Barry Warsaw, “Get Your Resources Faster With
An introduction to Python 3.7’s new
importlib.resources, which incorporates
some (tiny fraction) of
pkg_resources into the standard library.
Alex Gaynor, “Learning from Failure: Post Mortems”
Failures happen. Make use of them.
VM Brasseur, “The Human Nature of Failure & Resiliency”
A self-described greatest hits of the history of research into failure (biased toward Western cultures). I think this talk is useful in conjunction with the last one, as Brasseur advocates for “pre-mortems” (i.e., answering the question “what went right, and why” compared to a post-mortem’s “what went wrong”) as a tool to identify latent errors.
Amanda Sopkin, “Randomness in Python: Creating Chaos in an Ordered Machine”
Three major take-aways to this talk: don’t use
for things which actually need to be random (counterintuitive, arguably a
failure in naming), use
secrets if you must, but better to use
cryptography if you need cryptographically-secure randomness.
Raymond Hettinger, “Dataclasses: The code generator to end all code generators”
For when you need something a bit better than a (named) tuple, but want to
avoid the normal boilerplate of regular classes. See also:
Nina Zakharenko, “Elegant Solutions for Everyday Python Problems”
Aimed at intermediate developers, a quick overview of magic methods, “method magic,” generators, context managers, decorators, context decorators, and named tuples.
There are a few additional talks that I liked enough to devote full posts to; look for those soon!
If your work won’t let you spend even 10-20% of your time keeping up to date with your profession, quit your job and find a better one!↩