reformed Mac and Linux user, resurrected Windows enthusiast8y
Your question shows a lack of understanding of personal computing history.
When IBM was developing the PC, the dominant microcomputer operating system was CP/M. It was designed to work on 8-bit processors such as the Intel 8080 and Zilog Z80. IBM chose to use a more powerful microprocessor, the Intel 8088 (an 8/16-bit hybrid). Digital Research, the developer of CP/M, had a 16-bit OS called CP/M-86, and IBM wanted to use it (this is extraordinary, of course, like the whole development of the PC…a small, almost-independent development team building a home computer out of off-the-shelf components, completely uncharacteristic of everything IBM had ever done).
It’s now the stuff of legend that for any number of reasons IBM and DR didn’t reach an agreement for CP/M, and IBM approached Microsoft, which was developing BASIC and other languages for the IBM PC, to provide the OS. Gates and Allen went to Tim Patterson at Seattle Computer Products and bought his embryonic 86-DOS, renamed it IBM-DOS (and MS-DOS for non-IBM clones), and the rest is history.
DOS was the first cash cow for Microsoft; the magic letters “IBM” legitimized the personal computer, and Microsoft had no reason through most of the ’80s to rock that particular boat. The very nature of the market, thanks to the domination of personal computers running MS-DOS, dictated that improvements be based around the hardware and software that drove that market, which is why versions of Windows prior to Windows 95 were graphical shells sitting on top of DOS, and why versions from Windows 95 through ME had 16-bit underpinnings. Not until Windows 2000 did Microsoft abandon 16-bit DOS, making Windows 2000 both a consumer and server OS with a 32-bit (Windows NT) foundation.
As has been pointed out in other answers here, Microsoft didn’t release XENIX until the 286 came out, and it was never intended to be a consumer product. Unix at the time was an academic OS, used primarily for research and academia. As has also been mentioned, when Microsoft developed their higher-end OS, NT (which came out after MS and IBM split up over the development of OS/2), its roots were in VMS, not Unix. If Gates was ever even interested in using Unix in 1981, he wouldn’t have been able to in the time IBM gave him to come up with an operating system for the PC.
18.7K views
View 47 upvotes