The Best Programming Advice I Ever Got (2012)

People who get upset if you touch their code. I think in your other comment you missed the fact that there was already an ongoing debate about this particular part of the code, and it wasn't going anywhere. In such a case a demonstration that the other method is faster is more compelling than more fruitless debate. Speculations on other reasons why the code should've used 2 processes is useless, as it is clear from the article that this change made it possible to get things done faster, no need to argue with that.

I have been there in that sort of situation where managers twist their eyebrows at an efficient or productivity-enhancing solution. The essence of the java enterprise consulting business is fixing an unnecessary complicated mess with a similar, new buzzword backed mess.

So the interesting thing here is how bureaucracy and efficiency seem like natural enemies; a subject which always reminds Terry Gillian's Brazil movie where the terrorist guy is the one who fixes things.

> In my opinion I think the author did try to address the big picture.

Sure. But grandparent's point is that he did it backwards: first he made the changes he wanted to see, and then he demoed it up the chain and got himself in hot water. Had he first taken time to develop a thorough understanding of the situation and the environment in which it had origin, he'd have been able accurately to evaluate how best to take action in order to bring about his desired result -- or indeed whether it might instead be best, despite what had originally seemed to him a desperate need for improvement, simply to cultivate the detachment necessary to take no action at all.

I think that's probably telling in itself. I've GTFO'd twice; one time was by my own choice, while in the other I had to be helped along by the organization a few months after I failed to follow my sensible instinct to resign. Both times, once I was finally out the door at last, I found myself feeling like a weight had come off my shoulders.

I could be wrong, but I don't see any way it's possible for that feeling to come along with a lurking realization that I'd made a mistake. (Certainly it never has in my experience, at least, and I've made plenty of mistakes.)

(In the following, we will stipulate that your "rogue programmer disrupts Photoshop, resulting in a better product and getting half of Adobe fired" situation is plausible, for the sake of illustration.)

The half of Adobe you posit getting sacked wouldn't say it was worth it, though, and from a certain perspective -- namely, their own -- they'd be right.

After all, Photoshop is already the unquestioned leader in its field, and the closest competitor isn't even within shouting distance, so what good does it really do Adobe's business if Creative Cloud's Photoshop component sucks 20% less?

(Yes, no one is happy with the change from Creative Suite to Creative Cloud; this I concede. But those who are unhappy with it aren't decamping to the GIMP or some other laughable knockoff; instead, they are buying CS6 licenses, and they aren't going to be convinced to replace those with Creative Cloud subscriptions by a mere 20% improvement in Photoshop.)

Meanwhile, there are a whole lot of people earning livings as part of Adobe Inc., who, in your "rogue programmer disrupts everything" scenario, suddenly won't be.

This, for good and necessary reasons, makes every one of those people a political enemy of your rogue programmer; after all, whether he means to do so or not, the effect of his actions is to critically imperil those people's ability to maintain themselves, their families, and the position in life to which all of same have grown accustomed.

(This might seem to you like a mild inconvenience. If so, I can only assume either that you're not getting paid well, or that you're only so recently getting paid well that you've yet to accustom yourself to it, and in either case that you're also not carrying a meaningful amount of debt. Hell, I'm not paid all that well, and I carry less debt than anyone I know, and I can assure you that my recent layoff was nonetheless far more than a "mild inconvenience".)

Getting back to our illustration: Of course, I'm sure our lovable rogue doesn't intend to get all those people laid off; indeed, given his laserlike focus on the quality of the product for which he considers himself responsible and his evident and total innocence of matters political, it probably never even occurred to him that such a thing could happen. This doesn't matter at all. Whether he intended it or not, he's put a huge chunk of the corporation which employs him in danger of being laid off.

Expecting those people to react in any other way than by attempting to squash him, by whatever means prove necessary and lie within their grasp, would be foolish; when you threaten a man's life, or something which a man regards as vital, i.e., essential to life, it is only basic sense to expect that he will react as strongly as is within his power. Speculate all you like on why this should be the case; regardless, it is a fact of human existence, and to regard it in any other frame of mind save acceptance is to waste your time and mental effort on utopian fantasy.

And half of a company the size of Adobe is a group of people with a very great deal of power, indeed. Even if we scale down from Adobe as a whole to just the Photoshop project team, at least some of those people are perforce going to be in the higher ranks of management. In fact, since our rogue is a Photoshop developer, at least some of those people are going to be in his direct line of command.

No points for guessing what happens to our lovable rogue when, through his innocence of the political realities of any large human organization, he places at risk the jobs of one or more of his direct superiors. Even if, by some impossible miracle, he manages not to lose his own job, then he's certainly not long for the company, because among the people he's managed to alienate are many of his peers. This means he's got more people looking to put knives in his back than he could keep up with, even if he realized he needed to be looking over his shoulder.

Of course, our lovable rogue being not entirely without some sort of rudimentary sense of interpersonal relationships, even if he isn't shuffled aside immediately and laid off as soon as practicable, he'll realize that his only sensible option is to resign, because there is no longer any chance that he'll be able to work productively with his erstwhile colleagues, who now consider him about as safe to be around as old dynamite that's sweating nitroglycerin.

So our rogue will quit his job, lick his wounds, and find somewhere else to work, and odds are the lesson he'll take from his error is, from the perspective of his future prospects and his ability to advance his career, the single worst one he could possibly choose: specifically, that office politics are incomprehensible, deadly, and evil, that anyone who engages in them is therefore necessarily an awful person and best avoided, as are office politics in general, because an honest hacker can't understand the game well enough to play it, and it'll only end in tears.

None of which, of course, is true. But all of it accords perfectly with the default attitude among software engineers toward office politics, and I'm starting to think that might have a lot to do with the way that the default position of software engineers, when it comes to dealing with management, is "screwed".