It was before my time, but according to Wikipedia:
"SCO XENIX for the PC XT shipped sometime in 1984"
"The first version of OS/2 was released in December 1987"
It was before my time, but according to Wikipedia:
"SCO XENIX for the PC XT shipped sometime in 1984"
"The first version of OS/2 was released in December 1987"
The software that many BBSes ran in the 90s (at least in my area code and the surrounding) was one of (a) Telegard[0] (b) T.A.G. or (c) a Forum hack of some kind[1]. These were all DOS-based and were impossible to get working in a two-node setup under Windows. Desqview and Desqview/X worked relatively well but still meant I had to shut down the BBS when my Dad wanted to use Lotus 123. Everyone I knew that had a multi-node setup ran OS/2 for a while. It was easily pirated, and inexpensive enough for the relatively honest amongst us compared to commercial UNIX offerings but the important reason was that it easily ran my current BBS software.
[0] And its many hacks, Renegade being a large one. Ironically, I don't think the original Telegard was multi-node. I ended up re-writing Telegard, myself, and added multi-node support using some of the leaked, I think 2.25, version.
[1] Run by the "Elite" :P
Xenix was originally developed by Microsoft. It offered DOS compatibility: https://news.ycombinator.com/item?id=12661405
[1] http://bitsavers.org/pdf/apple/lisa/xenix/SCO_Price_List_Dec...
The teacher would bring a single tower (486 based I think), plug it into monitor and keyboard, and the whole class would take 15 minute slots at it.
We would prepare our exercises in MS-DOS, with Turbo C 2.0, using mocks for the UNIX APIs.
Then use those 15 minutes to replace the mocks with the real deal, and have a go at it.
Failing the exercise, it was back to MS-DOS / Turbo C, while trying to figure out with pen and paper what might have gone wrong.
Then people complain nowadays when their build takes longer than 1 second.
My father wrote in Fortran during the 70's while at the university. He told me that he left the programs (stack of punched cards) with the "computer operator" on Friday and went back to get the "results" on Monday. Sometimes, the result was simply a syntax error message.