4 ms·
People (and I include myself) are lazy and will present that laziness via technical arguments and pretending not to understand/remember what you are saying. It'
by arielweisberg 10y ago
People (and I include myself) are lazy and will present that laziness via technical arguments and pretending not to understand/remember what you are saying. It's the steady string of previously debunked straw-man arguments or orthogonal unrelated objections that is the tell tale signal. If you keep pushing you are just burning relationship currency.
If you dig in for every single little thing your going to be the bad guy org wide because people don't have enough context in most development workflows and aren't going to put in the effort to understand the technical or timeline issues at stake. You will just be the one who makes a fuss all the time. I've watch some solid people get pushed out due to this.
As long as the harm is minimal I'll point it out and let it slide if the objections go off the rails even if the cost to do it right/better is less than the value realized from doing it right/better. Discussion isn't free. When the cost of dragging the discussion out is commensurate to the harm or value realized then that is when I will tax both myself and others involved in the discussion to address it.
This is one reason to have these discussions before code is written and people are invested in not having to make changes. If you are brutal and break things down into small chunks delivered regularly the latitude for things to go sideways is reduced and discussion is more productive.
- jacquesm 10y ago> Discussion isn't free. Exactly, that's the key insight to have. Making the discussions more important than the project is a pretty good way to derail anything, there has to be some sense of proportionality.