Even when I am actually doing work, I often spend at least half of my time reading and investigating existing code, not coding. Why? Because I want to:
1) Really understand what is going on in the existing code base
2) Make sure what I plan to do is in line with the philosophy of the current code
3) Make sure I'm not replicating something already implemented
4) Give my brain enough gestation time to allow lateral thinking to happen
In fact, often I'll spend some time reading, go off and practice juggling, then come back and whip off an initial prototype, then go and do something else distracting (read Hacker News) then revisit the problem again the next day. Most of the time, I'll figure out an even better way to do it.
1) Really understand what is going on in the existing code base
2) Make sure what I plan to do is in line with the philosophy of the current code
3) Make sure I'm not replicating something already implemented
4) Give my brain enough gestation time to allow lateral thinking to happen
In fact, often I'll spend some time reading, go off and practice juggling, then come back and whip off an initial prototype, then go and do something else distracting (read Hacker News) then revisit the problem again the next day. Most of the time, I'll figure out an even better way to do it.