>>>Just try to remember the other personal computers at the time - the 8 and 16 bit world of then. There weren't any interchangeable parts. The guys couldn't even agree to the floppy size - Schneider was pushing 3", Atari and a few others were on 3.5".
>>
>>Floppies were 5.25 or larger - but formated differently. 3.5 came much later...
>
>My point exactly - even after switching to smaller floppies, they still didn't have a standard format. Atari, though, could read DOS floppies, and could write to them, but whichever floppy it formatted wouldn't open on a PC.
>
>>>And since the advent of IBM PC, you could have six machines from different manufacturers, disassemble them, shuffle the parts, assemble at random, and at least four of these would work.
>>>If that's not a revolution, I don't know what is.
>>
>>Then Compaq was the real revolutionary, as they about defined how to clone without opening the doors for lawyers...
>
>Maybe, but it was a step away from interchangeable parts. Compaq had some cards which wouldn't work on other machines, and some standard cards wouldn't work on some of their machines.

Compaq was an early example of "embrace and extend". Some of its innovations were more successful than others:

- IIRC until its 386/20, all micros ran the CPU, memory and expansion bus at the same clock speed. That machine separated (at least) the CPU bus so it could run faster. There was a temporary plateau in performance when everyone was stuck at 386/16; that machine pointed the way forward

- I believe they introduced so-called "IDE" hard drives. Until that time you had smart HD controller cards and dumb drives. IDE put the smarts on the drive itself. At the time they were a lot more expensive than regular drives, and compatible only with Compaq controllers but it could certainly be argued they were merely a little ahead of the times

- They used !@#$ Torx screws for their cases :(((