Re: [tlug] Re: [OT] Say _no_ to the Microsoft Office format as an ISO standard

["Stephen J. Turnbull" <stephen@example.com>]

Josh Glover writes:

 > > Worse, they are known for producing really nice betas,
 > 
 > You must have tried out different betas than I did. ;)

I'm with Ken.  The last Windows beta I tried was NT 3.1J.  It was
smooth and much better than Windows 3.11 until it came out of beta, I
got the comp of the release, installed it, and the first boot fired up
the Buslogic driver recently patched to use undocumented instructions,
which confuxed my American Megatrends SCSI hosed, which proceeded to
wipe sector 0 and the next 6000 or so sectors of my hard drive.  All
of which was the Windows OS, fortunately all data I was pretty sure I
wanted to wipe anyway.  All typos intentional. ;-)

 > Um... no-one was using Unix on a PC at that time.

Well, of course not.  It wasn't until 1985 or so that Intel released a
CPU (the 80386) that it made sense to port Unix to.  The 80286 could
be done, but the 80286 Unices (IIRC SCO had the first) were hardly
more than proof of concept.  I think Steve Baur had an AT&T 3B2 in the
early 80s, though.

 > Agreed, though I believe 386BSD hit about that time,

386BSD became usable as a personal computer OS on commodity hardware
only around 1992 or maybe '93; it lagged Windows 3.1 by at least 3
years, and didn't have a fully working X11 distribution at that time,
and almost no working drivers.  PC-DOS was 1981, MS-DOS a couple years
later when the clones took off, Windows 2.0 was 1985 or '6, Windows
3.0 was 1987 IIRC, 3.1 was '88 or '89.

 > but unless you already used Unix at work or uni, you probably had
 > zero chance of learning about it.

There was a README on WUSTL's archive or something, or maybe I saw it
on comp.os.msdos.desqview.  I tried it '93, but it wasn't until '95
that I needed a server OS, and DESQview/X on DOS was 100 times more
stable and 1000 times faster than X on 386BSD.  (Memory starvation
problem on my part; I have no idea what would have happened if I'd had
a box with 64MB, I only had 32MB at the time. :-)

 > Nor could they, really. Linux was, as anyone who tried it during that
 > period will attest, for C hackers only.

Hrm.  I think I tried 0.99.7 in early '94, but it didn't have X.  Then
SLS Linux came out in mid-'94, with X.  I moved from that to BOGUS
Linux, then to Slackware by Jan 1, '95.  I guess I was sort of a C
hacker (I did half of the the DESQview/X port of Ghostscript, not that
it required much) but I never hacked kernel or libc code.  The closest
I came was bug reports (one to the people who wrote the BusLogic
driver, to tell them about my experience with "undocumented
instructions", and one to Don Becker about the 3C509 driver).

I think it was really more a matter of having a silicon thumb than
actually being a hacker.