←back to thread

210 points dakshgupta | 2 comments | | HN request time: 0s | source
Show context
dakiol ◴[] No.41842172[source]
I once worked for a company that required from each engineer in the team to do what they called “firefighting” during working hours (so not exactly on-call). So for one week, I was triaging bug tickets and trying to resolve them. These bugs belonged to the area my team was part of, so it affected the same product but a vast amount of micro services, most of which I didn’t know much about (besides how to use their APIs). It didn’t make much sense to me. So you have Joe punching code like there’s no tomorrow and introducing bugs because features must go live asap. And then it’s me the one fixing stuff. So unproductive. I always advocated for a slower pace of feature delivery (so more testing and less bugs on production) but everyone was like “are you from the 80s or something? We gotta move fast man!”
replies(5): >>41842203 #>>41842403 #>>41842705 #>>41845436 #>>41846116 #
resonious ◴[] No.41846116[source]
I honestly don't really like the "let's slow down" approach. It's hard for me to buy into the idea that simply slowing down will increase product quality. But I think your comment already contains the key to better quality: close the feedback loop so that engineers are responsible for their own bugs. If I have the option of throwing crap over the wall, I will gravitate towards it. If I have to face all of the consequences of my code, I might behave otherwise.
replies(2): >>41846890 #>>41849775 #
1. isametry ◴[] No.41846890[source]
Slow is smooth, smooth is fast?
replies(1): >>41854226 #
2. samarthr1 ◴[] No.41854226[source]
Greetings, Phil Dunphy