Curated Content October 2021
A few pieces of content I thought were worthwhile in the month of October.
95%-ile isn't that good
A great article from @danluu about how performing at a higher level than most people that are interested in the same things that you are, whether software development, sales, or Overwatch isn't all that hard. If you avoid playing badly more often than not you'll outperform the vast majority of people, including those that are trying to play competitively. It's only at the highest levels that it's no longer about avoiding playing badly, but instead playing well.
The Oxymoron of "Data-Driven Innovation"
The first of a pair of great articles this month from @HeyChelseaTroy. The first part about the failured induced by using data to drive product decision making in organizations is interesting, but the second part of how the innovation really happens when you serve the folks at the margins. I don't have data to judge how well this paradigm fits, but Chelsea makes a compelling case, and I'd love to either see or carry out the research on this topic.
A Rubric for Evaluating Team Members' Contributions to a Maintainable Code Base
I'm still mulling over the actual rubric, but this has been something I've been thinking about quite a bit lately, as it's attempting to answer the question of how do you know that folks on your team preempted fires if there aren't any visible fires? Worth a read, and if you don't adopt this rubric, consider what this would look like in your organization.
Working Backwards - Colin Bryar and Bill Carr
Regardless of what you think of Amazon, Bezos, etc, the way that Colin Bryar and Bill Carr packaged up the ways you could setup to operate a less dysfunctional organization gave me some intersting food for thought.
The idea of a Bar Raiser as a way to ensure hiring standards are kept by pulling someone into the interview process who doesn't feel the pain of needing someone to join the team immediately, and the 6 Pager rather than the PowerPoint as a way to think deeply about a problem were very interesting to me and ideas I'm attempting to incorporate into the way I run organizations.
Principles of Product Development - Donald Reinertsen
The way Reinertsen provides insight into the product process, and how it can be informed by learnings outside of lean manufacturing. Things like how queue theory, informed by networking, can help improve our product decisions was incredibly interesting and ought to be a foundational text for folks attempting to operate an agile or lean product development environment, especially software.
Specification by Example - Gojko Adzic
While there wasn't a ton of new information for me personally in this this one, the packaging, and my personal distance from the Ruby community where Behavior Driven Development (BDD) is practiced in a more central way was refreshing.
While not exactly a book about BDD, it is a book about making sure you don't screw up the communication with the product team/customers/stakeholders and actually deliver what users want. If you're not familiar with the concepts, or if you feel like your team could use a guide to ensuring you're delivering what users need, while minimizing miscoummunication this book is a great place to start.
Designing Data Intensive Applications - Martin Kleppmann
This is a great guide for anyone looking to level up their software engineering and architecture starting intially from much lower level building blocks, composed upward into complex systems and how they should interact, including the tradeoffs that you make across the chain. A must read for folks in software.
Distributed, Scalable, Fault-Tolerant Video Streaming with Riak Core, Elixir, and Kubernetes
I'm biased, as Jace is a great member of my team, but I thought this highlight of how we've built our video platform at Savi Solutions was excellent, and how Riak Core and Elixir made that a much lighter lift than it would have otherwise been.
No jokes this time around, but instead a great thread on how to quickly allow folks to acquire expertise. Hint: it isn't through classroom learning. Useful to consider in the context of the value of pairing and mobbing/ensemble programming.