Re:An experiment in crowd-sourced songwriting
by SwellJoe on Thursday September 01, 2016 @01:05PM
(#12406990)
The majority of the people are completely lacking in taste the majority of the time.
Also, the majority of people have no musical training; not even a few hours tinkering with a music app to have a vague understanding of what makes a song meaningful; basic stuff like tension and resolution and song structure. This is effectively a random selection of notes within the key, and I bet "no note" will never win the vote, so there will be no interesting rhythmic variation. A simple script would work as well as these 65,000 people.
That said, there probably are interesting (though questionable in their artistic merit) ways to crowd source an artistic work. Focus groups for films and TV shows and such have been happening for a couple decades. There have been interesting surveys of popular harmonies and melodies, which could be applied to scientifically build songs that fit what is popular (and I'm sure it has been done). The uncanny valley is probably where you'd end up in most cases, though; at least for now.
Some interesting work in the area of understanding what crowds like:
Re:An experiment in crowd-sourced songwriting
by cpeterso on Thursday September 01, 2016 @02:43PM
(#12407813)
Somewhat related:
"The Most Unwanted Song" is a novelty song created by artists Komar and Melamid and composer Dave Soldier in 1997. The song was designed to incorporate lyrical and musical elements that were annoying to most people. These elements included bagpipes, cowboy music, an opera singer rapping, and a children's choir that urged listeners to go shopping at Walmart.
Re:An experiment in crowd-sourced songwriting
by 0xdeadbeefbabe on Thursday September 01, 2016 @12:06PM
(#12406427)
> The song is currently in skeleton form. When performed by artists, they are at liberty to use their artistic interpretation.
That's a wise move for which the crowd didn't get to vote. Same goes for the chord structure. It's as if an individual is trying to make the crowd look good?
Re:An experiment in crowd-sourced songwriting
by TheOtherHobbes on Thursday September 01, 2016 @01:25PM
(#12407180)
This would be a lot more interesting if you could "buy" pretend "shares" in each possible note and then get pretend "rewards" if your choices turned out to be correct.
Re:An experiment in crowd-sourced songwriting
by md224 on Thursday September 01, 2016 @01:04PM
(#12406978)
Shameless plug: if you're interested in crowdsourcing, check out my subreddit AskOuija (https://reddit.com/r/AskOuija). It's a platform for linguistic collaboration at the most granular level possible: individual letters. This diffusion of authorship leads to some pretty interesting (and pretty stupid) results. Also it gave me an excuse to write a Reddit bot.
Re:An experiment in crowd-sourced songwriting
by exorcet on Thursday September 01, 2016 @01:01PM
(#12406952)
Lololol. Guess "When In Rome - The Promise" is the new Rick Roll?
The tune wad immediately recognizable, but i didnt know which song. The lyrics tho - "hmmm..na..hmm.. The right words to say" I did remember. Googled "the right words to say lyrics" and bam - the promise - perfect match.
So was original song created via wormhole travel to grab crowdsourced song from the future? Or is the crowdsound song intentionally being shaped to match? Or all one big coincidence?
Re:An experiment in crowd-sourced songwriting
by akeck on Thursday September 01, 2016 @01:29PM
(#12407208)
I couldn't find a TOS. There was no click-through license that granted rights one way or another. So... if they get something really good, defending their copyright on the piece will be difficult?
Re:An experiment in crowd-sourced songwriting
by sova on Thursday September 01, 2016 @02:22PM
(#12407656)
They own the platform so.. I don't think it would be difficult. How else could it come into being without their hosting of the platform? Then again, if there were a class-action suit... perhaps, but it seems more like a task of fancy than a task of profit. Anyway, music is kinda a different ballgame thanks to ASCAP
Re:An experiment in crowd-sourced songwriting
by dublinben on Thursday September 01, 2016 @02:32PM
(#12407722)
Owning the platform doesn't make them the author of the piece. Each member of the crowd can claim to be a partial author, which is a copyright nightmare.
Re:An experiment in crowd-sourced songwriting
by peeters on Thursday September 01, 2016 @12:35PM
(#12406694)
The first time you try it, it's a mildly successful novelty. The second time you try it, 4chan causes the site to get sued by Metallica for copyright infringement.
Re:An experiment in crowd-sourced songwriting
by anotheryou on Thursday September 01, 2016 @12:24PM
(#12406599)
This is why crowd sourcing for anything touching "taste" fails. Even suggestion engines fail with this (I don't like the cliche artists of the genres I listen to).
Re:An experiment in crowd-sourced songwriting
by SwellJoe on Thursday September 01, 2016 @12:50PM
(#12406857)
Yep. This is why every station I have on Pandora plays Pearl Jam, the Cure, and Bad Religion. All range from "OMG, again! why!?" to "mild boredom" in how I feel about them. No matter how many times I dislike Pearl Jam and Eddie Vedder songs, they keep coming back; something like 30 or 40 dislikes, now, but the will of the crowd is that someone that likes the things I like is clearly supposed to like Pearl Jam. (One could argue this is a problem in their algorithm. By the third or fourth dislike of an artist, that artist should probably be purged from the possibilities.)
Re:An experiment in crowd-sourced songwriting
by anotheryou on Thursday September 01, 2016 @01:03PM
(#12406976)
Especially pandoras metrics just don't cut the subtleties of "good taste". E.g. listening to something like Bach, with many interprets, there will be just a few that play it to my liking, even though most superficial (every single note) stays the same.
Another problem might be the inability to split a personality. Many engines take the 7 people that behave most like you as a reference, but might compromise when trying to find people that like exactly the genres you like. (if you try to find people that listen to the same set of genres your pool of candidates gets so small, that the similarities within the genres are very few).
Re:An experiment in crowd-sourced songwriting
by fao_ on Thursday September 01, 2016 @12:42PM
(#12406764)
> This is a gimmick
That's probably what it was intended to be. Are gimmicks bad now?
> A melody doesn't even have 65k notes
If you had followed the link, you would see it works via a vote for the next note / word in the lyrics.
> It's like having 65k people writing a text. There's no structure, leadership, direction, etc
Many, many successful things have worked without leadership or structure. I disagree, there is direction. People choose the note/word that is most appealing to them, etc.
Also, the majority of people have no musical training; not even a few hours tinkering with a music app to have a vague understanding of what makes a song meaningful; basic stuff like tension and resolution and song structure. This is effectively a random selection of notes within the key, and I bet "no note" will never win the vote, so there will be no interesting rhythmic variation. A simple script would work as well as these 65,000 people.
That said, there probably are interesting (though questionable in their artistic merit) ways to crowd source an artistic work. Focus groups for films and TV shows and such have been happening for a couple decades. There have been interesting surveys of popular harmonies and melodies, which could be applied to scientifically build songs that fit what is popular (and I'm sure it has been done). The uncanny valley is probably where you'd end up in most cases, though; at least for now.
Some interesting work in the area of understanding what crowds like:
http://www.hooktheory.com/blog/i-analyzed-the-chords-of-1300...
http://articles.latimes.com/2012/feb/03/science/la-sci-hit-s...
https://en.wikipedia.org/wiki/Hit_Song_Science
"The Most Unwanted Song" is a novelty song created by artists Komar and Melamid and composer Dave Soldier in 1997. The song was designed to incorporate lyrical and musical elements that were annoying to most people. These elements included bagpipes, cowboy music, an opera singer rapping, and a children's choir that urged listeners to go shopping at Walmart.
https://en.wikipedia.org/wiki/The_Most_Unwanted_Song
Most Unwanted Song: https://www.youtube.com/watch?v=-gPuH1yeZ08
Most Wanted Song: https://www.youtube.com/watch?v=McIfIx29tSg
That's a wise move for which the crowd didn't get to vote. Same goes for the chord structure. It's as if an individual is trying to make the crowd look good?
The tune wad immediately recognizable, but i didnt know which song. The lyrics tho - "hmmm..na..hmm.. The right words to say" I did remember. Googled "the right words to say lyrics" and bam - the promise - perfect match.
So was original song created via wormhole travel to grab crowdsourced song from the future? Or is the crowdsound song intentionally being shaped to match? Or all one big coincidence?
https://www.youtube.com/watch?v=oOlDewpCfZQ
Another problem might be the inability to split a personality. Many engines take the 7 people that behave most like you as a reference, but might compromise when trying to find people that like exactly the genres you like. (if you try to find people that listen to the same set of genres your pool of candidates gets so small, that the similarities within the genres are very few).
A melody doesn't even have 65k notes
It's like having 65k people writing a text. There's no structure, leadership, direction, etc
That's probably what it was intended to be. Are gimmicks bad now?
> A melody doesn't even have 65k notes
If you had followed the link, you would see it works via a vote for the next note / word in the lyrics.
> It's like having 65k people writing a text. There's no structure, leadership, direction, etc
Many, many successful things have worked without leadership or structure. I disagree, there is direction. People choose the note/word that is most appealing to them, etc.