Stumbled upon this (or, rather, was shown it). Read that before you continue. Here's my take on it:
The difference between programming and, say, sports, is that programming tends to be an all or nothing plan. You can't build half an algorithm and get half the satisfaction out of it; half an algorithm is, essentially, a broken algorithm. Zero points. And when you're dealing with programming, nine out of ten times, your algorithm will be broken.
The positive feedback only comes upon finishing the program, and then it has to be absolutely perfect.
Let me explain from my point of view. If you've followed this blog at all, you'll know I do a ton of time-crunching game jams. Imagine this: It's the last hour in what was just a 72 hour jam, where you sacrificed life and limb to finish your game. You finally submit it within the last few minutes, breathe a sigh of relief, and go test it.
There's a game-breaking bug after the first level.
All of the feedback you get is generally positive, because the indie game community is generally awesome and supportive, but all you can think of is if I had not typed that one variable wrong I would've had a functional game.
The problem is even worse in the work force, because due to the flexibility of digital media/software in general, clients work closely with their programmers to make it worth the project cost. Hence, every flaw is pointed out. If the program is not 100% perfect then it's not worth their money.
Now, the absolute worst is when you're working for a large company.
On really large projects, administration tends to partition individual pieces of the issue to different programmers, so that your own job is fairly manageable. However, it also means that your own job is fairly insignificant. Imagine being the guy that optimizes shaders in Halo. There really is no positive feedback, at all. After slaving on pushing assembly code around for months, once the game's released, you can play it and say, "You see how the light refracts off of mossy rocks? I made that twice as fast."
Not very satisfying.
Now, here's where I disagree with the original link. I mentioned that nine times out of ten, your efforts result in absolute failure. But that other ten percent is the greatest feeling in the world, when you've slaved a week, or two weeks, or even a few hours on a program, and it comes to life and it's brilliant, it's all really worth it.