When we are too tired to make good decisions, it can be a good decision to decide to stop making decisions for a while. --RonJeffries on the XpMailingList
This is why even a partner is not enough to help follow this rule effectively, and we fall back on rigid numbers, like FortyHourWeek, as a reminder.
A better way of stating the rule is WorkWhenYoureFreshStopWhenYoureTired.
Here's how I know I'm adhering to the spirit of the rule -- I don't look back often and say, "I knew I was being ineffective, but I was too far into the problem to realize it." --KentBeck on XpMailingList
I find that there are different kinds of work that I do better at various times of the day. In the morning I am best at analysis. In the late morning and early afternoon I'm best at the details of programming. In the late afternoon I'm best at RandomActsOfLearning.
I think it depends on the person. Personally I like to enjoy my work and not be forced to do the work. Usually I like to FrontLoad my time so I can rest up a bit easier later on in theory. I sporatically get RandomActsOfLearning done duing my FrontLoad period and have to deal with less people at once to boot. I still adhere to WorkWhenYoureFreshStopWhenYoureTired, I am still young for now and I can have longer runtimes than FortyHourWeek, but having RandomActsOfLearning in the middle helps relieve tiredness and makes you go on longer.-- ArchimedesTrajano