Recent Posts

  1. ()

    This post is part of a series reimplementing one of my past projects using technology available these days; for a general overview, please see the prologue. Continuing on from last time, the design of the system as a whole is, intentionally, pretty boring. I think boringness is underrated by developers as a whole; I actually want my projects — especially my work projects! — to be as boring as possible because it buys you predictability. … (read more)
  2. ()

    Fridays are for learning.1 Another batch of videos from PyCon US 2018 (see the YouTube channel for more), for the week ending May 25, 2018. Barry Warsaw, “Get Your Resources Faster With importlib.resources Watch this video at YouTube. 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” Watch this video at YouTube. Failures happen. … (read more)
  3. ()

    Fridays are for learning.1 These are the first batch of videos from PyCon US 2018 (see the YouTube channel for more) for the week ending May 18, 2018. Larry Hastings, “Solve Your Problem With Sloppy Python” Watch this video at YouTube. I think Larry has a good point (things you build for yourself to automate personal tasks don’t need to be held to the same standard as code you write for others to consume); I like to use these opportunities to try out new tools. … (read more)
  4. ()

    Fridays are for learning.1 These are some interesting links for the week ending May 11, 2018. A short list this week; I’d encourage you to keep an eye on PyVideo for recordings of talks done this weekend at PyCon US 2018. (Presumably this page will start existing at some point.) Next week I plan to start rounding up the talks I found most interesting. On structured concurrency in high-level languages, continuing our series of posts thinking about this topic. … (read more)
  5. ()

    Fridays are for learning.1 These are some interesting links for the week ending May 4, 2018. Lots of prominent new releases this week! Flask 1.0!2 Rancher 2.0! Fedora 28! NetHack 3.6.1! Congrats to all. Jupyter received the ACM Software System Award, joining such highlights of computing history as NCSA Mosaic, TeX,3 make, and more. Heady company for sure, but well-deserved. Interpretable machine learning is going to become more and more important; in my opinion black-box “computer says no” models will not be tenable much longer. … (read more)
  6. ()

    Recently I’ve been thinking about how much easier it is to achieve quite complex things these days compared to back in The Olden Days. In fact, it almost feels like we suffer from a surfeit of mature, high quality pieces that we can use to build almost anything.1 There’s one project I worked on in the past — many years ago now — that I still think about from time to time; how would I accomplish those same tasks now, with everything that’s easily available to a professional today? … (read more)
  7. ()

    Fridays are for learning.1 These are some interesting links for the week ending April 27, 2018. Warehouse is finally, officially, the new PyPI. Well done to everyone who helped make this happen! On making a blog with more privacy. An overview of one of my favorite new(er) Python standard modules, pathlib. Python 3.7 implements PEP 553, a better way of inserting breakpoints in your code. Maybe we can figure out when IoT devices are compromised using machine learning? … (read more)
  8. ()

    Hmm, a new domain, a new theme… looks like this site’s finally got a piece of that pie.1 2 Watch this video at YouTube. And by “that pie” I mean “my time”; cobbler’s kids and all, you know. ↩ Okay, let’s be real, though: if you are of an age with me, this is what you actually thought of as soon as you read the phrase “moving on up”: … (read more)
  9. ()

    Jake VanderPlas’s keynote for PyCon US 2017 was about “The Unexpected Effectiveness of Python in Science”: Watch this video at YouTube. The main thrust of Jake’s talk was about how astronomers have been able to use Python and the culture of the community around the language to do better science, including being able to provide other scientists access to the software used to perform experiments and analyze and manage data. … (read more)
  10. ()

    It’s that time of year again... time to catch up on this year’s PyCon US talks. I’m planning to post and comment on some of the ones that made me think the most. To start, I really enjoyed Lennart Regebro’s “Prehistoric Patterns in Python”. Watch this video at YouTube. Even though I think most of these old patterns have largely gone away, sometimes the “Old Way” of doing things can resurface. … (read more)