I have a car, an iPod, and a long commute. I mostly listen to music while driving, and my car (a 2006 Volkswagen GLI) has a useful but quirky way of interfacing with an iPod: the five CD buttons on the radio console are mapped to the first five playlists on the iPod, in alphabetical order, and the sixth button is dedicated to whatever playlist the iPod was currently playing.

The key to making use of this is to use playlists that start with a number 1 through 5, so each button corresponds to a playlist I want to hear. The playlist I use the most, however, is “4 - Haven’t Heard Recently.” It’s a smart playlist, which has the rules:

Match all of the following rules:
Last Played is not in the last 4 weeks
Media Kind is not Podcast
Kind does not contain video
Kind does not contain movie
Last Skipped is not in the last 1 months
Album does not contain Seminars About Long
Time is less than 33:00
Kind does not contain PDF
Genre is not Comedy
Limit to 100 items selected by random

This playlist always contains songs that I haven’t heard in a while, so my music playlist while driving feels fresh (even though there are albums I will play over and over again, but rarely so). I limit the playlist size to 100 items, so when I deplete them, I have to sync my iPod again and “refill” that playlist with more random songs I haven’t listened to recently (either on the iPod, or through iTunes).

One curiosity, though, is that sometimes a song appears again in the playlist, at the very top, even though I definitely listened to it. Why would that happen? It looks like the last-played attribute of those songs wasn’t being updated.

The issue was that duplicate items aren’t updated properly. If a song appears more than once in your library, for some reason the last-played rules from the iPod won’t work because only one of the two items is updated. Find the duplicate item, delete it, and things are back to normal (and you are using less disk space).

See, this side effect acts like a filter on your library. As you listen to music, entries that appear multiple times are caught in this net, and you can deal with them as you find them. Yes, you could write an AppleScript that did the same thing, but that’s not the point: you didn’t start out to filter out duplicates, you started out just listening to music, but filtered out duplicates as a consequence. There must be other examples of this, with greater consequence than compressed music files.

(And, this process also filters out music that I hate, but those songs aren’t quite as well tracked as duplicates)

WOD has been on the app store for two full weeks now. It went up on October 8, after being submitted on September 30.

In the first week, I got lots of great feedback about the app, in email and through ratings on the App Store. The feedback was so good that I submitted version 1.1, which has some bug fixes and additional features, this past Monday (still waiting for approval).

To my surprise, it’s been selling reasonably well: right now, it’s sold 287 copies, most in the United States, with some in Europe, Great Britain, Australia, and Canada. This means I’ve personally made about $400 (+/- for exchange rates…). Not bad. It’s beer money. I’m mostly just surprised every morning when I see that it sold a consistent number the previous day. No big jumps, no drop-offs so far.

As far as I know, this has come purely from browsing the App Store or by word-of-mouth, because I don’t see a lot of traffic on my own web server and haven’t found much on the web linking to it (CrossFit West Santa Cruz let me put up a link on their blog, and another CrossFit affiliate picked up the link, but little else). I don’t know how big it will get, or when I’ll saturate the market for people who would want to buy it.