If we don't explore the ends of the spectrum deliberately
we’re not going to know if our thinking is in fact a balancing act in the middle of the spectrum or actually at one end of it (without us even knowing it).
»The problem with optimizing the parts of the system that are not the ones that should be optimized
(non-constraints) is that it makes things even worse in the part of the system that is the one that should be optimized (constraint) and consequentially the whole system.
»Batch size and responsiveness
Besides other benefits, reducing batch size increases the responsiveness of the system.
»It's impossible to not have high WIP and longer lead times in async work
In async work, if you stop working on your thing to respond to another thing, you get interruption, longer lead time for your thing, and higher WIP. If you mute notifications you get longer wait times and high WIP for the other thing.
»Instead of complaining about big PRs to review
you should do reviews way faster and you’ll get way smaller PRs to review.
»XP teams are very rare environments where WIP is not skyrocketing
Funnily, concept of WIP limit is invisible and unknown to most of XP teams. They do it without even knowing or trying to do it.
»Teams that are afraid to change the code
often tend to also experience lack of trust from stakeholders.
»Psychological safety and trust go down the drain
in teams that are not able to sustainably deliver high-quality software. And XP and Continuous Delivery are preconditions for that.
»Doing Continuous Integration is a suboptimal intervention point in the system
Trying to do more of Continuous Integration by trying to continuously integrate is a suboptimal intervention point in the system (can be a monitoring point, though).
»While we realize the need
for minimizing (lead) time to recover from incidents and the way to achieve it, somehow we fail to realize either the need for minimizing the lead time to value for our customers or the way to achieve it.
»