@cford@toot.thoughtworks.com @roundcrisis@types.pl I'll have a go at a brief reply while we're working on the write ups! Firstly this is a question without an answer in Residuality - the goal is neither to act on stressors nor to keep them in the back of your mind, so the only answer is you don't distinguish either way. The goal is to handle _unanticipated_ stressors. Since we can't anticipate the specific stressors that will occur, to try and give ourselves more capacity to handle the unexpected, we look for _attractors_ - common patterns of business reaction and system design that handle _multiple_ stressors from the roughly half of the large list of random stressors you did anticipate. You then validate that those attractors help with unanticipated stressors by testing the new design against the other half of your stressor list. Even though you did anticipate these ones, they weren't used to come up with the system design, so they are a proxy for the probability distribution of future stressors and business reactions. Now you have a design and a set of architectural changes linked to the sets of stressors that prompted their inclusion, you still have to choose which to actually implement a point in time - and residuality is silent on this - you do so using standard roi/prioritisation techniques given the current business context. But not with the goal of addressing specific stressors from your initial list. What residuality has given you is a set of architectural elements that you can draw from when responding to the needs of the current day, confident that they will improve your ability to accommodate unanticipated stress in the future. OK... not so brief! But I'll hit send anyway and happy to clarify or answer further questions if this prompts any other thoughts :)