The 360/20. It's seems unlike to me that it was
really the inspiration for
the minicomputer families. They had already build minicomputer-class
machines in the past (e.g., 650, 1401, 1620/1710), and were fairly
successful with them.
Maybe not an inspiration, but certainly a rebirth. S/360 was supposed to
do all the jobs for business (not counting process control or lab use).
I think what caused them to scale up their efforts
was more a matter that the perceived they were losing a lot of market
share to DEC (and other minicomputer vendors).
I could see this.
There were some intermediate stops along the way, such
as the 1130/1800,
and the System/7.
I think these were process control machines (great for running
factories). I think IBM always treated these as a whole different branch.
There is an amazing lack of information on the S/7 on the web.
One of the most interesting things about Series/3 was
that they invented
a new punched card format for it.
Not all of them used the things, however.
I recently acquired one, and wish I had time to test
the "unhackable" part.
The newer machines are based on the PowerPC architecture (with some
extensions), and may be more "hackable" than the old CISC machines.
I don't think having a PowerAS based system will help. I would think the
older machines would be easier to play with.
William Donzelli
aw288(a)psfn.org