This is another example of "Bloatware is Proprietary" philosophy which, along with "Worse is Better," are the main reasons software is bloated despite lacking features (so users have to constantly reinvent wheels) and being of poor quality (aka sucking). Web browsers are so bloated that it's easier to write a C++ compiler and port the browser engine than it is to write a browser. C++ is so bloated that it's easier to write an OS and port GCC than it is to write a C++ compiler. It all goes back to UNIX in the 1980s and the way AT&T replaced open systems with UNIX systems, originally as a way to lower the bar for software quality and get more licensing fees with a minimum of work.
https://en.wikipedia.org/wiki/Open_Systems_Interconnection
>Prior to OSI, networking was largely either government-sponsored (ARPANET in the US, CYCLADES in France) or vendor-developed and proprietary standards (such as the System network architecture (SNA) of IBM and DECnet of Digital Equipment Corporation). An Experimental Packet Switched system in the UK circa 1973, also identified the need to define higher level protocols. The NCC (UK) publication 'Why Distributed Computing' which came from considerable research into future configurations for computer systems, resulted in the UK presenting the case for an international standards committee to cover this area at the ISO meeting in Sydney in March 1977. OSI was hence an industry effort, attempting to get industry participants to agree on common network standards to provide multi-vendor interoperability.
This scared AT&T because this was a Right Thing design made by smart people and none of their existing "programmers" were capable of implementing anything like this. UNIX companies needed a plan, so they came up with POSIX so they could call UNIX an open standard. It was proposed as a way of solving compatibility issues in the UNIX community, but the weenies didn't stop there. UNIX weenies then co-opted the term "open systems" for themselves and were able to force their shitty hack onto real operating systems, much like what's happening with JavaScript and web browsers today. Because "Bloatware is Proprietary," computer companies found it much simpler and cheaper to pay licensing fees to AT&T than to clone UNIX themselves to become "POSIX compliant." IBM's MVS with UNIX was OpenMVS, DEC's VMS with UNIX was OpenVMS, ICL's VME with UNIX was OpenVME.
Instead of good standards based on decades of academic and industrial experience that actually fixed problems with existing systems, AT&T and the rest of the UNIX companies like Sun chose their standards based on the number of users because "incompatibility" gave them an excuse not to fix them. More users makes bad standards harder to replace, so inferior quality software (worse) enables them to extract more licensing fees (better). Sun understood this phenomenon very well when creating Java. Brendan Eich even described it by fearmongering that if he didn't finish JavaScript in 7 days, the web would have been stuck with something worse. This is the anti-Right Thing.
Another thing you may have noticed is that OSI is about distributed computing, which is from a 1973 paper. UNIX weenies still try to claim that Plan 9 is somehow an advancement in distributed computing, which is complete bullshit.
After describing how much we know about making
computer hardware which is fault-tolerant...
So, what other problems do we have?
<strategic pause>
UNIX.
<mumbles of agreement from the audience>
Unix is a standard brought to us by academia....
<goes on to point out the obvious contradictions
to implementing fault-tolerant systems on top of
Open Systems -- continually equating the concepts
of "Open Systems" with "UNIX" as the propoganda
would have us believe... finishes up with pleas
to academia for Open Standards which are
compatible with Fault Tolerance...>
Why do "we" get all the blame? AT&T spawned this virus.
Perhaps we are at fault for not rising to the occassion and
writting an anti-toxin which vaccinated machines to cure
them of the virus. Hmm, perhaps that was what RTM's worm
was really after (after all, it did only affect UNIX
machines...).