Some observations on life and computing
The number of parallel programs I have embarked on is less than
the number of deadlocks I have created.
I have never written a program with pointers where I did not try to
traverse a circular list.
In life, everything is important.
Everything else is not.
Parallelism decreases as maturity increases.
Parallelism decreases as flexibility increases.
H. T. Kung
I want to build everything as simply as possible, since everything
After floating point, no one ever thought again.
Don't let that trouble you. I refuse to discuss that assumption.
Existence is trivial from the finiteness of the universe.
I don't know how to teach it to students, and I can't do it myself.
Maybe those two are related.
I took a vow 27 years ago never to do another divide.
The good Lord made graduate students, and if He ever stops, we're dead.
It would be strange to me if shared memory machines were the way to go.
You have to stand there with your wire-wrap gun before a programmer
will take you seriously.
Even if the machine is a flop, you will need a good programming
environment for it.
VAXes are the marijuana of the university crowd.
This is like collecting statistics on undetected murders.
If we lose a data record, it's of no consequence. We have plenty more data.
Don't worry about semantics. I'm dealing with abstract engineering concepts.
You could have a hot computer, a warm computer, a cold computer,
each running off the heat of the previous one.
If God, having some kind of fit, threw the gearshift of time into reverse,
the spoon would fly off the table.
The only reason no one ever knew this before is because no one wanted
to know this before.
It is now crystal clear (to a very small number of people).
This is sort of a terribly interesting idea.
To become a physicist, you must sign in blood that you won't be upset
by things that make no sense and can't be explained.
When you widen the diameter of knowledge, you increase the periphery
A good idea is not necessarily a bright idea.
It never hurts to have a supporting argument for something people
are already doing.
Testing is not a religion I have to believe in.
This isn't yesterday's lecture. He's getting his arithmetic right.