So when you do your "research" on google (rather than having had any actual technical experience with the thing itself) you're quite likely to get real confused about which part played the role of the compiler (the p-code generator) and which part played the role of the operating system (the UCSD Pascal system-- an operating system written in pascal, NOT a compiler!)

And yes, the Lisa and Mac OS were bad hacks of the UCSD Pascal OPERATING SYSTEM, which was, yes, compiled with a different compiler than the original p-code generator. They didn't want to use the p-code system, because otherwise people would be able to run the derived Lisa and Mac OS's on, um, Modcomps and Teraks! By just copying the p-code to a different machine. That was why the p-code system was so great, it was like Java WAAAAY before its time.

But to think that the p-code system was the SAME THING as UCSD Pascal is just as stupid as thinking C is the SAME THING as Unix.

1. That if Apple obtains an exclusive license to all UCSD Pascal future source development and thereby makes it impossible for either university researchers or other commercial outfits to use it, it is "incorrect" to call this a "purchase." It's a license . Sheesh!
2. That compiling an OS on a new machine with a different compiler is not a "port" of the OS. Thanks, MacJerry, or MacJefRankin or whatever your name is for informing me that port of UCSD Pascal to the Lisa and MAC were NOT PORTS...because they used a DIFFERENT COMPILER.
3. That because a lame idea someone published in their thesis resembles a product developed by PARC, PARC stole the idea. Even if the original lame ideas were developed into another product by another company that was discontinued after six months. This is PROOF that Macintosh did NOT steal its UI from PARC!

I have also learned many interesting things about SGI's, even though I developed for them for several years.

1. Code that compiles out of the box on several versions of gcc and several distros of linux, unix, and windows...but dies a horrible death on IRIX C...this is the fault of the code, not a problem with IRIX CC.
2. That lack of support for nested template instantiation under IRIX CC is a GOOD thing (tm).
3. That having four different incompatible floating-point representations on the same machine (IRIX again) is a GOOD thing.

Oh, where would I be without the 3l33t /.ers to tell me these useful things. They're almost as helpful as the trolls on net.women.