polishing
We’re polishing up three big releases for Big Cartel at the moment. Polishing can be a slog because, in a sense, the hard bit is out of the way—you’ve figured out how to build whatever it was you were trying to build. It feels like it’s time to move on to the next challenge, except that the user experience isn’t where it needs to be yet. And, in the end, the experience is what matters (no matter the technical hurdles you’ve already overcome). In any case, polish and QA are the order of the week.
Similarly with button-down bird, I’m submitting test files to the printer today. I’m a little apprehensive about getting the proofs back from the printer. I’ve never really designed for print, and I know things are going to look entirely different on paper than they do on my screen. Hopefully, we won’t have too much trouble matching color, though.
I spent last evening messing with ePub. What a horrible format. There is little to no documentation and lots of XML. I ended up taking apart the few non-DRM files I had to see how they worked (which I was only able to do once I figured out they were just zip files). ePub kind of reminds me of when the web was young. I guess we’re just going to need a Zeldman to come in and advocate for proper standards.
Using standard ePub for a picture book looks terrible—the format is intended for text-heavy content. Luckily, I found just enough information on Apple’s extensions to the format to create a fixed layout ePub which looks not unlike the print edition of our book. If I can find good documentation (or a non-DRM example), I’d like to eventually include a narration track as well. I am not even going to try to create a version for Nook or Kindle at this point—I don’t think it would do the artwork justice. It will be interesting to see whether we sell any books via iTunes, though.
Last weekend was all Transitflow. Friday evening and Saturday I read up on everything from RabbitMQ and AMQP to host configuration management to Chicago’s transit data APIs. There are a lot of options out there, but luckily I had a good idea of what I was looking for. Sunday, I put all of what I’d been researching down on paper, then Tinderbox, then OmniGraffle. We now have a pretty solid architecture for the backend of Transitflow and a starting database schema. Next, I get to start building it.