How do you personally measure "a good day's work?" Is it based on gut feelings or do you have specific hours that you work? Also, do you follow a strict schedule, or leave your day open for "sporadic" opportunities?
As to your first question I don't have a good answer. I cannot think of a way to quantify something that I can only intuit about. You can think of any metric by which to measure the progress of a programmer and you will find an endless debates about its usefulness. It's inherently a qualitative value in my experience.
In general though I am motivated by accomplishment. I feel good about doing a good job. It is its own reward. I feel that I've had a good day's work when I've contributed something worthwhile that has improved the code, the company or my team in some way, even if it's small.
I keep my work days structured around the traditional 9-5. At first it was out of habit but I've since embraced it as a discipline. The difference for me, I believe, is that without a long commute and being stuck in an office far from home there is very little cost associated with following "sporadic" opportunities. I don't leave time out for them because the benefits tend to out-weight the cost of being away from my desk. I can get up and take a walk down to my local café, chat with the barista about local events and pick up some flowers for my wife on the way home. This relieves stress, breaks up monotony and gives me time and space to connect the hunches together in my mind. I may come back from that outing with the solution to my problem or I may not; but at the very least I will be reinvigorated to take on the next set of tasks ahead of me and it cost the team almost nothing and gained quite a bit.
The way I see it my employer was looking for someone who was smart, experienced, creative and responsible. I do everything I can to cultivate that in myself. They're not paying me to trade hours for lines of code. They need someone with my experience, knowledge and intuition to help them solve problems and build a solid product.
If the solution to our problems as programmers could be solved by rote then I'd be out of work and in a different profession.
This applies to traditional work too, doesn't it? Sure, a brick layer can measure based upon how many bricks but I don't think it makes sense for programmers to measure a "good day's work" in terms of just time spent coding ... I think it is more about progress. Even if you didn't write any code, did you make some conceptual breakthroughs that will lead to better coding tomorrow?
For me, it's not based on the work, it's based on the mood/mind. A "good day" is one where I was thinking clearly and solving problems and choosing good paths with ease.
In fact, the amount of code written can often be inversely proportional to the quality of your work in that day.
There simply is no meaningful correlation. It usually takes the talent of another skilled professional to measure the quality of your work, which is one of many reasons why management exists.
