Day[9]TV

Even though I don't particularly care about Starcraft II, I quite enjoy Sean Plott's enthusiasm for the subject and his amusing antics on this daily video podcast. I'll often have this on in the background, not particularly paying much attention and often skipping entire episodes at a time.

EngineerGuy

Another video podcast, but a much easier one to regularly watch since each episode is only 5-10 minutes. Bill Hammack explains neat engineering details behind a variety of cool technology in an extremely accessible manner. The video format is well suited for the topic, as the visualizations add a lot to the explanations.

Ruby Rogues

This is my programming podcast, and something I use to generally keep tabs on concepts/technologies/etc that I might find interesting. Not every weekly topic is of interest to me, so I'll pay a varying amount of attention to this depending on what the hosts are talking about. One of my favorite parts of this podcast is the "picks" segment at the end of each show, where everyone will chime in with something neat that they recommend.

Three Moves Ahead

A podcast about strategy games and their design (mostly video games, but occassionally venturing into board game territory as well). They often have game designers as guests to talk about their games and design choices.

99% Invisible

This podcast discusses aspects of design in the everyday world (mostly with architecture) that you probably never even notice. It's another one that's easy to keep up to date on, with 5-10 minute episode lengths. I liked this one enough to contribute to the Kickstarter campaign to produce a third season.

A friend of mine was recently talking about doing interval training, and it sounded like a good way to play around with Go.

Because I always forget, this is for my future self:

I think this is the cleanest way of using Python on OS X, short of installing a new version of Python via Homebrew. (For Ruby, I like to use rbenv and ruby-build, since it's handy to be able to use both 1.8 and 1.9 versions for my personal work.)

With Webby being deprecated, I intended on moving this blog over to Jekyll, as it's supported natively by GitHub Pages. This would have allowed me to use a single repository for this blog, instead of having one repo for the source and a second which would hold the "build" that would be hosted by GitHub.

Unfortunately, I quickly bumped up against Jekyll's limitations, mainly due to the Liquid templating language that Jekyll uses. To accomplish what I wanted to do, I would need to add my own plugins to Jekyll/Liquid. This more or less defeats the original choice of Jekyll, as I wouldn't be able to use a single repository. And once that became obvious, I figured I may as well evaluate other static site generators.

Some research on Google quickly revealed two main candidates that appealed to me: Middleman and nanoc. After fiddling around with a test blog and some sample posts in both, I settled on nanoc. Middleman felt more powerful, but I preferred the simplicity of nanoc. (Perhaps because it's more like Webby, which is what I was familiar with.)

As I was upgrading the "backend", I also wanted to pretty up the frontend a bit. Not the smartest decision, as I was now also investigating HTML/CSS frameworks alongside the static site generators. But since I know far less about the former, I chose Foundation, mostly since it allows for a semantic grid using Compass. (Although I knew I didn't want to use Bootstrap since it's so prevalent across the web.)

And finally, here's the end result of all that work! It didn't take me one day, but it was rather satisfying nonetheless. The hardest part was getting everything to work together, and figuring out how nano's rules worked, but as usual, most of the answers were just a Google search away.

One last note: if you do have to use two repositories for hosting a site on GitHub, I highly recommend using a submodule for the build. For this blog, I added the GitHub Pages repo (kejadlen.github.com) as a subdirectory of the main repo (kejadlen.net) so that I could build the site straight into the build repo. (Or you could add a rake task to dump the output dir into the build repo. Either works, but I find this more convenient.)

The source for this blog is available on GitHub here.

I assume that by now, pretty much everyone has heard about the four women's doubles badminton teams being disqualified for throwing their matches. I wanted to resume blogging by writing about switching to nanoc, but couldn't resist chiming in on this topic instead!

Being an ultimate player who strongly believes in the "spirit of the game", I'm disappointed in the players. But as someone who also very much believes in "playing to win", I can't ignore the faults in the tournament format either. I guess that although I understand the situation, I'm still disappointed in pretty much everyone involved.

Sirlin has some excellent commentary in his post on the topic, Playing to Win in Badminton:

Whenever a "spirit of the game" clause is enforced, it's a failure of the rules. I might go so far as to say it *always* shows wrongdoing on the part of the tournament organizers.

If they want pools to have top 2 that continue to a single elimination bracket, then the exact seeding must not be knowable ahead of time. (To avoid cheating from the judges, the exact rules for determining the single elim bracket could be written down ahead of time for that particular event, and held by a trusted 3rd party to verify that they were followed, and not made-up on the spot to favor someone.) That would have solved this particular problem, though even then, round-robin pools have other problems that I mentioned with lame ducks and kingmakers. Double elimination pools would solve those problems because there are *never* lame duck players in a double elimination bracket. Anyone still in the bracket can win. Also it reliably determines top 2, which is what's desired here.

(As an aside, I wonder what he would say about ultimate's "spirit of the game" concept.)