Itunes, Billboard, and the marginalization of black music and black audiences in America

not to be all fact-check police but the Carly Rae Jepsen thing is almost certainly unrelated to this, considering it happened in 2012: http://www.carlyraemusic.com/2012/10/make-your-own-lyric-video-and-win-a-skype-call-and-signed-goodies/

(different song, of course, but same idea.)

I assume they count any videos where an artist/song has been recognized by YouTube.

yeah that seems to be the gist, any recognition is registered and paid for. i'm just curious if there is or ever will be a statute of limitation.

the whole 'lyric video' thing seriously is out of control though. Justin Timberlake is on-camera, occasionally mouthing words of the song, in the "Suit & Tie" lyric video, but because the words are on the screen and it's not a super high budget production that's just the lyric video, and the 'real' video was something else.

A board member of yore weighs in

http://t.co/YYHOtAKK86

pretty sure this lyric video was made by a 12-year old

https://www.youtube.com/watch?v=D6S-0SbIVgs

that's my niece you're talking about!

watching 2 seconds of a 4-minute YouTube always counts towards its views and always has, though. Lex bringing that up raises a whole other issue about all this.

― D4y0 (some dude), Friday, February 22, 2013 6:03 AM (3 hours ago) Bookmark Flag Post Permalink

this is not actually true from what I understand

ha well you def know more about YouTube than me, i thought all view-based traffic was measured by the "open the page and you're counted" rule. what is your understanding?

Looks like some dude is right?
http://www.quora.com/YouTube/How-far-into-the-video-does-a-user-need-to-watch-for-YouTube-to-count-it-as-a-view
http://www.atlantaanalytics.com/practicing-web-analytics/how-does-youtube-video-view-count-work/

Anyway, I wonder how many songs this year are going to come with dance moves attached.

you say that like half the songs out there don't have dance moves attached to them already

think the thing that annoys me most is the 30 sec thing? even if you ignore the fact that checking a song out on youtube doesn't equate to actually LIKING it, at least limit it to people consuming the actual full song

being forced to listen to a song over and over again on the radio doesn't equate to actually liking it either! also looool that you want to limit it to people listening to the "actual full song."

Maybe they'll somehow work YouTube Likes and Dislikes into the system.

this harlem shake song. does it really only last 30 seconds? (i only heard about it yesterday sorry if im asking a dumb question)

someone 'liking something' and something being popular aren't the same things regardless

ie like the billboard chart never strictly reflected 'the most liked' music

anyway related to this: http://www.npr.org/blogs/thetwo-way/2013/02/21/172644482/tv-ratings-agency-nielsen-will-begin-measuring-online-streaming

this harlem shake song. does it really only last 30 seconds?

it's normal length (4 mins or something) but some of the meme videos that are being monetized / counting toward it's #1 chart position are that short

it's 3:18

ahh ok. I did wonder.

is it really less than 4 minutes, it felt like it went on for 10

Has there ever been a song as short as 30 seconds at #1? Or is there a minimum limit? I know in the UK in the 90s they changed the maximum to 19mins 59 secs because the orb,fsol etc were releasing 40 min singles and charting.

Looks like some dude is right?
http://www.quora.com/YouTube/How-far-into-the-video-does-a-user-need-to-watch-for-YouTube-to-count-it-as-a-view
http://www.atlantaanalytics.com/practicing-web-analytics/how-does-youtube-video-view-count-work/

― marc robot (seandalai), Friday, February 22, 2013 10:15 AM (1 hour ago) Bookmark Flag Post Permalink

ahh interesting—it does say in analytics you can see what point people stop watching

that feels a bit creepy

I would assume that billboard is smart enough to count the views per week, not just the cumulative total. In which case they must be getting data directly from youtube I would think. And if so there's no reason they couldn't also get U.S. only data, filter out the plays that are too short, etc.

The shortest ever #1 hit in the US is "Stay" by Maurice Williams & The Zodiacs. Something like a minute and a half.

or about three harlem shakes, which is the new convention for measuring time

3 malted harlem shakes please

and one shamrock shake

https://twitter.com/McDonalds/status/303230001342472192

being forced to listen to a song over and over again on the radio doesn't equate to actually liking it either! also looool that you want to limit it to people listening to the "actual full song."

feel like this is as good a place as any to confess than, on purpose, my iTunes is set to only count plays that play 100% of the song/track

Been waiting for that. xp

tbf don't most radio programming surveys only play the first ten seconds of a song to ppl surveyed to see if a track is 'known'?

so is that why so many songs start with a chorus now?

that's not why, but a lot of songs do do that to try and hook people as quickly as possible

also while i can totally believe (and assume) billboard's methodology here is off in principle this move is an obv good thing, in terms of trying to capture snapshot datapoint of a song's popularity, relative popularity of songs for a certain week in time right? it's a move toward greater accuracy (and unlike weighing sales v airplay ratio, youtube and streaming is obv under airplay and the only question is how you calculate and weigh it in that number), unlike the moves w/ r&b, country, etc charts.

starting songs w/ chorus is old as the hills
https://www.youtube.com/watch?v=QoF-7VMMihA

also a good move in that it simultaneously dilutes the weight of radio airplay, reflecting the greater culture (something radio itself has been reflecting increasingly for a little while now - hello fm talk radio, goodbye rock radio).

there's no such thing as greater accuracy, there are just different ways of defining 'popularity'. and the old ways of defining it are becoming increasingly absurd.

yeah obviously it's a standard way to do a song, i'm just saying people writing big budget pop these days have openly said there's pressure to go straight for the hook. xp

personally i think any argument for this move just highlights how silly it was that they never made MTV and other video channels a factor

there's no such thing as greater accuracy
defining terms is kind've a first step, pretty sure billboard has an idea of what they mean by popularity. also will let everyone else in the lab know we can feel free to just go by guesstimates and eyeballing stuff, who needs assays and a280s.

personally i think any argument for this move just highlights how silly it was that they never made MTV and other video channels a factor

― D4y0 (some dude)

too true, ridiculous at the time but even more ridiculous now when '20 million views' can have chart impact and '20 million views' would've been what a heavy rotation video would've pulled on a friday night way back when. they may have just been reflecting the prerogatives of their readership (radio lobbying for mtv to be excluded would be plausible to me). not that the record industry was blind to the power of the huge, popular national radio station, the mtv oral history has many tales including label interns and flunkies attempting to stuff the ballot box w/ 1-800-dial-mtv, etc.

personally i think any argument for this move just highlights how silly it was that they never made MTV and other video channels a factor

the major difference being that mtv was basically just a promotional outlet that actually cost the labels money (to make the videos, etc) while youtube is actually a revenue stream. gangnam style probably made about $1 million from youtube, which is the equivalent to selling a million singles. why shouldn't that be factored in pretty heavily?

there's no such thing as greater accuracy

surely the data that youtube collects is much more objectively accurate than whatever voodoo nielsen does with radio.

too true, ridiculous at the time but even more ridiculous now when '20 million views' can have chart impact and '20 million views' would've been what a heavy rotation video would've pulled on a friday night way back when.

I think you're vastly overestimating the size of a typical friday night mtv audience when they were playing music videos. 2011 vmas were mtv's biggest audience ever with only 12 million viewers http://www.mtv.com/news/articles/1669953/vma-2011-ratings-history.jhtml

it speaks volumes that PSY had to get a billion views to generate income comparable to a million single sales

defining their terms doesn't make it more acccurate because 'popularity' isn't something that can be accurately judged. the numbers they plug into their formula can be more or less accurate but the number it spits out is never going to have anything to do w/ accuracy. if they had a clear idea of what they mean by popularity they wouldn't be radically changing that number crunching machine multiple times in a year.

there *isn't* some clear idea of popularity because the way media is consumed is changing basically year to year. spotify was barely a thing fairly recently. it could be replaced by something different soon. etc. that doesn't mean these things shouldn't be included, it just means billboard won't be able to come up w/ some magic formula that lasts a decade unless the way people consume media stops changing so quickly.

xps

it speaks volumes that PSY had to get a billion views to generate income comparable to a million single sales

how many radio plays or mtv views do you think it takes to convert into a single sale?