I worked at Xerox Office Systems Division 1984-86. At Xerox they made hardware 10 years ahead of its time, and got the software finished in 5 years. Then, the

I worked at Xerox Office Systems Division 1984-86. At Xerox they made hardware 10 years ahead of its time, and got the software finished in 5 years. Then, the computer was 5 years too early, and too expensive. This was the problem with Xerox Star 8010 Workstation, and Macintosh Lisa which was very similar. What Bill Atkinson & Jef Raskin did at Apple was to make a screen-based picture language, store the pictures in the resource forks of files, and copy the Smalltalk software virtual-memory system, reducing the memory footprint by a factor of 4x. Star 8010 workstation didn't have this amazing advantage, and was too expensive (you needed 512KB or better yet 768KB to get anything done; the Mac needed only 128KB). Since most of the cost of a computer was RAM, the Apple innovations reduced the price 4x, to $2500, roughly, making the computer viable. It was not useful until Bill Atkinson wrote MacPaint and Apple wrote MacWrite and MacDraw, so you could do WYSWYG word processing on it for only $2500, plus the cost of a memorywriter printer. So the Mac priced like a PC but it did something new, and more importantly, it was a blast to play around with that computer, and internally the software was 20 years ahead of anyone else, it was a joy to read the programming manuals. Many companies (even Microsoft with MS-Word) immediately switched over to Mac development, and after getting stuff working on Mac, figured out how to port their software to the IBM PC. I got my first Mac (a Mac II color machine) in 1987. It was light years beyond the best Xerox-internal prototypes in that year. It lacked virtual memory, but with 1GB of RAM, 80 MB of Disk, and better software than the Xerox Star (MS-Word, by Charles Simonyi who invented Wyswyg/Bravo at Xerox), it didn't need it. I had left Xerox to go to grad school.