PARC's founding director, George Pake, was an outstanding physicist in the area of nuclear magnetic resonance. Dr. Pake had been serving as provost of Washington University in 1969 when he was courted by Jack Goldman, Chief Scientist at Xerox. If Jack Goldman was chiefly responsible for Xerox founding, and generously funding, a research center, then George Pake was chiefly responsible for siting PARC in Palo Alto -- 3,000 miles away from Xerox headquarters.
Xerox PARC was the birthplace of many foundations of modern computing, including many aspects of the Graphical user interface (GUI), the mouse **, the WYSIWYG text editor, the laser printer, the desktop computer, the Smalltalk programming language, Interpress (a resolution-independent graphical page description language and the precursor to PostScript), and Ethernet.
Xerox is often critized for failing to commercialize on their amazing inventions. A favorite example is the GUI, initially developed at Xerox PARC for the Alto and then commercialized as the Xerox Star by the Xerox Systems Development Division. It is deemed a failure because it only sold approximately 25 thousand units. The first successful commercial GUI product was the Apple Macintosh, which built on the earlier Apple Lisa -- also a flop commercially.
What is misunderstood is the impact of those systems. It has taken at least two decades for much of the technology in these systems to be surpassed. The commercial viability of the interfaces and technology that PARC pioneered has become the standard for much of the computing industry, once their merits became common knowledge.
It is legend that Xerox management consistently failed to see the potential of many of the PARC inventions; and while there is some truth to this it is also an over-simplification. They certainly understood the value of laser printing, and advances coming from the non-computer focused part of PARC. (Most people don't realize there was a large number of researchers working on materials science at PARC, for example pioneers in LCD production.)
The work at PARC in the years since the early 1980s is often overlooked, but major work since then includes Ubiquitous computing aka Pervasive Computing, and Aspect-oriented programming to name but two.