pragmatic Test Driven Development

As a dev, I write mostly unit tests and care about test coverage increasing over time.  My view has shifted recently.

Although web UI tests can be fragile, they present a lot of bang for the buck, coverage-wise.  It turns out they also have a lot of unique business value; that is, if we're testing what our users see, then when we change something we can immediately know that business value is being affected.

- I recommend view-level tests. Django directly supports and advocates this style. They're fast and easy to create like Unit tests, but have direct user-level (thus business-level) value.

- for UI tests, set an ID on everything that you are testing. (Never ever use XPath.)  This makes UI tests more durable and reliable. 

- Kent Beck has been the main proponent of TDD from the very beginning, as part of Extreme Programming. “I get paid for code that works, not for tests, so my philosophy is to test as little as possible to reach a given level of confidence”

- I really enjoy Brian Okken's entirely pragmatic view on testing:
My reaction to “Is TDD Dead?”  His podcast is in my Top 3: Python Testing

great developer podcasts: Accidental Creative

I commute in Los Angeles. Since I started listening to podcasts on my drive, the long rides have become a lot easier, and I am inspired and smarter and more knowledgable. I highly recommend using podcasts as a tool to expand your horizons.

Here are my favorite podcasts.

Oddly for a senior DevOps guy, my #1 is not technical at all. One podcast above all others serves to inspire me and help me move forward in my career: Accidental Creative by Todd Henry. His aim is to assist the working creative professional be productive and organized. His audience is the corporate "create on demand" class: designers, marketers, writers.  As a DevOps guy I find his work refreshing, it helps me open up my brain, to relax and focus my whole being on solving mysterious code or server problems.

Each short (ten minute) podcast has a clear idea. He then briefly and directly supports that idea, often by interviewing someone.  Recently there was an episode on doing "deep work."  By dedicating a few hours at a time, we can have great ideas that we can flesh out during the normal work week. That is, in contrast to the traditional work style of trying to make progress by herding chickens, one creates a vision then works towards it.

This is similar to John Cleese's theory: creativity is a shy turtle. Your creativity is shy, so if you create a reliable "safe haven" for it, it might poke its head out and share something awesome with you. This sounds exactly like the "deep work" idea of setting aside reliable quiet time to work on larger ideas.