Sure We're Open Source - Not
Alas, apparently that silly press release last week has totally confused the writing fraternity into thinking the 1990's were actually the 1980's (see Fun Friday - Homer's Illiad to be "Improved" for Silicon Valley). Aside from the fact that Clinton and Reagan were both absolutely adored by the American people, I don't think the 1980's were really enough like the 1990's to easily confuse the two decades, do you?
In this piece in cnet, "Unix got its start at AT&T, but Sun co-founder Bill Joy was instrumental in an open-source variant developed at the University of California at Berkeley. For half the company's history, Sun used this BSD version of Unix in a product called SunOS". I wonder what they've been drinking today?
Back in the 1980's, BSD required an AT&T source license to obtain source code - a considerable expense for a company. SunOS required an AT&T license as well. At Symmetric Computer Systems, to sell our BSD-based Symmetrix 375 computer required an AT&T license and we only used BSD - not System V! Afraid it was true for everyone. But that's not all.
BSD just wasn't open source then. You paid to play - big time. Our first investment from venture in 1983 came in the door in $$$ and went out the door in license and prepaid royalties to AT&T, just like at Sun - want to see the check stubs? They were part of an exhibit several years ago on the story of a BSD workstation startup circa 1984. (One young man who I chatted with at the time said "It wasn't a golden age, but.... well, it was a golden age, wasn't it").
It took years and a careful process by Berkeley to document and remove proprietary source from BSD, as it was very intertwined. Even with best efforts, they had a lawsuit from AT&T / USL that was only settled in 1994. And it took years of uncompensated effort (not backed by Sun - actually, Compaq was the one who donated machines) to create an open source X86 version - thoroughly documented in one of the premiere (code and all) hard tech magazines of the time.
And it wasn't until 1992 that a completely open source BSD became available to the public. It's called 386BSD, and it's well documented in Dr. Dobbs Journal from 1991 onwards.
Oh, by the way, Solaris isn't BSD, just like Mach (from Apple) isn't BSD. Solaris is almost an "un-BSD" in many ways architecturally - a classic Fred Brooks "second system syndrome" in action. So invoking BSD when they were so eager to dump it seems a bit ahistorical, doesn't it?
So let open source Solaris stand up for itself - it doesn't have to hide behind BSD's history to shine on it's own. And it really does have to stand on its own to beat Linux at their own game.