Monday, July 20, 2015

Music news round up #2

This week, here are some more music articles that I found interesting to read:

1. An article about a scientific study looking for the universalities in music across disparate cultures. The result? Small scales, arching or descending melodies, rhythms based two or four beats, and men singing. Of course, there are many outliers, and this article is dumbed down from the actual research. And I wonder if the focus on men might be a bias by the people who put together the body of work they analyzed, the Garland Encyclopedia of World Music, though probably not.

2. An article about the lack of transparency in the music industry about where the money goes. "Anywhere from 20-50 percent of music payments don't make it to their rightful owners." You can hear a somewhat related story in a recent Planet Money podcast about manufacturing the song of the summer (rebroadcast from 2011) and how much money the industry gambles on making a hit.
 
3. An article about rape and abuse of teenagers by people in Rock & Roll. If you weren't aware already, an industry focused on sex can have some really bad side effects. Eye-opening and horrifying, perhaps, but not surprising. (P.S. unlike the other two, this article doesn't have a picture of Taylor Swift).

Happy reading and/or listening!

Tuesday, July 14, 2015

The ethics of field recording

Ethnomusicology has changed as the world has become more egalitarian. Instead of the white man showing up in a small village, setting up a wax cylinder or microphone, recording some volunteers, and then selling the recording for their own scholarship or profit with little or no attribution, native performers are now getting more credit and monetary rewards for being recorded.

But how do we correct the wrongs that have been done in the past? Well, museums are giving back pieces of art work and historical objects that had been stolen. Sometimes, this is called repatriation (though the Elgin marbles are not going back to Greece anytime soon). Can we do that with recorded music, too?

Well, NPR did a story recently about someone trying this approach, going to a village to give back music recorded 65 years before:


http://www.npr.org/2015/06/28/417462792/in-a-kenyan-village-a-65-year-old-recording-comes-home
 

Besides a delightful vignette of villagers singing their version of Jimmie Rodgers the man-eater, this story touches on a limit of this approach of musical repatriation: who has CD players? Not these villagers.

In librarian terms, this is a recording preservation issue—in order to preserve and play back a recording, you need a specific machine, and those machines aren’t as prevalent as the recordings. Case in point: they don’t even make VHS players anymore (in fact, I would bet that there are more 33 1/3 players than VHS players nowadays). CD players will probably die out faster than CDs do (I’m giving them another 30 years). But there, I guess, is a root of the field recorder’s problem—when the recordings were made originally, these villagers were not ready to keep and maintain the recording machine or preserve the recordings; and it looks like they still don’t really have a preservation plan. Does everyone making a field recording need to also teach that culture how to maintain a playback machine and store the recording? It seems a tall, perhaps impossible order in some cases. But even if the culture could keep and preserve the "original" recording, music recordings are inherently a copy, so we can’t really return the object as is the case with physical art.

So, while music repatriation is an interesting idea, and could certainly work in some cases, I think we’ll have to use other approaches to paying people back for music exploitation most of the time. Or at least, when an ethnomusicologist finds a new music-making populace and before they start to record, they should think for a while about ethics and how a recording could best benefit that particular population.

Monday, July 6, 2015

Signifying Sound and Fury's Apple Music review: discovery limits and metadata woes

It's all shiny and colorful now...but what's on the inside?

This week saw the release of the long-awaited Apple Music, a music streaming service. Although I’ve never tried a streaming service before (besides Pandora Radio, which is a very different model), I decided to try out the free trial period, and I’ve been playing around with it this week.

Although the New York Times’ initial reaction stated that the Apple music’s design was strong, I would argue that there’s some big problems with discovery and metadata (which is a big part of discovery), among other small issues.

Discovery


The first thing a person encounters when signing up for the service is a bunch of bubbles with artist names. I thought these bubbles were weird—setting aside the limited genres, even when I kept hitting the “more artists” button, I was surprised at how few artists were featured. Also, once you’ve selected some artists as your favorites, liked bubbles stick around on the screen while the new artists come in, limiting the number you can pick. Why did I need to pick favorites among such few artists? Pandora’s system, with thumbs up and thumbs down when you here the song, makes much more sense. Supposedly, the new heart system (you can heart both songs or albums), similar to Pandora’s system, also leads to more personalized “For You” results, despite me adding lots of hearts, I haven’t really seen any change in my “For You” recommendation tabs. Also, even when I have gone back and un-liked on artist I initially had liked, the recommendations did not change. And why can’t we heart by artist, as well?

While you can always search for artists or albums (though this was difficult to find, too), the other main problem is the limited ways to browse to discover music. There are only two ways I can see to browse music: the “For You” tab (a list of playlists and albums created by your preferences and supposedly your “hearts”), and the “New” tab (a curated list of new music). Why aren’t there any other ways to browse music? It’s pretty easy to browse by genre in the iTunes store; why can’t this functionality be used for streaming music? And if I browse music genres in the iTunes Store, iTunes doesn't make it easy to stream that music; you have to go back and search in Apple Music (which involves a couple of other clicks). Also, why not, like Spotify, allow Apple Music users to see playlists from friends (note: you can share playlists, but it is limited and hard to do) or search user-created shared playlists? (Apple Music also has lot of “radio stations” that I haven’t explored yet, but those again are curated lists).

Metadata


But a larger issue here is metadata. Pandora is meticulous about their metadata, creating an entire genome of metadata about a song. As I already know from the data I normally get when I download an album, Apple’s metadata is limited to say the least. I frequently change genres and have my own list of genres that I use because my own collection is so big, I need to discover it again myself. Some of my albums I’ve even recently bought don’t have album covers uploaded, to say nothing of the beaten-up albums I bought used at library sales (which form a large part of my collection). I don’t know how well Apple can coordinate good radio stations or playlists when their metadata and genre distinction is so bad. One of the worst offenders: iTunes thinks that “Artist1 (feat. Artist2)” does not mean that "Artist1" is the artist of a song; sometimes iTunes even creates a different album for these qualified statements of responsibility.

Apple Music albums seem to have even less metadata than the iTunes store version of the album, such as the day the album is released, the popularity of tracks on the albums (which I find very useful), or reviews. Wouldn’t it be easy to just transfer that information over? But of course, neither platform includes musicians on the record or composition credits or liner notes.

Other problems


Besides the main problems of browsing and metadata, there are some other problems:
1. I was forced to sync my music with iCloud, which I have resisted until now. This sync took a very long time because I have a lot of music, and was not very effective; Apple is very bad at matching your songs to the versions they hold in the cloud, probably because of the bad metadata.

2. I signed up for the family account, but it was not intuitive at all how to add other people to account. In fact, I discovered that other family users have to be Apple users to share the account; so this option won’t work for people with iTunes on their PCs (iTunes homesharing was removed, too).

3. One creepy thing: Apple Music Connect figured out which music artists I’m already following on Twitter. On the other hand, if they are going to be that creepy, why couldn’t they use that information to tell which artists I already like to add to the "For You" recommendations?

4. Unlike many people, I’m not connected to the internet all the time, so streaming doesn’t work the best for me on the road. Apple Music, however, does allow users to download songs for listening later, which would be awesome except that my iPod touch is too old to have the latest OS or iTunes app.

Take-away


My take-away for Apple Music: like other streaming services, you can find a lot of music, but it is hard to use, doesn’t allow for music discovery beyond their narrow scope, and does not use metadata at all well, even metadata that is already available to them. Right now, I’m planning on enjoying many albums I haven’t yet had the chance to hear (from lists I’ve curated myself from friend recommendations and new music podcasts), but at the moment, I don’t plan to renew once my free trial is up. On the other hand, maybe even the bad service is worth $10 a month—though I’m not optimistic that much of this money will trickle down to the artists making the music.