Miscellaneous Operating Systems/Hardware

Does anyone have any experience with this?

http://en.wikipedia.org/wiki/Interactiv ... orporation

It descends from SVR3.2.

I have Interactive Unix 2.2, but haven't got round to trying it on anything yet.

Believe I loaded a version onto a spare 386SX PC sometime in 1993. The OS didn't make any impression more persistent than how insufferably slow that machine was.

However it simply could not be as cretinous as SCO from 1992, where I had to use loopback NFS mounts to make up for the lack of symbolic links...

I ask because, for some odd reason, I'm getting an urge to set-up a 386 Un*x box. I think Interactive Unix might be a bit too esoteric for me, and I will likely have better luck with an older Slackware or FreeBSD distro.

Coupled with SunOS for Sun 386i systems, this would make it three Sun OSs that ran on 80386 processors; Solaris, Interactive Unix and SunOS for 386i.

I'm occasionally tempted to acquire a first gen (1987-88) Heath/Zenith 386 system and put *NIX on it, as this was the first "killer micro" I encountered. And it was pretty impressive, given that the Apple LaserWriter was faster for single-user jobs than my college's VAX-11/750 (provided you could do it in PostScript, with no storage).

Then I typically stop drinking for the night, and the urge passes by morning...

(We did later get a Sun 386i, and I am occasionally tempted by that notion as well.)

smj wrote: Then I typically stop drinking for the night, and the urge passes by morning...

Atleast you have something to blame for this temporary madness, smj

That headline font looks awfully apple-y.

However it simply could not be as cretinous as SCO from 1992, where I had to use loopback NFS mounts to make up for the lack of symbolic links...

You must be kidding. Please say you were kidding.

ClassicHasClass wrote:

However it simply could not be as cretinous as SCO from 1992, where I had to use loopback NFS mounts to make up for the lack of symbolic links...

You must be kidding. Please say you were kidding.

Sorry, but it's true. Sadly the accounting system I was charged with moving over from SunOS was built on top of Informix RDBMS + Informix 4GL, and had a number of paths* hard-coded by the knuckledraggers who gronked it up -- those paths had to work unless I wanted to be stuck going through the 4GL code. Needless to say, I wanted that project over ASAP so I could get sent to a client and therefore be immune from recall.

Anyway those NFS loopback mounts were the only way I could make it work. To be honest, I'm not even sure that crawling horror even supported longer than 14 character filenames (e.g. within a directory)...

Other people seem quite find of later versions of things marketed by SCO, and I'm sure they were much better. But this is why I would never, subsequently, have anything to do with that company - long before the (in)famous lawsuits filed under the company's name...


* The environment was, overall, in questionable condition. The local brain trust had paths like "/net/stilgar/disk3/foo/bar/..." hardcoded in dot-files, scripts, executables, this accounting package, etc. The Sun-3 named "stilgar" - among others - had been retired some years prior to this episode. The files from it had simply been copied over to a Sun-4/280 server, exported, mounted directly from the fstab on all other hosts under the path shown above, and left to rot in perpetuity...