I go to the coffeeshop with paper and pens only. Then I try to kill the ideas around what I'm working on right then and there, using a UX-centric focus. I've re

I go to the coffeeshop with paper and pens only. Then I try to kill the ideas around what I'm working on right then and there, using a UX-centric focus. I've realized that my personality centers this paper design process, and works best if I try not to consider too many implementation decisions at this stage. The result should be something approximating a functional spec, with features and their rationale, asset lists, wireframes, etc. - a document that describes something that "should" already exist. The coffeeshop time is when I am most allowed to space out, enjoy the moment, doodle some things. What I'm really doing is batching together all the biggest emotional decisions of the day so that my remaining work is smoothed out - it's all execution without having to stop and decide. Upon my return I turn the indicated design into Trello cards. When I want to do labor I look at the cards and pick one. Speed and terseness is essential for my retaining focus - having the paper design prevents me from compromising quality as I try to take shortcuts. If I get too much into storytelling while sitting at the computer, I may romanticize what I'm doing instead of doing it. If I have to send emails for business I try to keep a short timetable of a few minutes each; likewise with code. When I get tempted to narrate future work in my code, I deliberately make the comment barebones - a list of titles, for example, just enough to be a reminder. The cost of filling in the gaps in the comment could go to finishing the feature. If I need an additional level of documentation, that signals that I should go back to the paper design and the cards. This is only most possible with solo work and a high level of familiarity with everything I'm doing. If I have to learn, then I have to go through a different process of playing and practicing before I can proceed so rigidly. And with team projects things are necessarily different.