Your post naturally gravitates toward the question, ‘what’s good enough to save?’, and I like that question. Digital objects tend to pile up and are often in desperate need for some curation (and deletion) in addition to preservation. I think that digital archivists avoid the MP3 strictly because it is compressed audio; but what is the value, in reality, of saving an oral history in WAV (or FLAC) vs MP3? One case is simple: we want to save the original. However, many audio recorders output directly to MP3. So, aside from the argument from original source, what extra value exists in the sound of the human voice speaking in uncompressed audio?
One argument against the MP3 is its actively enforced patent. However, the MP3 U.S. patent will expire at the end of 2017, so maybe things will change.
I think oral history is a case where MP3s make a lot of sense– the audio files are often large, particularly if interviews are longer (Most of my interviews are over an hour long). I would expect that little-to-no post-production is done on the recordings, apart from some volume adjustment (e.g., adding EQs/filters/effects), so there’s less “data” there to analyze.
I think there are circumstances where you would want to go less-lossy and even analog if that was possible, particularly when working with musical recordings. This would take a lot more storage, and in turn, cost a lot more money. It’s worth noting that the difference in file formats when it comes to MP3s vs. WAVs also depends on the hardware on which you listen to them. To hear the difference in these files, you need to have amplifiers and speakers/headphones that really can give a good “image” of the frequency spectrum. Why have .WAV files if your researchers are trying to hear the minutiae of a recording on earbuds? An archive could hope that they might be able to provide a professional listening environment in the future, or count on users to provide their own, but the hardware needs to be considered along with the file format.
Jeff – I think you’ve hit the nail on the head when you say that the goal “good enough for most people, most of the time.” So much of library, archive, and info science is coming up with some sort of standardization that we think can be most useful to the most number of people.
But Kirschenbaum’s notion of the “heterogeneity of digital inscription” highlights the fact that we’re not dealing with uniform materials. While many of us info professionals may long for standards, it’s not a desire that inherently maps well onto digital objects. We may have format/file preferences, but we can’t control how materials are created.
Your post naturally gravitates toward the question, ‘what’s good enough to save?’, and I like that question. Digital objects tend to pile up and are often in desperate need for some curation (and deletion) in addition to preservation. I think that digital archivists avoid the MP3 strictly because it is compressed audio; but what is the value, in reality, of saving an oral history in WAV (or FLAC) vs MP3? One case is simple: we want to save the original. However, many audio recorders output directly to MP3. So, aside from the argument from original source, what extra value exists in the sound of the human voice speaking in uncompressed audio?
One argument against the MP3 is its actively enforced patent. However, the MP3 U.S. patent will expire at the end of 2017, so maybe things will change.
I think oral history is a case where MP3s make a lot of sense– the audio files are often large, particularly if interviews are longer (Most of my interviews are over an hour long). I would expect that little-to-no post-production is done on the recordings, apart from some volume adjustment (e.g., adding EQs/filters/effects), so there’s less “data” there to analyze.
I think there are circumstances where you would want to go less-lossy and even analog if that was possible, particularly when working with musical recordings. This would take a lot more storage, and in turn, cost a lot more money. It’s worth noting that the difference in file formats when it comes to MP3s vs. WAVs also depends on the hardware on which you listen to them. To hear the difference in these files, you need to have amplifiers and speakers/headphones that really can give a good “image” of the frequency spectrum. Why have .WAV files if your researchers are trying to hear the minutiae of a recording on earbuds? An archive could hope that they might be able to provide a professional listening environment in the future, or count on users to provide their own, but the hardware needs to be considered along with the file format.
I’ve shared this before, but it’s worth sharing again. In case you’ve never attempted to differentiate MP3s from WAVs, here’s a test: http://www.npr.org/sections/therecord/2015/06/02/411473508/how-well-can-you-hear-audio-quality
Jeff – I think you’ve hit the nail on the head when you say that the goal “good enough for most people, most of the time.” So much of library, archive, and info science is coming up with some sort of standardization that we think can be most useful to the most number of people.
But Kirschenbaum’s notion of the “heterogeneity of digital inscription” highlights the fact that we’re not dealing with uniform materials. While many of us info professionals may long for standards, it’s not a desire that inherently maps well onto digital objects. We may have format/file preferences, but we can’t control how materials are created.