Let’s discuss attrition. Attrition gets the most negative attention at workplaces. Folks talk in hushed tones about who is leaving, who is going where, the types of offers they are getting, and what could be going wrong with your team or company. For those leaving, there is the usual drama of long emails and LinkedIn posts about all the good things they are leaving behind. No doubt. It’s natural to feel emotional when leaving a team with whom you spent your most waking hours.

With respect to @martin_casado’s a16z.com/2021/05/27/cos…, I must share some perspectives:

1. There is no tech that can help repatriate. The Kubernetes reference is a lie we tell ourselves. We keep pouring vast amounts of intellectual and real money at this, and other such lies. 2. Companies that operate successfully (high agility and manageable costs) on-prem must spend upwards of 3-5 years of some of their best engineering talent to mature their core infrastructure (start with the primitives) architecture and engineering.

Very few can afford this.

Having worked in several matrixed situations for solving cross-cutting problems, I couldn't agree more with this article by @jeremiahlee on the so-called Spotify model.

"Failed #SquadGoals" - jeremiahlee.com/posts/failed-s… (1) Matrixed work requires certain assumptions to be true

(a) Everyone can thrive in ambiguity
(b) You can naturally and organically determine who should make a decision
(c) You can also naturally and organically know who is accountable for delivery

These are rarely true.