In addition to developing the cellular automaton that inspired a lot of the people who invented the modern computer industry, Conway made contributions in a wide range of mathematical subjects and in physics, producing important results even when he was past retirement age (even though mathematicians are known for peaking early).
Brought up is the CV famous people death thread. Here’s my post.
I realized later that this is the closest I’ve come to knowing someone who has died from this. Been in the same room with him, etc. But no personal interaction IIRC.
Back in 1971/72, the final project for the assembly language class I took at MIT was programming the Game of Life for our PDP-1, the same one where SpaceWar was developed on. You were judged by if it worked, and also if you handled interrupts fast enough so there was no flicker on the big CRT.
I got an A.
Sorry to hear of his passing. A true pioneer.
The last time he was in Montreal, maybe two years ago, he gave a talk at the University of Montreal (on how to generate the formulas for sums of squares, of cubes, of fourth powers, etc.) and the next day I got co-opted into driving him to see an old friend of his (well-known mathematician John McKay whom I’ve known for years) who was and is in a nursing home. Although I had met him before, this was the first time I had spent any time in his presence. Even then, he mostly used a wheel chair, although he could walk a short distance.
I just called a friend of mine who knew Conway well to give him the news who told me that McKay is also ill with Covid-19. Nursing homes are the worst places to be with this pandemic. But McKay and Conway are the first two people I know personally that I know have it. Apparently, McKay seems to be recovering, at least at the moment.
I was just remembering that the first time I met Conway he taught me a game he had invented he called football. The only thing I remember about it is that it is played on a go board.
Conway gave a talk at McGill I think in 1971 about the game of life. He invented the “game” in order to verify a conjecture he had made about the existence of a self-reproducing Turing-complete automaton. To this end he needed a couple of things, like a spaceship and a spaceship factory. He knew that it would require massive amounts of computer time, which was extraordinarily expensive the late 60s to discover these things so he made a game out of it and offered small prizes for interesting discoveries until he had what he needed. People with access to main-frame computers (there were no other kind) spent an awful lot of computer time doing this. John McKay arranged for the whole thing to be video-taped. Ten or twenty years ago, I realized what a wonderful historical artifact that tape would be and tried to track it down. McKay had no idea what had happened to it and the McGill AV office looked for it in vain.
I was a math major in college for a while until I realized that I enjoyed reading about math more than I enjoyed doing it. I still read a lot about math and it’s amazing how often Conway’s name comes up and in how many different contexts. He gave math a sense of fun that I get from hardly anyone else. Sad to hear that a major supplier of fun is gone from the world.
Omg, I bought Stewart Brand’s Amazon.com, half of which was about hacking Spacewar, at the Harvard bookstore in 1974, and it changed my life. I ended up getting a Ph.D. in theoretical computer science, and, yadda, yadda, I work at Red Hat. Brings back memories. Thanks!!
MrsRico and I programmed at a large financial firm in 1979. I found a worn copy of that 1970 SciAm with Conway’s LIFE featured. The bosses didn’t want us writing games in COBOL so I first graphed it on our home greaseboard, alternating colors per generation. Slow… A Heathkit H8 arrived for us later and I did write a COBOL-80 version. Somewhat faster… Didn’t early digital firms tout LIFE as a benchmark for their sizzling systems?
I had a friend at Berkeley in those far-off days who later went on to become a Somebody in computer cryptography. He figured out an algorithm that updated an entire row of a Game Of Life field all at once.
The field was encoded with just one bit per pixel. The algorithm took three consecutive rows as inputs, and consisted of a series of boolean and shifting operations, the end results of which was the next generation of the middle row.
He presented the algorithm to me simply as a sequential list of those boolean and shift operations. I coded it in assembly language on two machines: A Control Data 6400, using one 60-bit word per row (and with 60 rows), and a PDP-10 using six 36-bit words per row for a total of 216 bits per row (and with 216 such rows). In each case, the field was a torus: the top and bottom wrapped around, and the left and right edges wrapped around.
The algorithm ran like a bat out of hell!
I have no recollection of how I displayed the field on the 6400. The PDP-10 had a display monitor that I was able to program somehow. That version of the program also had a single-step mode, so you could view one generation at a time. With the 216x216 field, you could play with some awesome configurations!
I’m pretty sure I still have that algorithm around among my papers somewhere.