I work 6hrs a day usually and spend the rest of time experimenting with my own DSLs for future side projects or enjoying life whereas before I'd go to work before the sun came up, work in a windowless office under constant unachievable deadlines then leave at night. I had money but no life.
And I could ask you the same question: the startup founders and oil industry would live comfortably enough if they made a little less and paid him better; why is that not enough? Why should they be making more?
I also believe the guys who started ITA software wanted a problem domain where they could justify using Lisp.
So there are a few precedents for finding an excuse to use more expressive languages to solve challenging problems, in a way that makes money.
Perhaps there's something to that in programming languages as well; I sometimes wonder if the reason for the explosion in startups and web applications is that for as much as I hate fighting with CSS and HTML templates, at least it's better than trying to make sense of the Windows API. And if you've grown up, as many young startup coders do, with the world of JavaScript and the web, it's what you know as much as Lisp was what guys like Paul Graham knew.
I intuitively knew as a child that there were parts of programming and using computers which "weren't work" for me, but also very much ones that were, and I foresaw the probability that I might cease to enjoy the "not work" parts after being ground down by too much work.
As I come back to it now, I still think that. Increasingly I think it really, genuinely, depends on the style and language of programming you take up, and what you do with it. When I code in something like Lisp or Haskell, it doesn't feel like "work," it feels like problem solving. I tend to focus on CS topics as well; I built a VM, and a couple of programming languages, because they too were fun to think about, not work.
For me, I have progressively gravitated to more and more expressive languages, while moving away from more imperative ones which descend into repetitive "design patterns". I burned out quickly on the world of web programming, which inherently seems to require a whole lot of tedious yak shaving if you're flying solo.
Of course, the flipside which anyone bothering to read this may already be hastening to chastise me with is that none of this is likely to necessarily lead to actual employment.
Someone has to do the "work." And the amount of "work" to be done is a lot greater than there are people who don't find it work, which is why there will always be a lot more demand for that. In part I suspect this is because some kinds of "work" are self-replicating (Java programming begets the need for more Java programmers ...), but I nonetheless find myself fearing that my chosen "not work" is merely a fool's fantasy; and I occasionally have guilt complexes about "why am I not just learning JavaScript so I can at least get a gruntwork job at Rovio or somewhere" ...