[TUHS] Early Linux and BSD (was: On the origins of Linux - "an academic question")

[Larry McVoy]

On Sat, Jan 18, 2020 at 03:35:40PM +0000, Jason Stevens wrote:
> Internet legend is that the rift was massive.????

It was and Jolitz was unfairly crapped on at a Usenix, the one after
the lawsuit.  At the time, everyone thought it was about copyright and
Jolitz told me that there was a lot of code that was unchanged from v7.
He knew that I knew the file system code and he said "Go look at bmap(),
it's byte for byte exactly the same".  He was right and there was other
stuff, it wasn't just one function.

This was after both BSDI and 386BSD were a thing.  At a Usenix conference,
the BSDI guys pissed all over him, put down his efforts on 386BSD,
it was obviously self serving because they wanted a business, but it
wasn't a nice thing to do at all.

Bill thought he had the goods on them for copyright infringement but
he held his tongue and took the abuse because he didn't want to help
the lawsuit.  In his mind, he was a martyr for the cause.

So, yeah, those guys didn't like each other.  I know all of them and
wished they had handled this better.

I heard about all this, I was probably at that Usenix or maybe someone
told me, I don't remember.  I felt bad for Bill and was in a position
where I could hire him, so I did.  He didn't really do anything for me
though we did have some fascinating discussions about the short comings
of C++ and mused about how a language would have to work to truly layer
"correctly".  The context was the SCSI subsystem, it was part replicated
protocol in each driver and part driver specific stuff and we were trying
to imagine a programming language that would let us have the replicated
stuff in one place and plumb in the specific stuff as needed.  Needless to
say, we didn't come up with an answer but it was a fun thought experiment.
He's a sharp guy, quirky, but sharp.