Spotify - anyone heard of it?

Well, you can look at the thing without listening to it. Is that less scary?

I live in South Africa, and am considering going through the rigmarole to get Spotify up-and-running on my connection here.

Is there something like a ILM playlist on there?

I would really like to see a list of what's currently on that that. Is someone willing to do me a favour and screengrab a list of what's on the ILM playlist at the moment?

I would like to see if it something that would be worth my effort, as that would be my primary reason for getting Spotify.

Somewhere in this thread may be the Spotify mix you are looking for: Rolling Favorite Tracks + Albums 2015

i like the discover playlists and regularly check them out - any time a track choice offends me i can always tell WHY the robot thought i'd like it. but it has put me in an ironic position of feeling like big brother's not doing a good enough job at watching me sometimes. "an exile on guyville track? I KNOW i've played exile on here before...weren't you paying attention, algorithm?"

Also essential for ILX listening: Listening to ILX Listen - 2015 Spotify Genre Playlists

Oh wow I didn't even know that thread existed!

croup i think it knows you've played just maybe not in a while. that could be useful, i'm sure there's plenty of albums/acts i listened to once cuz they were on aquarius records mailing list or whatever and enjoyed but never got around to listening to again and would enjoy 'rediscovering', the problem is spotify is understandably unable to distinguish between those albums vs albums you've only listened to once in the past decade because you wore them out long ago vs albums you listened to once cuz it was on a list and once was enough thx.

Wish Spotify could import my last.fm history to help with recommendations. I'd pay for that.

I also maintain a playlist of the "Your 11 favourite songs of the moment, fuck" thread: http://open.spotify.com/user/miketd/playlist/01h5cX8jtxgfJoaOBgYN4F

an article on Spotify / The Echo Nest / Listener profile Genre Clusters / Discover Weekly

Spotify is getting unbelievably good at picking music — here’s an inside look at how
http://www.techinsider.io/inside-spotify-and-the-future-of-music-streaming

on Discover Weekly

Spotify's work with The Echo Nest has culminated (so far) in this summer's launch of Discover Weekly, that custom playlist that magically predicts what I and every other Spotify listener wants to hear. Even though the feature is barely a month old, the overall reception in reviews and on social media has been overwhelmingly positive.

“Discovery can feel like work, and we wanted it to feel very human and natural like the selections that are powering it,” Matt Ogle, the guy in charge of Discover Weekly, tells TI.

Ogle previously worked at Last.fm, a startup that pioneered a lot of work on analyzing music data, and a now-shuttered startup called This Is My Jam, which let you choose one song a week to share with your friends. Since joining Spotify earlier this year, Ogle has tried to combine both approaches to blend big data with a human touch.

Ogle gives a simple analogy for how Discover Weekly works: You’ve been playing song A and song C a lot, but it turns out that when other people play those songs together in their playlists there’s a song B that you’ve never heard before.

Discover Weekly gives you song B.

“We look at what you’ve been listening to and what are the songs playing around these songs that you’ve been jamming on,” Ogle tells Tech Insider. “We’re trying to find the missing tracks.”

During internal testing, his team realized that if you don’t recognize a single artist in a playlist, you might question if it’s actually geared for you. That’s why the playlist is intended to have a mix of mostly new tracks with a few songs you’ve heard before.

“Having a little bit of familiarity is key to building trust. It can be exhausting to just listen to stuff you’ve never heard of before,” he says.

Now it's true that the songs on Discover Weekly are chosen by algorithm, not humans, but Ogle insists that this misses the point since the whole thing is built on data created by humans — it's just that algorithms are connecting the dots on a massive scale.

“There’s something compelling about this humans versus robots narrative: a lovingly curated playlist versus an algorithm screwing up your sexy time,” says Ogle. "That whole distinction no longer really describes how we work. Discover Weekly is humans all the way down. Every single track that appears in Discover Weekly is because other humans being have said, ‘Hey this is a good song, and here’s why.’"

As popular as the new feature is, it's just the beginning for Spotify.

“We want to make sure that as mainstream or hipster as you are, Spotify can cater to you,” says Ogle. "That will involve humans and machines, and ideally it involves them skipping down the street hand-in-hand, because practically that’s the only way to get things done.”

“I see Discover Weekly as one of the first products from this new era of personalization, but ultimately we’d love for everything you interact with on Spotify to feel like there’s a bit of you in it.”

collaborative filtering is cool yes

I don't know how you fit it in the Spotify UI but I'd pay more attention to their recommendations if they came with some context. Even an Amazon-style "because you listened to A and C" explanation would be a start.

Spotify insists on giving me a Discover playlist dominated by male-centric indie every other week no matter what i've been listening to & it really makes me question the data fueling this project

glenn, any insights re: whether this issue will be addressed?

it knows you too well

If you give each song a fair hearing, say three plays each, that's 90 songs, which certainly makes up a fair amount of my weekly listening. I wonder if there's a feedback loop going on - a self-reinforcing downward spiral of male-centric indie.

snrk

ha, i skip over the songs i'm not feeling

> I wonder if there's a feedback loop going on

i thought the same thing.

> is this all because it's using your playing of last week's recommended racks in figuring out this week's?
> ― koogs, Monday, 10 August 2015

i imagine it's very easy for this to happen.

and this data ends up in the pool for everybody else too. eventually it'll get to the point where you may as well just listen to steve lamaqc.

i just want a way to give dan deacon the world's biggest thumbs down, on spotify

I get the same thing with Gaz Coombes. Gaz Coombes!

Last few weeks I had some pretty interesting eclectic Discovery playlists, with music spanning a variety of time periods and genres, but this week's playlist was pretty standard Indie/Pitchfork fare. I'm going to blame this on the fact that I decided to listen to a Spoon album earlier in the week, and now it just thinks I like boring music.

my playlists are pretty cool, they resemble one of those CDs that come with Mojo Magazine with some techno randomly thrown in

I think this week is the first time my Discover playlist has had some tracks on it that had already appeared on a previous weeks discover playlist (that I've noticed)

(tracks are 'I'm in Love with a German Film Star' and 'Looking for Clues', both of which were on my first Discover weekly playlist because I remember being impressed/unsettled that it had successfully identified two of my favourite songs)

"We think you would like <unknown artist> just as much as we think you'd like Tame Impala"

Discover Weekly is just the first of many things we're working on, and a bunch of the others have more dimensions. More to come.

the bastards gave me the feelies. Need a thumbsdown urgently

how dare you

Discover Weekly is just the first of many things we're working on, and a bunch of the others have more dimensions. More to come.

can't wait for these. discover weekly has more often than not been bang on for me.

this week bar the feelies has been stuff i already really liked. no new discoveries but i guess they know stuff i do like

The Feelies seem an odd band to urgently dislike.

oh wait right down at the bottom there is a band ive never heard of called Lonely Kamel. The shitty name suggests stoner rock?

data mining for bragging rights: https://spotify-foundthemfirst.com

That puts me in the earliest 1% of listeners for Clean Bandit, Naughty Boy & Gorgon City, the earliest 3% for Charli XCX, 5% for Ariana Grande, 8% for The 1975 & Sturgill Simpson, 9% for Quadron.

actually, one thing that I would find interesting, though I doubt it's got broad enough appeal to be a priority for implementations, is something like those "artists unique to this state" maps, except with artists unique or at least particular to particular users (followers only, if this seems too invasive)

i'm a sturgill simpson 2 percenter. he should be required to come to my house and do a living room concert or something.

i like these kind of toys
top nine percent for rich homie quan, kacey musgraves, a$ap ferg, earl sweatshirt, quadron
top five percent for fka twigs, ariana grande, lorde, charli xcx
two percent for alunageorge, fifth harmony, becky g
one percent for FETTY WAP

i've actually been enjoying the discover weekly thing, to my surprise. found a few new (to me) artists through it already. also a lot of stuff i don't like at all but i'm willing to sift through it. however,

If you give each song a fair hearing, say three plays each, that's 90 songs, which certainly makes up a fair amount of my weekly listening. I wonder if there's a feedback loop going on - a self-reinforcing downward spiral of male-centric indie.

i think this is an otm problem, not just with spotify but with predictive AI in the first place. i've been reading nick bostrom's Superintelligence the last few days and have been going down the rabbithole of discussion on this stuff. one good thing i recently read was by jaron lanier (http://edge.org/conversation/the-myth-of-ai) and he talked about the recommendation feedback loop problem a little:

I want to get to an even deeper problem, which is that there's no way to tell where the border is between measurement and manipulation in these systems. For instance, if the theory is that you're getting big data by observing a lot of people who make choices, and then you're doing correlations to make suggestions to yet more people, if the preponderance of those people have grown up in the system and are responding to whatever choices it gave them, there's not enough new data coming into it for even the most ideal or intelligent recommendation engine to do anything meaningful.

In other words, the only way for such a system to be legitimate would be for it to have an observatory that could observe in peace, not being sullied by its own recommendations. Otherwise, it simply turns into a system that measures which manipulations work, as opposed to which ones don't work, which is very different from a virginal and empirically careful system that's trying to tell what recommendations would work had it not intervened. That's a pretty clear thing. What's not clear is where the boundary is.

If you ask: is a recommendation engine like Amazon more manipulative, or more of a legitimate measurement device? There's no way to know. At this point there's no way to know, because it's too universal. The same thing can be said for any other big data system that recommends courses of action to people, whether it's the Google ad business, or social networks like Facebook deciding what you see, or any of the myriad of dating apps. All of these things, there's no baseline, so we don't know to what degree they're measurement versus manipulation.

But it's important to understand it if this is becoming the basis of the whole economy and the whole civilization. If people are deciding what books to read based on a momentum within the recommendation engine that isn't going back to a virgin population, that hasn't been manipulated, then the whole thing is spun out of control and doesn't mean anything anymore. It's not so much a rise of evil as a rise of nonsense. It's a mass incompetence, as opposed to Skynet from the Terminator movies. That's what this type of AI turns into.

i'm a charli xcx one percenter and i blame johnny fever

anyway to listen to previous Spotify Discover Weekly playlists?

Create a folder and save them each week.

lol the only artists I got were iLoveMakonnen and A$AP Ferg

i'm a charli xcx one percenter and i blame johnny fever

I am comfortable with this.

Glenn, the new Peaches track "Dick in the Air" is attributed to the wrong Peaches. http://open.spotify.com/album/7F5bywB7OEgeoFYFf6CJPD

Create a folder and save them each week.

― MarkoP, Wednesday, 9 September 2015

I was on holiday so I was hoping to get last monday weeks one.

Ew on Peaches. Will pass that along.

disambiguating artists is a lot harder than it seems like it would be

i like these kind of toys
top nine percent for rich homie quan, kacey musgraves, a$ap ferg, earl sweatshirt, quadron
top five percent for fka twigs, ariana grande, lorde, charli xcx
two percent for alunageorge, fifth harmony, becky g
one percent for FETTY WAP

― Meta Forksclove-Liebeskind (forksclovetofu), Wednesday, September 9, 2015 12:24 PM (4 hours ago) Bookmark Flag Post Permalink

haha i'm identical or similar on all of those except quadron, fetty, alunageorge, 5th harmony and becky

1 percenter for charli, fka and banks

I'm at 1% on Charli XCX, FKA Twigs and Run the Jewels (who I can't remember ever listening to tbh). Also have Courtney Barnett and AlunaGeorge. Can't believe I was too slow for the Becky G train.

first 1% for fka twigs, k michelle, charli xcx, ten walls :/

obv more popular artist (all of the above meet a minimum threshold, none obscure really) makes it easy to get in that earliest percentile, would be curious who the hell i'm among the first 10 or 100 or 1000 listeners of if anyone

lol i wonder what percent of the first 1% of fka twigs or charli xcx listeners are on ilx or lurk at least

pretty sure they only compute it for artists that "made it" by some metric, but yeah

i'm pretty sure the other formula would just yield 'holy shit this old local band's one album is on spotify'