Hacker News new | past | comments | ask | show | jobs | submit login
How bad is your Spotify? (pudding.cool)
974 points by feross on Dec 23, 2020 | hide | past | favorite | 607 comments

This is hilarious...

> You are 26% basic. You're trying to be cool with Kapitan Korsakov, but your favorites are the same as everybody else's..

> Your spotify was please-read-my-manuscript-walmart-hawaiian-shirt-sitting-alone-in-the-cafeteria bad.


Really interesting project by all means. I hope they release the source - I'm curious about how it was built.

Edit: This same link was posted 11 hours ago https://news.ycombinator.com/item?id=25513886 here and is not at all getting the same popularity. Anyone got ideas?

> You're stuck in the early 2010s. You only listen to Obama-era jams like Nocturne in B-Flat Minor, Op. 9, No. 1 by Frédéric Chopin and Piano Concerto No. 21 in C Major, K. 467: II. Andante by Wolfgang Amadeus Mozart.

The output it spitted out while processing was pretty good though.

> oh great another Rise Against stan... > You've been listening to a lot of Logic lately. > u okay? > >Yeah why > no reason... > Of course Twenty One Pilots.

Not gonna lie, I'm feeling a bit judged by this AI:)

If it's any comfort I got the _exact_ same sequence of comments/questions but with different bands I listened to a lot in 2020..

Hope that didn't ruin the magic.

It kinda did. I was enjoying being judged but now that I know that everyone is, I feel less special.

My favorite part was knowing it was an AI but still feeling at least a little defensive at parts

> You've been listening to a lot of Logic lately. > u okay?

Asked the same thing to me, but about Radiohead.

Same , but with:

- Ryan Caraveo, u okay? - ...of course, K.Flay

Lol. Mine was LCD Soundsystem.

Asked the same thing to me, but about The Replacements.

For me it was Asking Alexandria, which fit pretty well as they have mostly dark songs.

What's moody about the placemats?

Well, if you consider drunken sprees a mood... :)

On second thought I do think Paul Westerberg has a strong melancholy streak.

At first I had stuff "Waitress in the Sky" or "Tommy Gets His Tonsils Out" or "Alex Chilton" or whatever in mind, but now that I think about it there's a dark undercurrent to plenty of the songs in the core Replacements canon: "Hold my Life", "Unsatisified", "I Will Dare", "Bastards of Young", "I'll Be You", etc. - not to mention of a handful of slow tempo tracks with obviously dark themes.

I don't normally associate the Replacements with mopey music but in retrospect I'm not sure why. It's kinda all over the place, under a very thin veneer of drunken cynicism.

Exactly. To your point, I'm thinking of 'Swingin' Party,' 'Here Comes a Regular,' even the lovely 'Skyway' is melancholy at heart. Read 'Trouble Boys' by Bob Mehr to find out why.

Clipping here, which definitely made sense.

I listened to Why Don't We _one time_ to see what they sounded like and now Spotify, and this, thinks I stan.

12 hours ago was what, 5am London, Midnight East Coast?

Timing is everything. If a tree falls in the forest and 75% of the potential readership are fast asleep, does it make a sound?

(Numbers are fictional - and possibly eurocentric - for the purpose of conversation. I think the point stands with or without their veracity, it's just easier for form a sentence with values)

In theory you could account for this, right? HN (and reddit, etc.) could score based on how many typical voters have seen it, or just bias directly based on average active times across a week.

Dang sends email to people suggesting they repost it if he thinks it should have gotten more traction, probably an easier system. I could see reddit doing something like that but hn seems pretty basic

> Your taste is so obscure that's so cool I bet you're super interesting..

After it picked the one very popular song I listen to in order to laugh at me for listening to non obscure things, ignoring the piles of weird random crap I normally listen to that nobody ever seems to know. Funny but not very accurate.

Overall all I got a good laugh though.

I think the microtonal stuff I listen to is probably too unpopular to even have reviews to go on.

It picked out Brendan Byrnes as one of my obscure artists, though he is probably one of the more popular microtonal artists. It was able to pick out my own band as one of my "obscure artists", so I don't think "too unpopular" is a thing for it, haha.


you ok?

[lol, jk.] [yeah, why?]

"You are 11% basic. Red Moon Architect and Ahab? Where do you even find this shit?."

Exactly what happened to me also

Curious, I learned something.

> You're stuck in the early 2010s. For you, music's been all downhill since Orden Ogan made The Things We Believe In.

Now I wonder if this is because my account is old, or if I really listen to that much old stuff, or if my favorite bands just don't release that much. Weird.

But I'm entirely fine with the other assments:

> Your spotify was renaissance-faire -too-much-power metal bad.

> Thank your obsessions with medieval rock and power metal (e.g., Sabaton and Dreamtale) for that.

> Based on your listening habits, I can also tell you your spotify was...

> hail-satan bad

> You are 0% basic. Ewigheim and Winterstorm? Where do you even find this shit?.

And I think winterstorm and ewigheim aren't even the most obscure bands.

Oh but there is a new stribog album. nice.

>Where do you even find this shit?.

I don't follow the recomendations, I click "Fans also listen to". Here is the rest of my assesment.

> Your spotify was moonshine-renaissance-faire bad.

> Thank your obsessions with Johnny Cash and medieval folk for that.

> Based on your listening habits, I can also tell you your spotify was...

> napping-in-your-mancave bad

> Here's what else is going on in your aural trash fire:

You listen to these tracks too much:

    Enter The Ninja by Die Antwoord
    Napalm Sunshine by Lying for Friends
    The Lachrymal Sleep by Doom:VS
    Thousands Of Bees by In Gowan Ring
    The Seer and the Seen by In Gowan Ring
You stan these artists to an uncomfortable extent:

    VNV Nation
You are 1% basic. Trepaneringsritualen and Svartidaudi? Where do you even find this shit?.

You're stuck in the early 2010s. You only listen to Obama-era jams like Ensigns Of Victory by Winterfylleth and Streamline by VNV Nation.

Analysis completed in 4.012 exhausting seconds.

Thanks for letting me see your music I guess.

Shutting down.

Try having kids:

> You listen to these tracks too much:

> - Rock-A-Bye Your Bear by The Wiggles

Which, to be fair, is my second favourite Wiggle song!

Oh, I have kids. I just give them their own accounts

Heheh love winterfylleth, you're not alone!

> You're stuck in the early 2010s. For you, music's been all downhill since Kalax made Time Lapse.

I got the same thing with a different song. What's funny about this one is it's from 2017.

Yeah I don't see what's wrong with that.

It's complaining about me listening to Eluveitie and Ost+Front too much (and Metalocalypse and Celldweller).

I'm sorry that I went searching for other german industrial metal bands when rammstein took forever to put out another album.

Everyone got the same message about the 2010s it seems. Don’t fail the Turing test. Or the reverse Turing test where an AI makes you question your own humanity. :)

> Now I wonder if this is because my account is old

I made my account this yeah, and i got the same "you're stuck in the early 2010s". I guess I just listen to a lot of music from then.

Interestingly most of the examples it gave me for being stuck in the early 2010s were from the second half of the decade.

That reads like high praise to me.

You are 11% basic. Red Moon Architect and Ahab? Where do you even find this shit?.

this is about how mine went. It's very concerned about where I found Necropanther and Wasteland Riders

> You are 0% basic. Have you considered there's a reason nobody listens to Mistabishi?.

> > Your spotify was ira-glass-in-your-airpods-manic-pixie-dream-girl bad.

Made me laugh!

i mean there was him loosing all those bookings after going on racist rants on facebook. i think he used to get banned from dogs on acid a lot for being a dumb asshole too https://mixmag.net/read/drum-n-bass-producer-mistabishi-has-...

*Adds Mistabishi to playlist

> You are 3% basic. MYST and Project One? Where do you even find this shit?.

Hilarious, especially since these are pretty much mainstream artists of my (not so small) niche. I'd love to see how it comes up with those.

> This same link was posted 11 hours ago https://news.ycombinator.com/item?id=25513886 here and is not at all getting the same popularity.

HN allows reposts of an article when earlier submissions haven't gotten attention yet. This is our way of mitigating the randomness of what gets traction off /newest.

It makes for a something of a lottery in terms of which submission ends up with all the attention, but it's on our list to implement some kind of karma sharing across multiple submissions. It's been on our list for years, but we'll get to it eventually!



> You are 5% basic. Talpa and Villagers of Ioannina City? Where do you even find this shit?.

> You're stuck in the early 2010s. You must have peaked right around Ratatat's Magnifique.

The last line is definitely something I will mention to my therapist ... :-)

Still, it validated my hipster taste, even though it made fun of it. Suck it bot, I'm cool and your programmers aren't.

Gosh I love Ratatat’s Magnifique.

It's about time for their next album. Got any recommendation? So far, I've found Ratatat are in their own league, nobody comes close to that very specific feeling and genre.

Well, unfortunately no. Ratatat is my lucky find. Heard their 'Gettysburg' while watching some random gaming clip that feature their music! and they're not my usual genre I've listen to. So Google is your best friend...

Other of my fav instrumental band is CHON, but they're more of math rock and not really the same.

Mine was wonderfully vicious too, though I am noticing we got some of the same synonyms (I was at 17%): "masters-in-creative-writing" instead of "please-read-my-manuscript" and I would much have preferred "walmart-hawaiian-shirt" over "tommy-bahama". That stung.

Hilarious, yes.

> You are 15% basic. Yeah, you've got some obscure artists like VanKatoen, but your real top ones are ultra-mainstream like The Weeknd..

> Based on your listening habits, I can also tell you your spotify was...

> dad-rock bad

> neverland-ranch bad

> nineties-dorm-room bad

> studded-leather-jacket bad

> wannabe-tenor-but-probably-a-bass bad

As a rock-listening dad who spent most of the 90s living in a dorm room, I felt spied on.

This was so much fun.

> You are 5% basic. Old Man Gloom and Distorted Harmony? Where do you even find this shit?.

I feel called out, but also kinda proud.

> You're stuck in the early 2010s. You must have peaked right around The Dear Hunter's Act Iv: Rebirth in Reprise.

Early 2010s were amazing era for Prog metal alright.

Ha! I also got 26% score and it was ridiculing me for listening to Ed Alleyne-Johnson [1] and David Hicken [2] even though they are not even mainstream performers :)

[1] https://www.youtube.com/watch?v=XJwAZMq08xA

[2] https://www.youtube.com/watch?v=rHtdqGy3g5Q or https://www.youtube.com/watch?v=YftAYq1AHiE if you are in a more upbeat mood

>Anyone got ideas?

The Act of HN submission. But broadly speaking US working time zone get a much higher chance of submission being upvoted.

6% basic. AI hates me. Apparently, Your spotify was overdramatic-instrumental-yoga-studio-heavy-eyeliner-post-punk bad. And this was just the beginning of the insults.

>You're stuck in the early 2010s.

oof, I'm not even angry lol

The algorithm is more basic than the music it passes judgement on so take the verbiage with a grain of punk-vinyl-you-never-heard-before.

i'm 1% basic, i'm clearly the coolest in da room

I'm guessing 1% is actually common. Mine:

"You are 1% basic. Oh wow Penelope Trappes and Titi Robin! Your taste is so obscure that's so cool I bet you're super interesting..

You're too trendy for your own good. You only listen to music made in the last year."

Another 1%:

"You are 1% basic. Have you considered there's a reason nobody listens to DJ Marsta?.

You're stuck in the early 2010s. You must have peaked right around Mez's Tyrone EP."

I got 0%!

But you ARE cool for listening to Kapitan Korsakov!

It rated me 6% basic... what's sad is that I'm now proud of that. I've resisted being called hipster, maybe inappropriately.

My spotify was 'heavy-eyeliner-post-punk-boomer-relaxation-dads-still-cool bad'

Huh, I wonder if this is just a mad lib that is being used to gather information about people?

It's sayin stuff like "You know music existed before 2019, right?", but also criticizing me for liking Stan Rogers, who died in 1983, Jay Z, and Trevor Hall, who seemed to peak in the early 2000s.

Maybe I just haven't finished my coffee yet, but I was kindof hoping that this would actually use AI to figure out the musical variety of my listening history and suggest more people to listen to or something like that. Seems like it was just a bunch of mad libbed insults though.

Yeah some of it seemed tailored and accurate, specifically the insults at the end (They pegged my music perfectly with "stoner-frat-bro bad"), but most of it was just pulling most played artists and songs and making vague jokes.

I had the same feeling. It made a spot on joke about how much AC/DC I listen to followed by a funny and relevant comment about too much Slipknot. But then it followed it up by calling me obsessed with two bands I can't even remember listening to in recent years. One of which I never really listened much to.

For me also, the big punchline revolved around a song from the Dirty Dancing movie that I had to actually look up to find out if I've ever heard before. Now I'm a sucker for 80s synth-pop, don't get me wrong, but that specific song I'm 99.9% sure I've never actually played on Spotify. So it kinda seems like they've hard coded some insults with pop culture cross-references. The "AI" be like

  select case (person%likes)
    case ('80s-synth-pop')
    print *, 80s_pop_insult_char
    case ('single-specific-artist')
    print *, person%likes_artist//'-stan'
    ! (...)
  end select

Yeah, it called out Todd Barry as one of the most listened to things, and I haven't listened to comedy on Spotify since I started listening to a weekly stable of podcasts like six years ago.

i did not understand the need for the hostile language.

i regret letting this access my account. i would recommend others to not give this tool access, it is not worth it.

Same. I gave it access to my data, it spat out a bunch of vague insults that just... didn’t mean anything.

It called me a “manic pixie dream girl” - I’m male.

It said I was “local-talk-radio-bumper-sticker-bitch“, which isn’t even an adjective.

I revoked access afterwards but like you I felt a bit conned.

> You know music existed before 2019, right?

It gave me the line about being stuck in 2010. Just an insult machine. It would actually be useful if it gave me songs it thought were more "cool" based on my current lists. Still, I'm getting a perspective by reading other people's amusement with the comments here on HN.

Anyone that wants to revoke the access this (or any previously used) app had to your spotify account, you can revoke access here: https://www.spotify.com/uk/account/apps/

Yeah, the absolute lack of any privacy statements makes me wary. They get access to my entire music history and personal info, and just say that they won't post anything? Pass.

To be clear, it looks like the oauth scope they're requesting only allows read-only access to your account. So it's not like they could actually post on your behalf without additional express permissions, barring a Spotify API bug (as far as I understand).

That said of course I agree they should have an explicit privacy policy regardless.

The privacy policy is here: https://pudding.cool/privacy/

> The Pudding collects your data so that we can:

> Personalize stories > Create content for stories from reader-generated inputs

Is this clear enough to forbid them from long term storage and passing the data on?

Only if it is law; and even then, tough to police.

As someone who listens to spotify all day while programming, here's my secret to success:

I create a playlist to fit a mood (for example, ambient electronic music with a good beat and no words), and I seed it with at least 10 songs.

Then, I go down to the "Recommended Songs" section and I just play whatever it recommends.

If I don't like a song, I hit the skip button on my keyboard. The vast majority of what it recommends and plays is tolerable but not great.

Every now and then (maybe one out of every 20-30 songs) I will discover a shiny new gem of a song that really resonates with me. So I drag that song up and add it to my slowly growing playlist.

This has the effect of subtly changing the recommendation algorithm for that playlist over time.

Eventually I end up with a playlist with about 100 songs on it, and recommendations that at least roughly align with the mood.

I tend to get bored of songs pretty quickly though, so when a playlist gets too stale, I will prune the old and tired songs from it.

This approach has worked pretty well for me so far, but I really wish that there was a way to explicitly tell Spotify that I "dislike" a song and to never recommend it to me again.

I'm dumbfounded people still put up with this mess on Spotify. Pandora's discovery, recommendations, and shuffle algorithms are objectively superior to Spotify - yet, everyone still seems to use Spotify none-the-less.

Discovery, recommendations and shuffle are at the heart of what made Pandora a thing... it's literally what Pandora was built on back when it was purely a seeded radio of sorts. Whereas Spotify more-or-less bolted on these features later in the product life... and it shows.

The "shuffle" alone on Spotify was enough to send me running. Have a playlist with 200 songs from 80 different artists? Well, we'll just play this one artist - the same album even! - back to back to back until you're annoyed enough to manually select the next song!

I understood people flocking to Spotify back when it was the only game in town that allowed you to play specific songs, make playlists, and "download" songs for offline (airplane) playback. But now Pandora does all this too - and has for the past several years. The two services are even priced about the same...

So, if Spotify is so bad, as would seem evidenced by all these HN threads that pop up from time-to-time, why do people continue to use it?

Speaking for myself? Content. Philip Glass has 27 albums on Pandora, 105 on Spotify. Nico Mulhy? 7 vs 14. Maarja Nuut. Only 1 album on Pandora. For her, Pandora makes no recommendations for others to listen to, Spotify lists 20. Pandora has no bio, Spotify does. Third Coast Percussion? 8 vs 10. I'm just randomly clicking in my library and doing a comparison, they aren't artfully chosen examples.

I recognize this isn't mainstream listening, and for all I know the ratios are reversed for more popular music.

I found the Pandora choices of songs annoying. I love Philip Glass; I don't want to listen to a bunch of other people playing triplets against eighths. It's derivative and boring in comparison. I do want to listen to his contemporaries that compose in different ways. Or let's take Nuut again. I came across her recently while surveying contemporary Estonian composers. Pandora doesn't let me do that so far as I can tell. I'm interested in what is going on there compositionally over the last 2 decades, I am not looking for another person that writes something that is very 'similar" (according to the Pandora algorithms).

Finally, I never use shuffle. So much of what i listen to are

I understand my use case is not the only one, but neither is yours! People select different apps because they provide what the person wants.

If Glass is your thing, I highly recommend Idagio. It is explicitly focused on Classical, and its filtering is focused on what you'd want out of a Classical-focused search.

For instance, with Glass - there are 171 albums on Idagio. And it breaks them down with works, instrumentation, ensembles, soloists, genres (piano solo, chamber, secular vocal. etc.), conductors, and recording date.

If you want to compare the 22 piano solo albums that Jeroen van Veen has recorded of Glass's music - easy to pull all of them up quickly.

Their encoding is also vastly superior to Spotify or (yikes) YouTube Music. If you have a decent signal-chain for listening, high-quality components, and enjoy Classical... Idagio is head-and-shoulders above the rest.

You could check out YouTube Music (edit: changed from "music"), which has a long list of Philip Glass titles that may be up there with Spotify.

I gave up on Spotify when they removed the widget functionality on Android. I know they returned it eventually, but not without complete uproar. It burned myself as a decade old (family) premium user though; I used to love them, but that complete disregard completely changed that.

Now, only a 3 month minimum (6 would help) completely free retrial would get me back. Thus far, none has been received.

(On the flip side I absolutely love YouTube Red or Premium or whatever they call it these days.)

Youtube's organization of music is horrible. It clearly wasn't made for listening to music.

I assume they were talking about Youtube Music (https://music.youtube.com/), which is made for listening to music.

I used Google Music until they discontinued it in favor of Youtube Music, and I ditched that crap immediately, the UI was horrible, the recommendations were BAD (Google Music was somewhat good), and I lost about half my playlist items as "not available". It is so bad, that after years of Google Music, I just ditched my playlists (what little there was) and moved to Spotify.

I find the YT Music recs actually pretty good, though the "recommending a song after I thumbed down" behavior blows my mind in its idiocy. Also, the fact that they don't distinguish between me liking a video on Youtube and liking a song in YT Music is also frustrating.

The UI is the worst I've seen on any music service or player, but I'd found quite a bit of music I like through their recommendations.

I reported the bug about recommending songs that I had downvoted. Apparently I had more thank 5k downvotes so it didn't consider them all in the filter. The "fix" was bumping the limit.

I used to pay for Google Music and transitioned to YouTube Music but eventually gave up because it has so many issues. I also tried Spotify but didn't like it much so now I don't know what I will use.

I just tried listening to some uploaded music last night. Except the Artists page seems to be in random order vs alphabetical and there is no apparent sort option. I'm not sure how that decision played out.

And the idea of integration of music and music videos is brilliant. I love listening to music with video on my TV in the morning.

The implementation and mixing music videos with other video content in my Youtube account is horrible on the other way.

Yes, thanks.

While Youtube Music is a bag of some great ideas and implementation that went horribly wrong I have started using it over Spotify recently.

Youtube Music implementation of Chromecast is spotless while Spotify every so often interupts with "Your spotify account is used on another device" (Not true just checking my phone). In general Youtube is also tolerant of using two devices with the same account for a few moments.

And no podcasts in the music app... Hurah...

I don't have anything to add but to say I'm glad someone else has discovered the beauty of coding to Glass's works.

Been listening to Glass's excellent score for Tales from the Loop for a few weeks now. It's my current goto album for coding.

IME Spotify's recommendations, while not perfect, are way, way better than Pandora's. Spotify is comfortable letting you explore niches; Pandora, when I used it, always seemed to try to hill-climb out of whatever niche you were listening to as fast as it could into whatever music was the most broadly popular at the moment.

Spotify won the same way Dropbox and Slack did. They didn't invent anything magical, they just had a client for every platform that worked when people tried it. Like web apps? Here you go. Linux nerd? Here is a deb repo. Have a smartphone? Cool, we don't care what kind.

Even now I googled "Pandora windows app", clicked a few times, and ended up on some Microsoft Store page. Interestingly enough if I add "mac" to my search terms I just get a download...

What's wrong with Pandora in the web browser?

That is like asking "What is wrong with driving across town to pick up your Amazon purchases?"

If you are in a highly competitive space that is effectively winner take all - you go where your customers want to be, not where your developers think they should be.

I prefer Pandora tab on my browser.

The browsers RAM usage and the challenge of finding it among the often many tabs and windows come to mind.

Eh, does the RAM usage of a single browser window even matter on modern systems?

Anecdata, but I open Pandora in it's own window, then minimize it. This makes it pretty easy to find if you have a hundred+ tabs open at once like I often find myself with.

I currently have 44 tabs open, including Pandora in it's own window, and Chrome is only using ~2GB of RAM - perfectly acceptable to me.

If it's like YTM in browser then no control with keyboard keys.

Can you control pandora in the web browser with your keyboard? can you jump to pandora in the web browser with a single keyboard shortcut?

Nothing is wrong with Pandora in the web browser, other people just like to install spyware and root kits so let them.

> Discovery, recommendations and shuffle are at the heart of what made Pandora a thing

That's exactly what I'm not looking for; I'm perfectly fine with making my own music choices. I know what I like, have an extensive "library" on Spotify, and play from just that.

The biggest thing I'm missing from Spotify (or indeed, most players, except the one I wrote myself years ago) is the ability to say "play random album from my collection". Play random track is too biased towards albums with 60 tracks vs. albums with just one track.

> "play random album from my collection"

Quod Libet (https://quodlibet.readthedocs.io/en/latest/) lets you do exactly that.

Can you use that on Android?

No. Quod Libet is GTK-based.

Pandora allows you to shuffle playlists. Might allow you to sort of achieve what you're after.

foobar2k wins again ;)

Does it have this ability out of the box? I don't remember it being there years ago (I looked for it), but that was 3+ years ago.

It's still the best music player I know of though, even when running with Wine on Linux.

AFAIK, I didn't have to add any plugins to achieve that feature. I believe you can set the order to be Shuffle Tracks or Shuffle Albums out of the box, but I might be misremembering if I needed to do anything for that to be there.

see: https://i.imgur.com/zL0oZsx.png

You can customize this feature in Advanced Preferences. Then, go to Playback -> Shuffle -> Album grouping pattern.

See: https://hydrogenaud.io/index.php?topic=113035.0

How great is fb2k for music discovery? Also their macOS player is kind of...bad. I'd recommend Cog for local files.

So does cmus :-)

mostly the same for me, but I would love to be able to browse record labels on spotify; it does not have the concept of a record label (!) though.

It does. At least some time ago on desktop version every album had a copyright (record label) listed on the bottom. Clicking it would search you all of the stuff released under that label.

None of them are links: https://i.imgur.com/IikFCNt.png

> Pandora is only available in the U.S. right now – but we are working on bringing our music service to other parts of the world.

I guess Spotify is still the only game in town :(

I didn't realize Pandora was region locked - what a bummer!

The rest of the comment still stands for all US based users, however.

I last used Pandora's radio in Canada around 2005. I'm questioning what it is they're "working on" for so long.

Probably license agreements with the labels... but that's pure speculation. Spotify seems to have solved that problem by throwing tons of cash at them.

You can see here[1] that Spotify has made agreements with specific labels for international rights. What a thing... only licensed to stream a song in some select set of countries... absurd in 2020/2021 and being an internet based service - but that's record labels for you.

[1] https://www.theverge.com/2020/4/1/21203324/spotify-warner-ag...

Spotify is too, just not as drastically. I opened an account in Prague and found when I got to the US that it wouldn't play anything. Curious how they knew I wasn't in CZ anymore, yet instead of just letting me play the US version, it wouldn't let me play anything.

This isn't true, may be got to do with specific set of songs? I've an US account, which I use in UK, India and Japan.

Spotify is great, people on HN like to complain.

Aside from having nearly any artist's catalogue available for almost free, their recommendations have helped me find a lot of great songs. I don't personally listen to the curated playlists but many of my friends love them.

Spotify objectively sucks compared to plenty of other services. Is it possible you haven't tried those other services? Having access to lots of music is certainly better than not but we've had services that do that since the early 2000s.

I subscribed to Rhapsody in 2004. It gave me access to "nearly any artist's catalogue available for almost free". As 2 examples of things it did better, instead of using a shitty recommendation system that when I pick some electro swing band it recommends death metal, instead for every artist it showed their lists of who influenced them and who they influenced. I found tons of great bands that way. Way more than i've found on Spotify's crappy system

And that's just one example. The now dead Google Play Music's recommendation system actually worked where Spotify's does not.

>Spotify objectively sucks compared to plenty of other services. Is it possible you haven't tried those other services?

It's just your personal experience that is different. Same for the other guy, same for me.

I've been on a paid accounts for Apple Music, Google Music, Yandex Music and even Youtube's adhoc solution for music. At least 1 year each (almost for entire time since the launch of Yandex music).

Spotify beats them all. Both recommendations and content (though we have to agree that content part can vary based on your location). Never had a chance to try Rhapsody.

>that when I pick some electro swing band it recommends death metal

That's how Google Music worked for me though, not Spotify.

Yandex had best UI but worst recommendation system (and lack of content).

Apple's recommendations are about as bad as Google's. Better UI but again - did not have music I wanted.

> Spotify objectively sucks compared to plenty of other services.

Is this a new thing?

"Everyone else's opinion * objectively * sucks".

Just because you write "objectively", doesn't make it so.

Opinions are, by definition, subjective

Haha that's funny, I guess I haven't experienced any of recommendations personally. Usually they're very good in my experience.

Pandora has good algorithms, but as a product is objectively terrible.

It is literally impossible for you to just sort the list of stations you have created. If you have 120 stations, you have to s l o w l y scroll through them in random order until you find the one you want. And even though there are thumbnails next to them (presumably to make them easier to identify), the thumbnails change. There's a hundred different usability problems, and as far as I can tell, all of them minor features that could have been implemented years ago.

If you don't pay for the premium version, it's a nightmare of pop-ups to prompt you to pay, to the point that you develop reflexes to pass the prompts. Then there's the actual ads, which sometimes seem to disappear for a while, only to barrage you intermittently for a half hour.

The whole thing annoyed me so much, and I complained on their support channels so often, that I dropped them and just started buying music. And of course I then had to pirate the same music I purchased because there's no way to just buy some MP3s so you can move the music around.

Imagine thinking recomendations can be objectively better than others. Personally I don't care for pandora recommendations one bit and love my spotify mixes.

Why can't they?

It can easily be proven they can, posit three systems:

1. Spotify

2. Pandora

3. No matter what, rickroll you on the next song

Cleary 3 is far superior to the other two! ;)

So if something can be objectively WORSE, there must be the possibility of something being objectively better.

Pandora isn't available outside the US, so it doesn't really matter how good it is compared to Spotify I can't use it anyway.

It doesn't matter how good the tech is if business/legal can't keep up with the competition.

For one, it's US-only. And the recommendations from Spotify so far are good enough for my needs and my whole family is on it so the work required to migrate from one platform to another isn't null so unless there's a major irritant I don't plan on switching.

> So, if Spotify is so bad, as would seem evidenced by all these HN threads that pop up from time-to-time, why do people continue to use it?

Aside from amount of content (which wasn't an issue earlier), I suspect the primary reason is that Pandora isn't available in Europe. I loved it so much (>5 years ago?) that I used as VPN, but this didn't quite work on my phone and ended up being too much of a bother.

Spotify's big advantage, like most monopoly-focused 'startups' is their monopoly. Not actually providing the best to their customers.

I can use Spotify without a VPN, it has a lot of stuff Pandora doesn't have, and I can share stuff with friends because everyone else also uses Spotify. It's difficult to compete with that.

Oh how the world would have been different if Rdio hadn't been acquired and killed by Pandora. It's recommendations were perfect.

Rdio went bankrupt, and sold assets to Pandora as part of that process. It wasn’t bought and killed.

Fuck they were the ones killed it. I don’t remember how the recommendations compared but Rdio’s UX was miles ahead. Pandora really should have abandoned their frontend and went all in on Rdio.

When I try to open pandora.com from Australia I get an empty page with the sentence "Whoops! You weren't supposed to end up here."[1]

[1] https://ibb.co/YNxhbkH

Pandora restricted access for me when I was living in the UK years a go and I could no longer access it. (I don’t know if this is the case) Spotify won for me by default, as I can actually access it.

> Pandora's discovery, recommendations, and shuffle algorithms are objectively superior

IIRC, these algos were part of a project called ~Muse which later forked, and there was a schism between the algos and clickstream data or the personal data and anonymized cohort data. Will have to review / re-search.

The trove of our searches, listens, and skips Pandora has / had did a good job informing recommendations and was an effective moat for my use.

A big part of the algos was the "Music Genome Project"[1] as well. This allowed their recommendations to be about the music itself instead of just what other people also listened to, leading to some amazing music recommendations/discoveries.

[1] https://en.wikipedia.org/wiki/Music_Genome_Project

Thank you this helps -- the "Music Genome Project" was part of this fork & schism. Pandora says little about it[0], and the open portions[1] were last updated in 2014.

At least someone's still at it, and all this helps me find my notes from that time.


[0]http://www.pandora.com/about/mgp [1]https://github.com/omgDB

My guessing is a lot of it went proprietary once competition started popping up trying to also make good music recommendations. "Secret sauce" and all...

I use the "radio station" features on both because I'm not a curator. It does seems like Pandora's version is a bit better but not drastically so. Then again I'm not a music monster, I listen to what the algorithms tell me until they push me too far :)

I'm going to check out Pandora, but out of the large ones (Apple Music, Deezer) Spotify has by far the best algorithmic recommendations.

Pandora decided a few years back to block international users. I don’t know if they’ve changed this - I haven’t bothered to go back.

Did they ever allow for international users?

Yeah, way back when (early 2000s?) it was a streaming personalised radio and it worked fine, at least in the UK.

Pandora is also US only unless you use a VPN.

I bit the bullet and just subscribe to both. Pandora for discovery and radio and Spotify for specific listening.

> Discovery, recommendations and shuffle are at the heart of what made Pandora a thing... it's literally what Pandora was built on back when it was purely a seeded radio of sorts. Whereas Spotify more-or-less bolted on these features later in the product life... and it shows.

Pandora wants to get rid of those features. All their advertising now is "if you pay for a premium account, you can play whatever music you want, on demand".

This reminds me really strongly of how eBay decided that (1) customers prefer to buy things at a fixed price [completely correct]; so that (2) they should try to prevent people from listing ordinary auctions, because customers don't like those. [No!] Instead, the goal was to just be the place where people would go if they wanted to buy something online.

Except of course that that completely undermines the concept of eBay. They were an established auction site that intentionally drove auctions away. And then they complained that people preferred to buy things from Amazon.

In both cases, it looks like a popular company with an established brand sees a rival that's even more popular, and concludes that the only path forward must be to implement features the rival offers while discontinuing or discouraging any features where they're doing better than the rival is. How does this make sense?

Pandoras was better but is now much much worse. Spotifys recommendations are the fold standard.

I did search for 5 random bands I listen to on Pandora, found a total of 0 albums and 0 songs.

Pandora's suggestion algorithm is much better than Spotify's. What you describe is what I used to do with Pandora. I migrated to Spotify several years ago.

Spotify generates some decent genre based playlists based on my listening history. Not as good as Pandora was.

Lately with the pandemic, I've created a super-chill, ambient play list (I started with Moby's Long Ambients and expanded from there) to relax before bedtime. This has been really helpful for unwinding in these stressful times.

However, an unintended consequence is that Spotify now thinks I listen to soothing ambient more than 50% of the time. My year in music is all ambient and my suggested play lists are all wishy-washy now. Apparently there is no "ignore this playlist when tracking usage." You can only manually disable tracking under "Settings > Social" and it times out after every session.

I haven't used Pandora in a decade, but when I did, it strongly preferred to play mainstream artists, and within artists, only a handful of specific tracks. They weren't bad choices, but I got pretty bored of them.

Spotify takes a lot more risks and I think its highly preferable.

A decade is a long, long time. The catalog is definitely on-par with any service out there today.

I would believe they've improved over time. But my comment wasn't that pandora had a poor catalog. It was that they wanted to play mainstream songs instead. Perhaps not even mainstream, but just very specific songs. They had a lot of artist's whole discographies but only a subset would ever come on. I can still remember a few of them for me:

royksopp - this must be it

trapt - headstrong

ladytron - destroy everything you touch

Some sort of remix of the hand that feeds by Nine Inch Nails (I think it was the photek remix or something)

I got these songs a lot on many stations, and nothing else by those artists ever, haha.

I don't hold a grudge from 10 years ago though

I loved Pandora back when they were experimenting, but then they stopped providing service where I live and Spotify is the only game in town :(

Yeah, I hate this too. I use citypop and synthwave when exercising, and they then “pollute” everything else.

My "philosophy" with listening to music while programming has been that if I can tell I like the song, then i'm probably focusing too much on it rather than the task at hand.

+1 I do something similar. As a general rule, I prefer to listen to songs I don't already know while working. If I'm humming or singing along, I'm doing it wrong.

I've actually discovered some of my favorite bands/songs this way.

Programming along, with background music playing - when suddenly I become "aware" of the song playing and how good it is. That's when I give it the "Thumbs Up" in Pandora, then go back to work. Might happen once a session if I'm lucky.

I do this too. It works well.

In addition I have several playlists, as a song might suit the current playlist, and another song might fit playlist B instead. So any playlist might be seeding several other playlists at a time.

Sometimes I go through many, many recommendations, listen to a song in short pieces quickly by skipping most of it. It sounds weird but that way I can pretty accurately tell if the song is shit or not (read: if I end up liking it). If the song passes my filter, I put it to some playlist. I just do this quickly and finally go back and listen to the new ones.

As for "dislike", there isn't one per song but there is a "Don't play this" per artist. I sometimes play children's songs or pop as requested, and after some time the recommendations include music I truly never want to hear again in any recommendations (Marcus & Martinus I'm looking at you).

Also, it feels like only the first songs in a playlist actually matter for the recommendations. I don't know if this is really the case. I'd hope the recommendation engine took a random sampling of the entire playlist and recommended based on those.

Edit: there's a "hide" for a track.

> I create a playlist to fit a mood (for example, ambient electronic music with a good beat and no words) ... what it recommends and plays is tolerable but not great. ... (maybe one out of every 20-30 songs) I will discover a shiny new gem of a song that really resonates with me.

SoundCloud Weekly[0] recommendations have been solid for months specifically with ambient: ~85% Good tracks with ~30% Resonating enough to Favorite. My other contexts / use cases include listening by Album and Artist, Genre, Mood- and Mind-Setting, and Discovery.

Spotify works well for Albums and Artists. HypeMachine is my go-to for Discovery and Genre. SoundCloud and BandCamp work well for diving into artists and labels discovered. Playlists and Favorites across these are best for Mood- and Mind-Setting, as are "special purpose" options like Brain.fm and Ragya.

[0]https://soundcloud.com/discover/sets/weekly::[username] [1]https://www.ragya.com/

> Every now and then (maybe one out of every 20-30 songs) I will discover a shiny new gem of a song that really resonates with me. So I drag that song up and add it to my slowly growing playlist.

I take a similar approach, but I've found that the recommendation algorithm is pretty bad, so as an extra step I click on the artist and spend a bit of time listening to their other songs (and add those too, in the hopes it'll affect the algorithm).

> This approach has worked pretty well for me so far, but I really wish that there was a way to explicitly tell Spotify that I "dislike" a song and to never recommend it to me again.

That's one of my main issues with the 'recommended' algorithm. 90% of it is stuff I already have in other playlists, or stuff I keep skipping. I really wish the algorithm did a bit more with all the data it has, but I imagine they optimize for sales (to us or promoting artists/labels) over our actual desires.

This is what I do, except using a general playlist. At ~ 300 songs almost all recommendations I didn't like. At current ~ 400 songs all recommendations I don't like and Spotify gave up with a passive-aggressive message sort of "I give up, add more songs if you want better recommendations". (Note that I have a wide variety of styles in the playlist; too lazy to break up in genres and also I kind of like the "surprise" factor)

The secret to my success to spotify is be a free user with adblock. On desktop browsers, there's no forced shuffle / variance. It just plays the album from start to finish, and, of course, no ads. Pick something manually you like. When you're done, pick a an artist from their list of similar artists. Play entire albums at a time. If you're not immediately dissatisfied, you'll probably feel good about the next half hour of music.

I do the same thing now, but the recommended section leaves a lot to be desired on mobile.

There's no way to prevent songs from getting suggested (other than maybe blocking a song entirely?). My recommendeds don't seem to get updated until the playlist has changed dramatically. And repeat songs get added to the section even if I haven't added it the umpteen times it was suggested prior.

Is the experience different on desktop or am I doing something wrong?

Anyone here remember turntable.fm? I do. Sad. JQBX.fm isn't the same either. That was a great way to get a higher "hit" ratio though.

I miss it so much. Nothing else has come close. I'm surprised Spotify hasn't added a feature like it.

>I really wish that there was a way to explicitly tell Spotify that I "dislike" a song and to never recommend it to me again.

You can at least for radio playlists. You can specifically say to never play this song or artist again. They used to let you thumb up or down on tracks which I think was a much easier UI.

Skipping songs doesn't seem to do enough for songs in your playlists so you have to go through the whole right click -> never play again workflow. I really wish I could just downvote individual songs as they play and tailor it that way since sometimes I'm just tired of a particular song, but maybe I like the song or the artist and don't want to click on 'never hear this again'. It just seems to repeat certain songs way too much based on your listening history.

My solution to me getting tired of tracks in my work playlist was to keep adding more. It's currently at 2885 songs, or 5.2 days of music. It's mostly video game music, since a lot of video game music is written to not be too distracting.

Though maybe I should say the majority of the playlist consists of songs I've imported into iTunes. Apple does a decent job at letting you seamlessly integrate your personal library with stuff from their streaming catalog.

I mostly use this approach as well but my experience with multiple services is that I eventually (and sooner, rather than later) end up hearing same 100 or so recommended songs. Maybe I'm too picky, but I just don't have great luck with recommendation engines anywhere I go.

> here's my secret to success:

> effectively, I manually make my own playlist.

yes, but my method at least introduces me to music that I would have otherwise been exposed to.

I recommend the app Mick Tagger for the drag process. Makes it easier and keeps you in the zone with programming instead of having to fiddle with a song and playlist.

Thank you for saying this. I've been trying to use Spotify for a year now and it leaves so much to be desired on process. You're doing gods work

I've had trouble with this. It seems no matter what I play, Spotify just wants to recommend three songs by Pavement on repeat.

> If I don't like a song, I hit the skip button on my keyboard.

Which button is that?

Share an electronica playlist and prove it ;).

Do share your playlist , if its public

Music preferences are subjective, because familiarity is a huge part of appreciation. Sharing my particular fingerprint of music appreciation with other people doesn't do me any good, because they will only criticize my preferences as they are not their own.

I have to give huge applause for the quality of writing. I genuinely can't tell how expansive the responses are, which is huge for AI.

It definitely has the UX feel of being a bit creepy on how well it responds.

My friends and I all did it, the punchlines all seem to be the same. I think it feels “creepy” if you attach emotion to songs (I certainly do!) and then listen to them often, but it’s really not all that hard to pick out your top tracks/artists to do that with. Anyway it’s still fun, I liked it

Exactly, I love the wittiness in writing. But wonder if the AI responds similarly to other users.

It looks like the API is overloaded.

With regards to taste, you usually need to find a suitable gateway into an unpopular genre to begin appreciating it. I always hated jazz until I turned 30, when I heard the melodic jazz of Avishai Cohen [1] and Phronesis [2], which led me to Eric Dolphy [3] and ultimately to free jazz outfits like Alexander von Schlippenbach and Peter Brotzmann. Having grown bored of everything else also helps.

I recently cancelled my paid subscription to Spotify after 10 years and I use Youtube a lot now. The Spotify web app is not great and Youtube recommendations are a lot better in my opinion.

[1] https://www.youtube.com/watch?v=D6KYbpDditU

[2] https://www.youtube.com/watch?v=lEw0swsmqYk

[3] https://www.youtube.com/watch?v=ln8naZpOJ0o

Great suggestions. You might know them but I want to add Nik Bärtsch's Ronin[1], The Bad Plus[2], Nils Petter Molvær[3] and maybe some Ibrahim Maalouf[4].

[1] https://www.youtube.com/watch?v=i8eL95fIQbk

[2] https://www.youtube.com/watch?v=iNI-2i6t7Zs

[3] https://www.youtube.com/watch?v=edT6QGBmkrc

[4] https://www.youtube.com/watch?v=wpg8jBFaj3c

The Bad Plus (version 1) is such an unreasonable concentration of extraordinary talent, I haven't yet found anything else that comes close.

Essbjörn Svenson Trio, all of their records are great, however Leucocyte is on a whole different plane; it has been my favourite record for the last 15 years. It's however quite moody, Tuesday Wonderland is far more accessible while also being extremely good.

Haven't listened to them in a while, but I also enjoyed listening to Tingvall Trio, though I remember them as being more atmospheric. There's also Vijay Iyer Trio, I specifically remember Accelerando well, but that's a long time ago.

I had a similar Jazz experience. Thanks to Spotify, I ended up going to Bad Plus, Brad Mehldau, Avishai Cohen and similar concerts and started to appreciate free Jazz and the local scene more, too. I found this incredibly enriching, and it has made me try to play the piano again.

I’ve never had a spotify account and whenever I want to listen to a specific song I go to youtube, even if I had already bought the song on bandcamp. It is amazing, sometimes I discover for the first time that the song has an amazing music video, or—better yet—a fan made music video (see e.g. https://www.youtube.com/watch?v=TOeL6PYrU8M).

I can’t speak much for the recommendations, as I am fine with getting my recommendations from the radio and I prefer listening to whole CDs over playlists.

I study the Spotify web app a lot. What is it about the app that you don’t like?

The problem with this service is that most times, I don't chose what I listen to. Spotify does.

I play music all day while working and I'd love to listen to more artists but most times, I'm stuck in the routine between my Release Radar on Fridays, Discovery Weekly on Monday, and Daily Mixes during the rest of the week. I add songs I like to a variety of personal playlists and sometimes I start a "Song Radio" for a song I like. But even then, I just end up getting the same songs Spotify seems to think I liked only because they've already played them to me over and over. Part of me wonders if their algorithm favours songs that costs them less in royalties...

Playlists made by real people seem way better than any Spotify playlist, but they are difficult to discover. I'd love a feature that recommends me playlists from people that share my music tastes. One example, is a playlist called "Sexy Bath" by Nicoleta Tataru.

I have no proofs, but I feel that liking/disliking songs does make difference (even though I'm still struggling to get rid of tons of bad covers and mediocre music in Slav languages)

Liking/disliking makes a difference, but for some reason my likes from 2014 seem to have an identical weight to my likes from today.

I haven't observed that with my account. I think it does change if you spend a bit of effort at letting it know what you like. Like songs you like, hide songs you loathe, skip songs that are meh.

> even though I'm still struggling to get rid of tons of bad covers and mediocre music in Slav languages)

What region is your Spotify set to? It seems to base the recommendations on the region (I live in the Nordics, but used the US version for some time and got recommended a ton of spanish music).

I'm in Israel but originally from Russia and listen to some Russian music and Spotify somehow assume Polish would be just as good for me. Leave alone decent amount of German music due to Rammstein.

I also got some Scandinavian stuff too but it usually fits my taste at least.

I wish the playlists were better but half the time when I search for a playlist to suit a specific mood, It'll have a minimum of 5000 songs on there which means that the overall average quality of the songs is probably going to be pretty low.

You know that the individual who created the playlist in no way hand curated that list, hell they probably haven't even listened to half the songs that they added to their playlist, much less vetted them.

>The problem with this service is that most times, I don't chose what I listen to. Spotify does.

This is one reason I never really clicked with it, or with Pandora, and why I have really enjoyed and used Apple Music. I'm not really interested in algorithmic playlists most of the time. I want what I want. I've been listening to music a long time; I can make connections on my own. ;)

I follow artists I like as a way of adding them to my library and "like" particular songs if I want them more accessible without having to go to the artist page. This is the way I listen to what I want (along with occasionally making playlists).

I actually really enjoy the ability to effortlessly discover new music I never would have found otherwise that Spotify gives. The recommendation algorithm can be weird at times, but I hear liberal use of likes and dislikes helps though YMMV.

As an aside, everyone else I've seen using Spotify tends to strictly build playlists and not really follow any artists. Not really sure what to make of that.

I don't mean to dismiss algorithmic discovery entirely, I should say. It's good if you want it. But mostly I play stuff I've gleaned from elsewhere (read that article, got a suggestion, whatever).

Funny, but I was hoping this might provide some insight into why my Spotify is so bad, even by my own tastes.

Spotify constantly queues and recommends songs to me that are so bad, I can’t even imagine how there could possibly exist any data indicating that any significant sample of listeners has ever enjoyed hearing them. Spotify has 5+ years of my listening history, and orders of magnitude more data from listeners all over the world, and yet every time I set it to recommend anything to me I just sit there pressing “skip” repeatedly until I give up.

I always blamed myself, thought I was just becoming old and curmudgeonly in my 30s. But yesterday I finally discovered the problem isn’t with me. I switched to Apple Music, and it queued up 50 songs I’d never heard, and I enjoyed almost all of them.

I can use my HomePod now, too. I’m really loving Apple Music so far, highly recommend it to anyone who thinks their Spotify account is permanently broken like mine was.

I'm not typically one to buy into conspiracy theories, but I really can't help but wonder if the Spotify algorithm favors songs that have more favorable licensing terms.

They have already announced they intend to let artists buy their way into playlists in the future https://pitchfork.com/thepitch/could-spotifys-new-discovery-...

Two years ago they were already shoving Drake all over paid subscribers' playlists[0]

I've paid for spotify for more than a decade and I definitely wasn't pleased to have every single playlist recommending an artist I don't like, and it was just ridiculous how EVERY[1] playlist had his photo regardless of the genre.

[0] https://exclaim.ca/music/article/drakes_face_was_all_over_sp...

[1] https://twitter.com/DanieleRose22/status/1013845491757568000

I had exactly the same feeling. The daily recommendations contain 80% from albums I've either already downloaded or liked (or auto-liked, which was also a very annoying thing I had for a while). The remaining 20% is recommendations of different songs of the same artists mostly.

With New releases it is different. If I ever play, say, a song someone else sent me to listen to in a different genre, then it screws up my recommendations here for months, sometimes forever.

I think that is the case for every freemium services and social networks: they (must) reward and optimize users for profitability.

I'm surprised that they haven't taken a page out of Netflix's book: sign small artists that sound close enough to what's popular to their own internal record label and push them in playlists?

That doesnt require a conspiracy (really hard), just an imperfect incentive structure (really easy)

Don't wonder. Just be realistic. Did you ever work in an IT business environment? Remember the situations when good taste slowly morphed into business interests? Yeah, that's natural in a capitalist environment. By no chance spotify dodged that bullet. Positivity and goodwill is fine but don't spare anyone from corruption.

Counter anecdote: I've built up playlists full of tracks from good recommendations from Spotify. The genres are relatively niche and electronic (sub-genres of house, techno, uk garage, electronica, etc). The only downside is it gets a bit too eager to recommend the most played few tracks from a given artist, but I've discovered a lot of new artists this way though.

I remember that my suggestions were pretty obscure the last time I tried the spotify algo. But actually it is pretty natural that they suggest niche or less known artists because those have shittier deals and are less expensive to stream. Also, in that position you don't want to promote the top artists for free.

> might provide some insight into why my Spotify is so bad, even by my own tastes.

That's just Spotify. The AI did a better job of figuring out what I like (138 and uplifting trance) than Spotify does (EDM). I blame it in Armin van Buuren: he is an extremely popular EDM DJ and he plays a very wide variety of EDM. If you listen to EDM chances are he has at least a few dozen sets you'd enjoy. That's the problem: he's prolific and he connects every genre of EDM to every other genre of EDM. He's a "super-connector" and I could easily see how this wrecks recommendation algorithms, all roads lead to Armin.

Basically: recommending music is not the same as recommending goods, and I think the same approach is being used.

Good point. I started to notice a lot of the Spotify recommendations sounded like half of an EDM track I might like, mixed with half of something that it should’ve never been mixed with.

Spotify also seems to think I’ll like literally anything with a house kick.

It’s an extremely difficult problem to solve though so they have my sympathy!

Nobody can really explain why they love one song and hate another, and the overlap of the Venn diagram between any two people is usually very small.

I can eat any dish at a restaurant and think it’s not very good, just ok, pretty good, great, or amazing. The same for any movie, TV show, painting, drink, book, article, ... but for a song, I either like it, love it, or hate it so much that I can’t stand it.

Music recommendation algorithms have such a narrow surface area to land on, and when they miss, they go right into a volcano.

Wow, a mention of Armin van Buren! I wonder how much discovery of his work is just happenstance. His track is the first thing on Google and Bing that gets thrust into your face when you search for "this is a test".

That's my default browser bar search term for checking that my interwebs are still wired up, and I've often thought that this was either genius or a really happy accident.

I don't think that's a bad thing per se, if the algorithm correctly identifies that you like EDM. We then come to the micro-genre discussion - how specific are you in your tastes, and how precise is the algorithm required to be?

EDM is an extremely diverse genre. Recommending music because it falls under EDM is like recommending music because guitars are used.

Fans of any genre say this. If you don't like metal then it's all loud guitars and screaming. When you do like metal there is fractal complexity of subgenres all the way down.

Except Phish fans. For them there is only Phish.

Spotify is fantastic for me with everything. My only complaint is their webapp causes random bugs and eats whatever resources it can find.

Just as a counter point.

Same here. Wonderful playlists. Constant exposure to new music. This discussion seemed to get people in the mood for the airing of grievances, so here we are.

Just a couple of months ago Spotify temporarily blocked playlist exporting and there was an uproar. And it turned out that a lot of the uproar was about those Spotify automagically created playlists. Read some sentiments, coupled with some Apple Music comparative opinions, here-


I guess it just turns out that maybe Spotify isn't for all people. Nor is Apple Music. Some discussions just draw out the edges from whatever side and it becomes the narrative.

Taste is a tough nut to crack.

If I remember correctly, they blocked API access for one user because that user was violating the API usage terms, namely exporting Spotify's content (their playlists) to competing services. I actually though that was pretty reasonable of them.

I... what? So you think Spotify was justified to stop people from... copying playlists? ?? what????

The playlists they (Spotify) created? Yes. I think there's an argument they own that data.

The playlists that users created? No.

I apologise if that distinction was not clear enough in my original post.

I wasn't aware that playlists can be copyrighted. And what's the difference between someone mechanically scraping playlists (because, quite frankly, spotify alter their playlists all the time -- I've experienced this with the Soul playlists they have), and someone hiring people via mechanical turk or getting 3 other people together to make text-based lists?

> I wasn't aware that playlists can be copyrighted.

I Am Not A Lawyer, but "a curated collection of units-of-art" seems like a very reasonable thing to protect access to, independently of ownership of the actual units-of-art themselves (overly-generic terminology is intentional, since although we're discussing playlists here, the same argument could be made for, for instance, "a particular framing/hanging of visual arts")

> And what's the difference between someone mechanically scraping playlists [...], and someone hiring people via mechanical turk or getting 3 other people together to make text-based lists?

What type of difference are you interested in?

* Effective difference in terms of output? None. * Spotify's desire to prevent people from doing so? None (they would want to provide both - and justifiably so, IMO, since, again, the playlist _itself_ is their own (algorithm's) creation). * Likelihood of evading prevention? The MTurk solution is less likely to get shut down, for sure. That doesn't mean that you have magically gotcha'd copyright and that corporation will cease trying to protect their IP, it just means that you have found the next step in the arms race.

> I Am Not A Lawyer, but "a curated collection of units-of-art" seems like a very reasonable thing to protect access to

I'm going to need a citation on playlists specifically being protected under copyright. I'm not sure how you can claim that what is effectually a simple plaintext list of things can be copyrighted. What original content there is being put under copyright?

As you already stated, they do not alter the music itself in the way that a rearrangement would, most media players provide fade-in/out functionality so they cannot claim that is unique, and they do not provide any supplimentary content to enhance the experience, at least not in the same way that a book published list, or internet top ten list would ordinarily provide some kind of commentary on the items.

What you're claiming here is effectually that, not even the content, but the mere titles of every single "top 50 foos" list on the internet can be copyrighted. Or that the rearranging of a book's chapters without any change to the contents, can itself fall under copyright

I'm not sure how anyone could think that this is at all a reasonable position to have? It's baffling, to be quite honest.

> I'm going to need a citation on playlists specifically being protected under copyright

And I'm not going to give you one, because, as I said, I Am Not A Lawyer (and even if I was, you're not paying me to be). But, regardless, note that I made no reference to copyright whatsoever - I said "this seems like a very reasonable thing to protect access to". I'm making no reference to law whatsoever - just to what _I_ think is reasonable for a corporation to protect. You are free to disagree.

> they do not provide any supplimentary content to enhance the experience, at least not in the same way that a book published list, or internet top ten list would ordinarily provide some kind of commentary on the items.

The collection _is_ the supplementary content. If a Spotify playlist was a quasi-random collection of tracks from across the entire catalogue, then you'd be right, but they're not - the playlists are curated and specifically chosen to fit some niche (genre, artist-relation, time period, etc.). By virtue of _being in a playlist called_ (e.g.) "1940's Smooth Jazz", the songs are demarcated as being a) relevant to the particular criteria, and b) of a high-enough "quality" (whatever that means) that they have been chosen. So, yes, the plaintext-listing of those song titles _would_ have some value, just like a list of "(only the titles of the) top ten sci-fi novels of 2019" would have some value. I would, for instance, value such a list more-highly if it came from someone whose taste in sci-fi I respect and resonate with.

> What you're claiming here is effectually that, not even the content, but the mere titles of every single "top 50 foos" list on the internet can be copyrighted

Again, I'm intentionally _not_ touching on issues of copyright as legal status - but, yes, I am absolutely suggesting that someone who has gone to the effort of curating a "top 50 foos" list, _and_ of associating it with a powerful taste-making brand like Spotify's, would be justified in perceiving that list _itself_ as a valuable and protection-worthy creation.


After the cut because it's less relevant to the discussion, but:

> that the rearranging of a book's chapters without any change to the contents, can itself fall under copyright

I mean...if you _don't_ see how it could be possible for someone to remix the content of an existing artwork into a conceptually-new work, by making statements _with_ the playing-with-form, then I think we're just looking at art from fundamentally different perspectives and are never going to agree. If you truly _do_ want to understand my perspective, you might try thinking about how the consumption of art is affect by the context of how it is consumed and presented, not just the sequence of bytes/soundwaves/visual-elements that compose it.

I believe they're referring to the playlists created BY Spotify

+1 for the general fantasticness. I use the app on my iphone - no problems whatsoever. They even recently updated their watch app to stream directly (without the phone being nearby). Apple music has had this feature for a while.

It feels like it's a self-fulfilling spiral of bad music. The more I listen to the recommended music, the more of the same crap keeps getting recommended to me. Even if I skip them.

I remember that Spotify actually had pretty good recommendations when I first started using it at the start of this year, because I had only listened to the songs I actually liked so far. But now I'm lucky if I find one decent song out of a hundred recommended ones.

Seems there is exactly the same sort of discussion on another HN thread about poor recommendations from YouTube, also leading into a spiralling decline to more and more rubbish

Yeah. I've commented about this in the past. I have two YouTube accounts in different languages. One has great recommendations. One is terrible.

I wonder for the one with terrible recommendations if I fell into a particular failure spiral: YouTube tried to give me some new stuff for a period, but I happened to not like any of it. The AI noted I was just clicking suggestions from my subscriptions. It eventually tried different recommendations. I didn't like those either, and clicked videos I'd seen before from my subscriptions. Eventually the AI decided I only like watching videos for channels I'm subscribed to, and now doesn't recommend me stuff outside of that. Even when I branch out into a new area, it no longer suggests related channels or anything like that.

It's not just media. Every year or so I reinstall the Swype keyboard and it works great. But after about a month or two it 'learns' from my jokey misspellings or starts learning a lot of my more obscure vocabulary and starts muddying up the very basic level of Swyping I actually want it to do.

It works great when it types out phrases as I swype and I go back and edit in whatever jokes, memes, or 10 dollar words I want to include. It starts to suck when it starts trying to include those into its predictions.

There's like a sweet spot of machine learning beyond which the machine gets more annoying than helpful.

Hilariously I just switched from Apple Music to Spotify and had the exact opposite experience! I wonder if recommendations get "stale" so you get bad ones, where as if you switch services you get recommendations based on different data sets so they're temporarily better.

In my experience Spotify's recommendations for my big anything-goes playlist are terrible, but the one's for my smaller playlist for mostly British D&B music (https://open.spotify.com/track/22Z4p2gRkDPqmw8DgVixML?si=jeb... - for a taste of the genre) can give very specific and quite good recommendations of smaller artists.

I can't speak to Spotify but YouTube Music seems to do a good job with recommendations. If I start by seeding it with a "radio station" built off an artist I enjoy a lot of the follow-up music is enjoyable and helps me to discover new bands. YouTube Music even has a free ad-supported tier, which is nice.

Hahahaha YouTube has 8 years of my Google Play Music listening history, as well as my entire music library I uploaded back in 2012 when Zune (later XBM/Groove) got killed.

I started it on "The Little Things Give You Away" by Linkin Park, and let it play. For a while it was alright, it gave me chill downtempo kinda emo music selections with some like newer stuff from Linkin Park and like Green Day.

Normally I listen to slow/soft shit from bands like 3 Days Grace, Breaking Ben, 3 Doors Down, My Chemical Romance, Good Charlotte and so on with my Linkin Park (especially the Minutes to Midnight album), so Green Day and Fall Out Boy aren't bad per se even though I like my older stuff. And I do have lots of playlists with GD + LP + FOB + other random pop punk bands.

But then it started throwing in shit like NF, Machine Gun Kelley, Eminem, Logic, then D12, Run DMC, DMX. The Smart DJ on my decade old Zune HD does a better job at building a playlist and recommending music (with the same library).

Pandora, Last.fm, GPM (after 8 years it finally started to "get" me FUCK YOU GOOGLE), Spotify, YouTube Music, Apple Music, etc. They have all done this to me, they all inevitably end up playing some sort of rap, or move to super modern pop music. I do not understand it at all beyond 'it makes them money'.

I do listen to rap music, and I do like listening to those groups and artists. I have quite a few playlists built with just that music, and aside from Eminem, they do not intersect with my emo/punk/rock playlists. If I start listening to Linkin Park, I am never going to want to go from there to early 2000s rap.

I. Don't. Give. A. Rats. Ass. if every other person on the planet listens to Linkin Park with Eminem and DMX. I don't and I tell them every time I play a song or build a playlist, or skip, or remove a recommendation what I like and don't like.

Why can my deprecated Zune appreciate that and keep my genre interests separate and accurate and a billion dollar AI algo can't?

I made the same assumption about what the link was going to be about.

My solution to terrible music algorithms this year has been to, well, give up on them entirely, and go back to the world of curated music. When Google Play Music died, and YouTube Music left an even worse experience in its place, I was done. So now people curate my music instad of machines. Whether that's in the form of letting the artist curate it for me with albums, letting myself curate my favorites with playlists, or letting a DJ curate music discovery on a streaming radio station, I find that humans do it better.

College radio stations (and their streams) are great, by the way. I can't evangelize them enough. Minimal or no ads, and some of the most interesting and creative sets I've heard in my life.

Music algorithms didn't use to (or seem to) be terrible when the initial data they fed on was already curated music collections. It allowed the algorithm to really learn what you liked then provide recommendations from people with very similar music libraries.

Now it seems that everyone is listening to songs and albums based on recommendations based on what everyone else is listening to, which is also based on recommendations, so you don't find what you really like, you find what everyone else likes and it creates a feedback loop of mediocre garbage totally unspecific to your tastes. And if you try to break out of the loop you land in a loop with people who "broke out of the loop" and you get caught in the cycle again with weirder shit. Rinse and repeat until it pigeonholes you into some inescapable nether of music you might nearly but not really be into.

I would also like to plug Green River College's radio station 89.9 KGRG FM. It has an app and I've never heard a better modern Rock/Metal station

Their recommendation engine is really bad IMO. I was hoping for a futuristic machine learning + listening based recommendation engine that could find similar songs based on what they actually sound like. Mostly they just find songs based on what other playlists your song exists in. It's really boring and not good at finding new music for me. It does OKAY and it isn't why I use Spotify, but I can't help but feel like they are missing the boat here. I remember working with Gracenote software like 15 years ago and it could fingerprint songs and find similar music at least that long ago.

e: my overall sentiment on Spotify is very positive, I just wish recommendations were better.

Check out SAGE from the guy at Hate5six. It's more punk/metal based, but if you're into that, you'll definitely find some new music.


Pretty good! I tried it with some random cross-genre combinations of stuff I like and it found a bunch of other fairly niche bands I like. Looking forward to trying out the other recommendations. Works fine for genres other than punk & metal btw.

Blog post below on how the hate5six creator made this algorithm. Boils down to using the lyrics and community listening habits from last.fm to build a similarity graph for artists, then community detection to group similar artists.


Similar experience for me - they always shuffle me most awful songs ever - I could never rely on their algorithms to the extent that no matter the mood/kind/playlist I choose, I end up skipping 20 songs that are just horrible and I just quit Spotify.

I wonder how much better algorithm could be if they'd recognise a skip within 10 seconds of a song as "no, bad choice, never play it again".

Honestly, I switched to funkwhale and I'm not looking back. I prefer to control my own "sound system" like in the old times.

After playing an album I select (usually soundtracks), Spotify falls into a tiny selection of 10-20 tracks that it plays on shuffle forever. It appears music discovery is incredibly poor.

...I'm also afraid to play something for my son, then the recommendations while coding will surely be polluted by that; just like if you play one or two kid's videos on youtube.

FWIW, I’ve had the exact same experience. Started a 3 month Apple Music trial and was so blown away by the number of bands I’d never heard of and absolutely loved that I am ditching Spotify. I somehow found my Spotify account recommendations either recommending songs of bands I’d already told it I liked (duh!) or new bands that I absolutely detested.

It seems to always give me the best songs first and last in the Discover Weekly and Release Radar auto generated playlists. But in the middle there are sometimes unbelievably ill suited songs for me. Maybe it’s the artificial intelligence attempting to probe the boundaries a bit.

I think it just gets stuck on old songs you played. My kids played "The Bacon Song" a few times on my spotify, you can imagine what my recommendations are like now. I wish Spotify had a way to remove "bad songs" from whatever algorithm is used.

I always thought my recommendations are that broken because I listen to two very different music directions. For now, I fixed it by finding new Music via YouTube and using artists playlists etc.. Apple Music is probably no option, though, as I'm exclusively on Android.

I've had the same problem. I like multiple genres but Spotify recommendations tend ever more toward overly harsh and angular electronica. I've tried retraining it multiple times by intentionally listening in a different direction but it never fixes the problem for long. Really frustrating.

Android does have an app for Apple Music though it's been years since I've used it. I was quite happy to pay the subscription fee but an update broke the whole app for me. It would play only a part of the song and then suddenly skip to the next one.

My music taste also diverges drastically but Apple seemed to handle it quite well IIRC.

sometimes spotify recommends songs that are so bad it makes me wonder is it some sort of test to check if i am even paying attention :)

i get good results from discover weekly for the most part though. I used to spend a lot of hours on social music sites like thisismyjam and the like, where I would stumble across a gem every 1 in 50, but with spotify it feels more like every 1 in 20 or 30, so I'm happy with that and it's also less work not switching between multiple services

other people I know have nothing good to say about discover weekly so I dunno. maybe I just have low standards!

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact