Bryan’s talks are unparalleled entertainment, but it’s hard for me not to
laugh at some of this with the benefit of more time passage.
There’s a lot of clean code worthy of inspection in Illumos, but I think
it’s bordering on the threshold of a historical project. Key subsystems
like the VM and network stack have seen little development since 2009. Sun
was a giant benevolent benefactor and nobody’s filled the void of working
on general infrastructure.
My gut feeling is Sun’s early ‘90s choices were inevitable death wish.
I’ve had the unsettling thoughts lately that Sun after this era was a
brilliant marketing company more so than anything else.
On Monday, October 2, 2017, <jason-tuhs(a)shalott.net> wrote:
Solaris: so bad I left the company.
Why was Solaris so much worse than
SunOS?
But that was a kick in the nuts to us engineers. The sytem v source base
was crap compared to sunos, a huge step
backwards.
So my crowd pretty much all left in disgust. There was a lot of
heartache over it. None of us knew about the business deal at the time, in
fact I think a lot of management didn't know.
[...]
Instead, they repeated the SunOS journey. Bryan and crew polished that
turd for years and got it sort of reasonable. [...] but they went for it
and got it better. Only to have it tossed away again. Yuck.
For anyone who hasn't seen it, there's a pretty good talk that gives the
two-minute version of what Larry has described, and then picks up the story
from there and runs with it over the next twentyish years:
https://www.youtube.com/watch?v=-zRN7XLCRhc
-Jason