Vikram Chandra - Geek Sublime - The Beauty of Code, the Code of Beauty

Mort does not want to understand how an entire system works in order to tinker with it. And my goodness, Mort hates reading documentation … Mort’s primary job is to frobnicate widgets — code is just a means to that end — so every second spent making the code more elegant takes him away from his primary job. 10

Mort lacks “mechanical sympathy,” that quality possessed by the best race-car drivers, who understand their machines so well that they flow in harmony with them.

To the Morts of the world, and even to the Elvii, Mel the Real Programmer’s programming is inscrutable and his mystique dazzling. The narrator of our epic is asked to investigate and change the behavior of a program that Mel has written. He reads through Mel’s code, and is baffled by an “innocent loop” which doesn’t have a test within it — as is usual — to break the loop. Code loops always contain a conditional test of the form “if numberOfLoops > 4 then break”; without such a construct you are trapped in an endless circling repetition. “Common sense said that it had to be a closed loop, / where the program would circle, forever, endlessly.” 11But Mel’s program doesn’t get stuck in the loop, it flows through, it works. It takes the narrator two weeks to comprehend Mel’s uncanny melding of code and machine, which uses the test-less loop and a programmer-forced malfunction in the system’s memory to position the next program instruction in the right location; such is the force of this revelation that “when the light went on it nearly blinded me.” After such knowledge, reverence is the only proper emotion; the narrator tells his Big Boss that he can’t fix the error because he can’t find it.

I didn’t feel comfortable

hacking up the code of a Real Programmer. 12

Despite the allusion above to “the *macho *side of programming,” the non-geek may not fully grasp that within the culture of programmers, Mel es muy macho. The Real Programmer squints his eyes, does his work, and rides into the horizon to the whistling notes of Ennio Morricone. To you, Steve Wozniak may be that cuddly penguin who was on a few episodes of Dancing with the Stars , and by all accounts, he really is the good, generous man one sees in interviews. But within the imaginations of programmers, Woz is also a hard man, an Original Gangsta: he wired together his television set and a keyboard and a bunch of chips on a circuit board and so created the Apple I computer. Then he realized he needed a programming language for the microprocessor he’d used, and none existed, so Woz — who had never taken a language-design class — read a couple of books, wrote a compiler, and then wrote a programming language called Integer BASIC in machine code. And when we say “wrote” this programming language we mean that he wrote the assembly code in a paper notebook on the right side of the pages, and then transcribed it into machine code on the left. 13And he did all this while holding down a full-time job at Hewlett-Packard: “I designed two computers and cassette tape interfaces and printer interfaces and serial ports and I wrote a Basic and all this application software, I wrote demos, and I did all this moonlighting, all in a year.” 14

That second computer was the Apple II, the machine that defined personal computing, that is on every list of the greatest computers ever made. Woz designed all the hardware and all the circuit boards and all the software that went into the Apple II, while the other Steve spewed marketing talk at potential investors and customers on the phone. Every piece and bit and byte of that computer was done by Woz, and not one bug has ever been found, “not one bug in the hardware, not one bug in the software.” 15The circuit design of the Apple II is widely considered to be astonishingly beautiful, as close to perfection as one can get in engineering.

Woz did both hardware and software. Woz created a programming language in machine code. Woz is hardcore.

Working in machine code is very hard, so assembly code was created by adding some mnemonics to machine code. Working in assembly code is still hard; doing anything complex — like making games — in it is insanely hard. On programmers.stackexchange.com, a user going by the nom de guerre “DFectuoso” asked, “Are there any famous one-man-army programmers?” 16During the ensuing discussion about lone coders, one of the participants mentioned Chris Sawyer, who wrote the hugely successful 1999 game, RollerCoaster Tycoon : “He had a little help with music and graphics, but otherwise RollerCoaster Tycoon was all him. Amazing, especially given the physics engine. Last but not least, the entire game was written in assembly language.” And another commenter responded, “He wrote that in assembly?! Jesus Christ. I think I need to go boil my brain now.”

Sawyer’s achievements are indeed brain-boilingly immense, but why and how does a lone Scottish geek toiling obsessively over his virtual roller coasters become a “one-man-army”? Why do the Einsteins of programming affect — in their online personas and sometimes in person — a blustery True Grit swagger? Therein lies one of the tales that Nathan Ensmenger tells in his illuminating social history of computing, The Computer Boys Take Over: Computers, Programmers, and the Politics of Technical Expertise. Ensmenger does for computing in North America what historians of science have done for other disciplines: locate the development of knowledge and technology firmly within a messy matrix of human agency and politics; there is no orderly teleological progress from triumph to triumph, only competing interests that struggle over authority, access, and power. The computer boys of Ensmenger’s title are the early software-builders, the pioneers; the irony, as Ensmenger shows, is that many of them were women. In fact, the earliest programmers were all women: the “ENIAC girls” were women recruited by the (male) engineers and managers of the Electronic Numerical Integrator and Computer. The creators of the ENIAC had a clear division of labor in mind: the male scientist or “planner” would do the hard intellectual work of creating the mathematical algorithms and structures necessary to solve a problem; the female “coder” would then carry out the “static” manual labor of entering this plan into the machine by manipulating plugboard wiring and thousands of switches. Numerous wires had to be plugged and switches flipped for each machine instruction; each problem required thousands of instructions, so wiring — which was the programming — would take several days, and checking the wiring would take another few. 17

This division of tasks of course echoed the hierarchies already present; men did the thinking and inventing, women were clerks. “The telephone switchboard-like appearance of the ENIAC programming cable-and-plug panels,” Ensmenger writes, “reinforced the notion that programmers were mere machine operators, that programming was more handicraft than science, more feminine than masculine, more mechanical than intellectual.” 18The planners considered the coding process so transparently simple that they couldn’t imagine that once in the machines, their algorithms might fault and hang, might need to be stopped. One of the ENIAC programmers, Betty Holberton, had to work very hard to convince John von Neumann that programs were complex and therefore fragile: