A quarter century ago Grace Hopper gave me a nanosecond.
Admiral Hopper isrenowned as one of the first women in computing, and was certainly oneof the most influential of all computer scientists. One of her 1953inventions – the compiler – isprobably today the most important tool we embedded folks use. Can youimagine being stuck coding in assembly today?
Born 100 years ago this week (Dec 9, 1906) she was the third personto join the research team of Professor Howard H. Aiken,who had requested her months earlier and greeted her with the words,”Where the hell have you been?” Then he pointed to the Mark I electromechanicalcomputing machine: “There's the machine. Compute the coefficients ofthe arc tangent series by next Thursday.”
The Mark I was a five ton electromechanical beast driven by a 50foot shaft and 4 horsepower motor, with no conditional instructions.Need a loop? Glue two ends of the paper tape together.
Active, energetic, she reminded me of Buckminister Fuller, anothervisionary I had the honor of meeting, and another who is now departedand missed. Though pushing 80 at the time, she was far more dynamicthan most of us much younger folks attending her presentation.
She gave each of the 20 or so of us there a piece of telephone wire- the solid copper kind with thin, color-coded insulation – that wasabout a foot long. “It takes electricity one nanosecond to propagatethrough that wire” she lectured, hoping (successfully) to awe us withthe miniscule nature of 1 nsec.
Now, of course, a nanosecond is a huge quantity. My $1000 desktopcycles at 2.6 GHz, or about 385 picoseconds. I wonder what the Admiralwould have thought of these advances, for she died on January 1, 1992,when desktop machines ran at a paltry ten megahertz (100 nanosecondcycle times) or so.
Yet today I still use her example when explaining the concept of ananosecond to non-techies. I hold my hands a foot apart, mentallycalibrating the distance to the remembered wire. That trivial bit ofcopper disappeared years ago, probably crinkled into a ball oflab-bench waste. I miss it, for it forged a bit of a connection to theearliest days of computing, and to one of the industry's mostinteresting people.
Admiral Hoper is also often erroneously credited with finding thefirst software bug – a moth fried in relay contacts of the Mark IIcomputer. Like original sin that's another legacy we've inherited, onenone of us relish. But her bug wasn't a programming error.
Happy birthday, Grace ” we miss you!
Jack G. Ganssle is a lecturer and consultant on embeddeddevelopment issues. He conducts seminars on embedded systems and helpscompanies with their embedded challenges. Contact him at . His website is .
Very fine article … on a very fine and luminary person!
However, Grace Hopper, being the penultimate engineer, probably wouldn't have been very impressed with your 385 nanoseconds. After all, if you divvie this up, this equals about 1/3 of a foot, or about 4 inches. Now, if you talk about 10 picoseconds, now you are talking! For this is equivalent for light to travel a mere 1/8 of an inch! Why that is a mere 100 GHz; which still hasn't been invented yet!
– Ken Wada
I still the nanosecond that she gave me. It's in a book I asked her to sign after her keynote at a DECUS meeting.
The DECUS newsletter had a picture of me bent over getting my book signed.
I really wish I had been able to get a copy of that picture.
– John Davies
I also had the privilege to meet Admiral (then Commodore) Hopper almost a quarter century ago. She gave a talk when I was attending the 1985 NJ Governor's School in the Sciences. After her talk she gave us those same “nanoseconds” made of telephone wire – I may still have mine packed away in my basement. I remember that we had decorated the ceiling in our dormitory with nanoseconds for the remainder of the GSS program.
I enjoyed her talk, but naturally feel a bit guilty that in my youth I did not realize how fortunate I was to meet her.
Thanks for the very fitting remembrance of one of the most influential people in our field.
– Michael Gilbert
Here is one for the physicists- the 'signal' does not actually travel at the speed of light in that wire, does it? Of course we know that light travels at 0.3 meters/nanosec in a vacuum (about the same speed in air) but slows down in a medium such as water. But the signal travels via electrons in that copper wire, and even in a vacuum tube, the electrons never get close to the speed of light when traveling from plate to anode (maybe i am showing my age; time to go revisit the Physics books?).
I will not be surprised if someone who is 'up on physics' can shed more light on this (no pun intended).
– Bob Sovacool
The way I remember her describing it was to demonstrate how many nanoseconds it takes for a signal to travel to a satellite. She made a motion as if laying her nanoseconds end to end toward the satellite.
– John Davies