Which only happened thanks to the AT&T tapes that were taken to Berkely, and the work of Bob Fabry.
If it wasn't for that, BSD Unix might never have happened.
> Anyone manufacturing new hardware had the choice of developing a proprietary operating system, which was a lot of effort and would die with the machine, or porting Unix to it.
Because UNIX's source code was available under a "cheap" license.
If it was an AT&T product like those mainframe OSes, that would never been a path they would take.
Portability was important for Unix's success. The other operating systems were free (as in beer) but wouldn't run except on the expensive machines they came on. ITS could have been free, and wasn't proprietary, and was (at the time) a better development environment than Unix, but it was limited to PDP-10s.
MS-DOS was proprietary but was more popular than Unix, because it ran on even cheaper hardware.
Nothing prevented the other companies to port their OSes, many of each were implemented in variants of Algol and PL/I, they just didn't saw a business value in doing so, and thus lost to the availability of an almost freely available OS, that universities which didn't want to pay for mainframe OSes, adopted to their existing computer labs, instead of getting more licenses.
Students at those universities then picked UNIX for their startups, for the newly workstation market, like Sun did.
There were only two operating systems written in high level languages before Unix: MCP and MULTICS. Burroughs mainframes had a 48 bit word, and Honeywell mainframes had a 36 bit word. Burroughs and Honeywell were interested in selling those machines, not in spending considerable effort porting their systems to competitors' hardware with smaller word sizes. There was no business value in doing this.
Unix, which was very much influenced by MULTICS, initially ran on 16 bit machines. It's a lot easier to scale up (to run on Vax) than scale down, which is why Stallman based his GNU project around Unix rather than attempt to port ITS.
Which only happened thanks to the AT&T tapes that were taken to Berkely, and the work of Bob Fabry.
If it wasn't for that, BSD Unix might never have happened.
> Anyone manufacturing new hardware had the choice of developing a proprietary operating system, which was a lot of effort and would die with the machine, or porting Unix to it.
Because UNIX's source code was available under a "cheap" license.
If it was an AT&T product like those mainframe OSes, that would never been a path they would take.