4 ms·

I forget what job it was I was interviewing for but I remember being stunned at the open lack of process they had in place for trying to bother unifying all the

I forget what job it was I was interviewing for but I remember being stunned at the open lack of process they had in place for trying to bother unifying all these teams to be on the same platform framework etc and instead deliberately let sub teams work with tooling that worked for them and didn't create a lot of overhead trying to coordinate and sync and share a common codebase. Every place prior to that day I had only heard of people going for long stretches to impose a company wide uniformity in tooling. All the costs associated with getting everyone on the same page was always taken as a worthy cost to pay and that future date in the promise land would have made it all worth it. The folks in this other company just flat out said not worth it, just get your stuff done, ideal development scenarios and perfect harmony and alignment have too much hidden costs to bother with. It's naturally a pro and con scenario but it was pretty refreshing to hear of folks so unconstrained by this self imposed constraint every other shop had imposed on themselves.

hn_user2 4y ago
I have seen two extremes of this. The one side is forced conformity, and the other side was complete chaos. To the point where each project was written in its own language. Imagine changing teams to help on a project and step 1 is to learn a new language and deploy methods. I feel like there is a golden middle where the team has decided on maybe 3 languages, one typed, one scripted, one cutting edge. And a few different deploy scenarios, and you can pick amongst them.
bennyelv 4y ago
I work at a company that's trying to come out of the other side of this kind of approach. Let me tell you, it might feel good while you're doing it but in 5 years time it will be a hot mess of unmaintainable garbage! There are hidden costs in trying to align as you go, but there are potentially company ending costs in not. You're right that there is no promised land, but the quest to reach it prevents you regressing into hell.

hn_user2 4y ago

I have seen two extremes of this. The one side is forced conformity, and the other side was complete chaos. To the point where each project was written in its own language. Imagine changing teams to help on a project and step 1 is to learn a new language and deploy methods. I feel like there is a golden middle where the team has decided on maybe 3 languages, one typed, one scripted, one cutting edge. And a few different deploy scenarios, and you can pick amongst them.

bennyelv 4y ago

I work at a company that's trying to come out of the other side of this kind of approach. Let me tell you, it might feel good while you're doing it but in 5 years time it will be a hot mess of unmaintainable garbage! There are hidden costs in trying to align as you go, but there are potentially company ending costs in not. You're right that there is no promised land, but the quest to reach it prevents you regressing into hell.