Re: : 20ish years old: Gray's survey of Computer Industry Laws

[warning: old guy telling ancient history.]

| From: James Knott <[email protected]>

| Perhaps they were referring to minicomputers, which were common back in
| the 80s.  They'd often be assigned a specific task and be used by a
| department, such as accounting.  This compared with the previous
| practice of everyone using terminals connected to a mainframe.

All these terms "mainframe", "minicomputer", "workstation",
"microcomputer", "departmental computer" were all descriptive,
connotative, and especially cultural.

This all relates the Clayton Christensen's Innovator's Dilemma.  Each
"band" of computing grew in capability and sophistication quite
quickly and as the bands overlapped, markets were created and
destroyed.

The PDP-8 minicomputer I used in 1968 was not more powerful than the 
Altair microcomputer of 1974.  But the culture of the minicomputer had 
advanced to the PDP-11/45 (introduced at about the same date) and could 
run UNIX (I first saw it running in 1974).  So the microcomputer could 
displace the mini from it's original niche but the mini was starting to 
displace smaller conventional computers.

(Personal annecdote: I was always interested in having my own
computer.  But I didn't buy a microcomputer early because they were so
useless for my purposes compared with the UNIX machines I used.  I was
able to buy my own UNIX machine in 1983.)

The VAX was a big success as a computer sold for business
applications, something earlier minicomputers were mostly technically
capable of but not culturally capable of.  It was a breakthrough
computer.

It was a big fat target -- many manufacturers tried to get into that
niche.  The big barrier to entry wasn't hardware.  It was partly
software: through a lot of work DEC had made VMS suitable for
business.  The HP 3000 and the IBM System/38 were other minicomputers
that were successful in business in that era.

To a technical user, the only respectable one of those was the VAX.
Almost all University Computer Science departments got a VAX to run
4.xBSD.

The small competitors hoped that UNIX would get them a cheap entry to
the VAX's space.  It mostly didn't work out that way: there wasn't
enough time between the reign of the VAX and the predicted reign of
OS/2.

A main architect of VMS transferred to Microsoft and his product was
Windows NT (Dave Cutler).  Somehow DEC supported / agreed / signed off
on this.

DEC tried to segue from VAX to Alpha (via OpenVMS and Windows NT) but
did a swan-dive.  The trajectory reminds me of Blackberry.  Both
companies had worthy products but execution details caused horrible
results.
--
The Toronto Linux Users Group.      Meetings: http://gtalug.org/
TLUG requests: Linux topics, No HTML, wrap text below 80 columns
How to UNSUBSCRIBE: http://gtalug.org/wiki/Mailing_lists