A year ago, every time I got in the car to go to work, I'd grab one of eight CD cases to provide my music selection for the commute and the workday. Today, those CD cases are collecting dust in the closet, while I carry their collected contents in a 160 gig iPod everywhere I go. I rarely use it in my office, since Grooveshark and Last.fm offer a vastly wider array of music. I know many others made this transition years ago, but I just got around to it, and I think I'm part of the masses in this transition, as opposed to the early adopters.
In any case, I've been thinking about the long-term effects of end of physical media. While reading music blogs, I've encountered a lot of discussion about this subject, but almost all of it is focused on the impact it has on the music business; surprisingly little is said about the artistic impact.
I approach music first as a listener. Deciding how to spend my music budget used to be a big deal - I carefully considered what I would most like to listen to, and I often guessed wrong. Now, the majority of music costs me nothing to hear - money is no longer a limiting factor. In fact, given the vast array of readily available music and my relatively broad tastes, there's no way I will ever get around to listening to everything I might like. I can listen to practically any artist on a whim, and I can choose to filter out all but that artists' best songs (measured imperfectly by popularity and various tagging schemes, but the technology is bound to improve, and already works tolerably well).
What I think this means to artists is that filler is dead. Let me back up, and say that the notion of "filler" is only meaningful from a listener's perspective. Generally, I don't believe artists intentionally release bad music. Most are aware that certain of their songs are better than others, but everyone tries to pick their best to publish. When I started working on my first album, the album format was still very dominant, and based on all of my past experiences as a music fan, I felt a sense of responsibility to give anyone who bought my album their "money's worth." I figured most listeners were like me, owning a limited number of albums, and listening to each multiple times. In that context, adding a decent (but not great) song to an album made it better, by extending the variety and collective play length of its owner's finite music collection. Now that everyone's library is effectively infinite, that kind of song is filler. In other words, when evaluating which of their songs to release, artists should set a higher bar for "good enough."
By extension, the concept of the album is obsolete. A lot of business-oriented articles address this, but from the angle of how to market singles versus albums. I think it's important in that it removes the need for a given number of songs to sound like parts of a whole. If, for artistic reasons, you want to release a collection of 12 songs together, you still can. But, you can just as easily release one song at a time, or three, or 85. I'm planning on taking the singles route, partly out of the desire to be more free with stylistic changes, but moreso because I have found the necessity of creating an album at a time to be paralyzing. I have a job, and other interests - the massive time required to get a set of songs ready at the same time was disruptive to everything else in my life, in a way that working on a song or two at a time is not. I have a backlog of partially finished songs, that were left unfinished at least in part because they didn't fit with the rest of what I had planned for an album. The need to wait for an album's worth of songs to be ready also served as an excuse that I used to give myself permission to finish them - now, there's no barrier.