The Book MP3: The Meaning of a Format by Jonathan Sterne offers an interesting and in-depth look at the various factors which had to come together over the course of a century for the file format which holds most of our music files today, the mp3, to come into existence, and subsequently prominence.  As dense as the book is–if there is indeed a less scientifically dense book on the subject, as Prof. O’Malley alluded to in class, why didn’t we read that instead?–learning about all these factors that at first glance would seem only tangentially related to the topic, such as telephony, turned out to have monumental impact on how most of our music sounds to most of us to this day.  Ultimately, I’m not sure what to do with all this information that I now have on the subject though: am I really supposed to consider the fate of the poor lobotomized cats which made my iPod possible when I’m using it to listen to the Beatles?  Sterne freely admits that people who are worried about the fidelity of the music on their mp3 players would be better served buying better headphones and/or speakers, as that will make a greater positive impact on how their music sounds to them than converting their files to a less-compressed format, and for far less effort. “Pre-echo” on percussive passages of music isn’t a thing I ever heard before this book told me to listen for it, and even now I have to be trying to find it in order to hear it.

However, where MP3 hits closer for me is in its description of the testing used to determine the compression algorithm for the format.  It seems that the groups who were administering the testing didn’t have enough of a range of music in mind.  While it’s certainly true that a limit of listening tests is that you cannot annoy the listener, it seems to me that you could have found some people who had the requisite knowledge to understand what they were listening for and wouldn’t be bothered by modern music, or music which focuses more on rhythm than melody, or polyrhythmic music, or whichever of the types of music that seem to be missing from the selected recordings that you want to go after.  At the very least, it would have been possible to have the expert listeners’ study be paralleled by another by a random sample of the population, which would tend to remove some of that middle-brow, “listenablility” bias, and then use some combination of the results of those two studies to determine which algorithm best reproduced the most music.