How bad is your Spotify?
pudding.cool
pudding.cool
> 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.
Ouch.
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?
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.
you ok?
[lol, jk.] [yeah, why?]
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)
The Act of HN submission. But broadly speaking US working time zone get a much higher chance of submission being upvoted.
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.
Not gonna lie, I'm feeling a bit judged by this AI:)
Asked the same thing to me, but about Radiohead.
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.
- Ryan Caraveo, u okay? - ...of course, K.Flay
Hope that didn't ruin the magic.
[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
> > Your spotify was ira-glass-in-your-airpods-manic-pixie-dream-girl bad.
Made me laugh!
My spotify was 'heavy-eyeliner-post-punk-boomer-relaxation-dads-still-cool bad'
> 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.
I got the same thing with a different song. What's funny about this one is it's from 2017.
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.
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.
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
VNV Nation
Winterfylleth
WEH
Rome
Falconer
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.
> 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!
> 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.
Other of my fav instrumental band is CHON, but they're more of math rock and not really the same.
oof, I'm not even angry lol
> 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.
"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."
"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."
> 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.
The output it spitted out while processing was pretty good though.
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!
https://hn.algolia.com/?dateRange=all&page=0&prefix=true&sor...
https://hn.algolia.com/?dateRange=all&page=0&prefix=true&sor...
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.
select case (person%likes)
case ('80s-synth-pop')
print *, 80s_pop_insult_char
case ('single-specific-artist')
print *, person%likes_artist//'-stan'
! (...)
end selecti regret letting this access my account. i would recommend others to not give this tool access, it is not worth it.
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.
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.
That said of course I agree they should have an explicit privacy policy regardless.
> 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?
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.
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?
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.
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.
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.
Spotify takes a lot more risks and I think its highly preferable.
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
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?
I guess Spotify is still the only game in town :(
The rest of the comment still stands for all US based users, however.
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...
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.
At least someone's still at it, and all this helps me find my notes from that time.
https://github.com/LDugar/MusicGenome/raw/master/Decoding-th...
[0]http://www.pandora.com/about/mgp [1]https://github.com/omgDB
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...
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 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.
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.
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.
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.
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.)
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 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.
The implementation and mixing music videos with other video content in my Youtube account is horrible on the other way.
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...
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.
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.
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.
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.
Is this a new thing?
"Everyone else's opinion * objectively * sucks".
Just because you write "objectively", doesn't make it so.
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.
It's still the best music player I know of though, even when running with Wine on Linux.
see: https://i.imgur.com/zL0oZsx.png
You can customize this feature in Advanced Preferences. Then, go to Playback -> Shuffle -> Album grouping pattern.
Quod Libet (https://quodlibet.readthedocs.io/en/latest/) lets you do exactly that.
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?
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.
It doesn't matter how good the tech is if business/legal can't keep up with the competition.
Which button is that?
> effectively, I manually make my own playlist.
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.
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.
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/
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.
It definitely has the UX feel of being a bit creepy on how well it responds.
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
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.
[1] https://www.youtube.com/watch?v=i8eL95fIQbk
[2] https://www.youtube.com/watch?v=iNI-2i6t7Zs
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 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.
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 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.
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 also got some Scandinavian stuff too but it usually fits my taste at least.
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.
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.
https://play.google.com/store/apps/details?id=com.apple.andr...
My music taste also diverges drastically but Apple seemed to handle it quite well IIRC.
Just as a counter point.
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-
https://news.ycombinator.com/item?id=24747636
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.
The playlists that users created? No.
I apologise if that distinction was not clear enough in my original post.
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'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.
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.
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-...
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'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 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.
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.
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.
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".
e: my overall sentiment on Spotify is very positive, I just wish recommendations were better.
https://medium.com/@hate5six/sage-an-artificially-intelligen...
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.
Except Phish fans. For them there is only Phish.
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.
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.
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'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.
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.
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!
It's less work to do what I've been doing all along: get recommendations from friends, family, articles, collabs, biographies, etc. (And it's a lot more fun.)
Do you _really_ think they wouldn't've already implemented it already if it were possible in a reasonable way?
- Adding things to playlists makes a big difference. If you really love a song, hit it with a heart, if you want to hear more but don't love it, add it to a relevant playlist
- skip mildly bad songs
- actively dislike terrible songs (hide it on mobile, hit the no smoking sign on desktop)
I got into a chill beats phase and it took a couple months for me to eradicate it from my discover weekly but now it's pretty on point. Just my 2 cents.
Also whenever I'm DM'ing a D&D game the majority of my daily mixes change to video game music for weeks because I use them as backing music.
I wish other websites had similar recommendation systems. Spend an afternoon watching shitposting videos and YouTube will only recommends low-effort videos.
It didn't identify that one of them is hip hop, and the other one is dutch hip hop, which obviously to me are the same genre, but that's alright.
It's got a classical music mix, one for rock, one for soundtracks and one for electronic/lo-fi.
It would've been cool if it knew to group electronic/lo-fi, classical and soundtrack together as my background work music, but I'm not sure how it would've learned that as normally I have to pick one of those genres while working, I have no playlists that mix them.
I see this suggestion a lot, but for whatever reason I never see an option on desktop to dislike songs. Unless I'm really blind.
Does it show up near the track controls?
I listen to a lot of ambient and instrumental music while I work, but I don't want recommendations based on this, so I always start a private session.
https://miro.medium.com/max/3836/1*C_ImLXQdMOGljMRhLcV-Yw.jp...
...was playing on the radio station, so that you could then turn on the radio and listen to it.
This was - unironically - the big thing they were using to try to get users to download their app. I can't even. This is so dumb.
Can we just kill radio already? Please?
Sure, and that's what Spotify IS doing, only they can do it for literally any song/artist/genre, instead of just the mainstream ones. If the RIAA wants to give me some melancholy lofi during my melancholy lofi listening, I don't really mind. I don't have to sit through an Imagine Dragons song, which is the radio alternative
https://www.theguardian.com/technology/2020/nov/03/spotify-a...
The recommendation features, albeit imperfect, are a big reason I'm still on Spotify. I don't think they're secondary at all, as I would have gotten bored by now and canceled my subscription. True, I could go through all the effort you mentioned, but realistically I wouldn't.
(I have some experience with Tidal, which also has apparently human-curated lists but they haven't at all at Apple's level. My experience with Spotify wasn't good, but it was also really limited, because it just didn't offer anything obvious over Apple Music to me -- the reverse is almost certainly true for someone who's already deeply invested in Spotify, though!)
I mean, maybe. I only listen to music when I'm a) driving; b) working. Neither are times I'm particularly interested in searching for new bands to listen to, so I'm going to end up relying on my streaming service's recommendation engine.
To each their own, I guess, but I’m naturally a “passive” music discoverer, using a combination of recommendations and similar artist radios and whatnot. It works fairly well.
I have a "Queue" playlist into which I copy my "Discover Weekly" every now and then; then I listen to that Queue playlist, which lets me delete the tracks I don't want, and keep the ones I do (you can't modify your Discover Weekly). Then, every time I refresh the Queue playlist from Discover Weekly, I first move the current Queue contents into a YYYY-MM playlist (e.g. 2020-12 for this month).
I've been doing this for several years and it's been great — I've got a timestamped record of what I was enjoying/discovering during those months, and it does seem to have helped Spotify to keep recommending me things I really do enjoy. (Weirdly some weeks seem _really bad_, like I delete virtually everything — but I guess there's some A/Bing going on there.)
(Also: I'll only like/heart tracks I _particularly_ like, and sometimes I'll just listen to my "Liked Songs" playlist; I think this helps teach Spotify what I really like.)
Noticed the same, but I think this is actually great, because it seems like there is some randomness that gives you different stuff. It’s like breaking your filter bubble.
Patiently anticipating how bad my taste is
(In Bender's voice) You think you're all high and mighty, just because you're an AI? Well, I guessed so.
EDIT On a real note, how do you feel about doing the same for my last.fm that has data since 2007?
I'm stuck in "Loading your music library...Spotify limiting how many people can use this app at once [...] I'll try again in n seconds" limbo :(
Hit a bug where it asked me “do you recognize this song?” And tried to play some audio, but nothing actually played. (Safari on iOS; could have been caused by an ad blocker)
It also absolutely disliked the amount of Daft Punk I listened too, and called me pretentious for listening to Periphery, this was hilarious.
Yup. That's me.
m000, you are my new best friend.
Liked those two, try these two.
My results and text pieces are pretty much the same as everyone else's here in the comments.
It just fetches some playlists, fills the blanks in its text pieces and that's it.
It's vaguely sentient.
It has some really bizarre opinions about certain songs - Guthrie Govan is basically a god amongst men for guitar players but it absolutely hates his work.
A final note: I assume this is a joke, but please don't try to objectively rank music.
They apparently just took public reviews, but I geniuinelycan not imagine someone disliking Guthrie Govan. Maybe the score is also inversely proportional to popularity/# of reviews found? It'd fit into the hipster gag
Like/Dislike ratio of 6600/98, they must be extrapolating from some bad data (or possibly extrapolating from the genre not being particularly melon-y).
Excuse me, I need to go and practice...
I keep hoping that someone will crack the micropayment idea and kill advertising on the web.
Trying! https://wiki.snowdrift.coop
Ok -- technically not micropayments and technically killing advertising is not a goal. But many of us are personally against advertising, and a viable alternative for funding free/libre/open works would certainly put a dent in advertising coffers.
How to flirt better. How to earn more money. How to be more lucky.
In the beginning there were lots of sex podcasts, seemed like everybody who had sex at least twice got its own podcast on Spotify, but now it's all filled with the generic stuff I already hate in the other places.
is pretty much the same reason I moved to linux because microsoft kept trying to push crap apps like candy crush on me even though I bought the "pro" licence. #everythingIsTerrible
They have many competitors, and tools to copy your music selections between services exist. For example:
Also, Spotify pays musicians less than most other services:
https://www.dittomusic.com/blog/how-much-do-music-streaming-...
Spotify’s pricing is also inline with competitors. I don’t get it.
I tried three competitors as I moved away from Spotify, and found they were all substantially better than it.
I care about music selection, recommendation quality, and library stability (not deleting stuff from my collection).
Just the other day I built my own podcast picker script that curates a playlist of subscribed podcast episodes in exactly the way I want (balance-wise/order-wise). Sounds minor, but it has been huge for my smart-home listening experience, given how terrible the out-of-box podcast experience is with Alexa.
The concept of "objectively good music" is hilarious enough that this is already a win in my book.
That's not you, that's bad AI :)
Spotify permissions it asks for make sense.
It was really confused by all the kid music I play. Why so much Raffi? Their training set must not include people with children. :)
> "You listen to 'Super Simple Songs' a lot. I mean A LOT. Are you ok?"
Fuck this site.
I suppose because soundtracks and 90s dance remixes are what I listen to while programming.
For soundtracks, I made a playlist that goes for 28hrs 30 mins full of:
* Hans Zimmer
* James Horner
* John Williams
* Alan Silvestri
* David Newman
* Danny Elfman
* Christopher Tin
* John Powell
* Michael Giacchino
* Ramin Djawadi
* etc
In terms of bad vs. good, though, music is like a lot of things in life, so subjective that you can't judge another person's taste. If for no other reason than people care about music to wildly different degrees. I've met people who say they don't listen to music. As in ever. Others cannot let a minute go by without having something playing in their head. It's akin to judging taste in food or recreational activities or movies. It's not really about what's good, rather, what gets a positive mental response.
That point aside, I'd say that what would be interesting is measuring how broad a person's musical tastes are. I.e., how much have they been exposed to and how much of that have they assimilated into their regular listening rotation. Pandora does a pretty good job suggesting artists I've never heard. When I hear one I like, I'll incorporate some songs into my regular rotation on Spotify.
If you must know, I have Bad Taste. The bad as in Hawaiian shirt-etc.-etc.-etc.... bad.
That said, I thought it was funny because I realize people making fun of me for not being cool doesn't have any effect on me anymore. Is this what middle-age is about?
It also focused on mocking my musical tastes being stuck in the early 2010s, but I haven't actually used Spotify since 2015 so I guess that checks out from its perspective.
Connecting my Spotify account to the Alexa my 3 year old has figured out how to use was a definite mistake.
very funny to get roasted by a machine though!
Thank your obsessions with Steely Dan, mellow gold, and Africa for that.
Based on your listening habits, I can also tell you your spotify was...
60s-acid-flashback bad
mercury-idolizer bad
parody-song bad
midwest-classic-rock bad
punk-with-grandchildren bad
But wait, it gets worse:
You listen to these tracks too much:
Calling Occupants Of Interplanetary Craft by Klaatu
Working Man by Rush
WEIT WEG by Rammstein
While My Guitar Gently Weeps - Remastered 2009 by The Beatles
The Dope Show by Marilyn Manson
Ninja Sex Party
Rammstein
Led Zeppelin
Steely Dan
Rush
You're stuck in the early 2010s. You must have peaked right around Ninja Sex Party's Under the Covers.
That's it. I'm done.
I need to go recalibrate my taste levels.
Shutting down."
Yeah, I'm a 90's kid who listened to their parents' music from the Classic Rock stations they always had on. If anything, I'm more stuck in the 70s; the early 10s is just when I stopped paying much attention to new music, though my wife is trying to change that. You can't shame me, A.I.!
What's funny to me, is that I actually discovered Klaatu via Spotify. I'd never heard them before, but they came up in one of my weekly playlists, and (obviously) I took quite a liking. "Calling Occupants" was my #1 song of 2020.
> Thank your obsessions with Childish Gambino and christian rock for that.
> Based on your listening habits, I can also tell you your spotify was...
> awkward-beatboxing bad
Bruh
My activity seems to have messed it up:
"You only listen to Obama-era jams like Sleepy Brown Noise by White Noise for Reading" err..
How do I delete my data?
Both have a lot of potential, as it would seem the former President and I have very similar musical tastes (that is: a very strong love for Stevie Wonder)
> You are 0% basic. Have you considered there's a reason nobody listens to Perihelion Ship?.
Screw you, they're great. Its Opeth with growling.
So... Opeth before Heritage? lol
Jokes aside, I think this is a really neat toy with a great sense of humor
Your Daily Mix # - that works better but it's mostly repetitive stuff I already have in my Library.
They need toggles of some sort, or some dial of some sort rather than put blind trust in these algorithms.
> sad-knowledge-worker bad
> wear-carhartts-to-your-desk-job
> remember-when-williamsburg-was-cool bad
The comments before were not enlightening enough, as I was music editor and dj on a local indie radio station, so it should be the other way round. The general taste sucks. Spotify's release radar only recommends shit, but nevertheless it's still the best to find the hidden gems. You won't find them on radio shows, other than KTRU or WFMU.
> Based on your listening habits, I can also tell you your spotify was...
> stoned-itunes-visualizer bad
> cling-clang-pots-and-pans-music bad
Gotta say, that's a pretty perfect call out of my average music listening.
- Your spotify was anime-slow-dance bad,
On "You listen to these tracks too much:" part, well that is because they are on the top of the playlist. I usually start my Spotify day by just playing the playlist (Liked Songs) so it goes just play the music on whats on top and go down to the first song I fav'ed.
- You are 2% basic. Oh wow The Contes and Crown Jewels! Your taste is so obscure that's so cool I bet you're super interesting..
This one got me :) I really like "The Contes" and "Crown Jewels"
- You're stuck in the early 2010s.
Hmm... Kinda not accurate. Does it know when the track was originally released to make accurate one? Like there was an album on my list that it was originally released on 1998-2004 but was released on Spotify earlier this year 2020.
"fuck marry kill.. fast" was fun but got off took me a while to figure it out, I quickly clicked "fuck" and "marry" on SCANDAL (why not right? haha) and thought the A.I. is dead, later I figured out that I need to choose on others too.
> Thank your obsessions with Of Monsters and Men and symphonic metal (e.g., Luca Turilli and Van Canto) for that.
> Here's what else I learned in that hellscape:
You listen to these tracks too much:
[too embarrassing]
You stan these artists to an uncomfortable extent:
You are 4% basic. Мёртвые Осы and Zabili Mi Żółwia? Where do you even find this shit?.
You're too trendy for your own good. You only listen to music made in the last year like...
It's amazing that it correctly spotted objectievely shitty songs on my playlist.
That might be the highest compliment I've ever received!
There is no 100% music (or any creative output) recommender because everyone feels differently about what 'energy', 'easy', 'powerful', 'touching', and 'rock' all mean, and to each person individually. These terms are meant to, and always will be, necessarily ambiguous; and that is how these terms work best.
AI currently cannot 100% give you a track without drums even if you check the 'no drums' box (because along the way one track with drums was put in the no drums box).
Incidentally I've used various searchers (now referred to as) AI for years to try and find tracks requested by clients, it remains somewhat helpful though unable to compete with a human first listening to the human making the request.
AI is improving, but it is going to be beyond our lifetime before it replaces communication, human to human, eye to eye.
Until AI knows what it means to be human, so it shall remain.
Thus: Your spotify was like-country-but-smoke-weed-with-your-gay-friends-drama-club-tonys-watch-party bad.
You can get around this by closing the account, and opening a new one, but this is too time consuming, since you need to do it frequently.
I thought it was a fluke at first but it has happened now with about 4-5 artists.
What's actually more fun is Spotify's yearly playlists with your most listened to songs. I always get a kick out of analysing those.
But I would not want to let that site log into my Slotify account.
Gnoosic is a site I love. It creates AI based recommendations without any login:
But I only use Spotify a) in the car and b) while riding my bike/bike trainer. I have an actual DRM-free MP3 collection[1] for the important stuff--which has a little overlap with Spotify, but not too much.
[1] Small - only about 20 GB - and has been contiguously curated, over five distinct "generations" in my life, since I was in 7th grade (1998), when I had a 486 DX4/75 that could just about play an MP3, downloaded over dial-up from a shady FTP server, over my SoundBlaster 16 if I wasn't running X or doing anything else.
> Your spotify was jazz-everyones-heard-of-jazz-snob-never-turn-down-a-microphone bad.
> You're too trendy for your own good. You know there's good music from before 2019, right?
By the way, I hated the Spotify ads so much I paid up for YouTube Music + Premium. Even though the Spotify Windows app is VERY usable compared to https://music.youtube.com. Good Lord, I hated those ads. I know the conspiracy theory that they were designed to be irritating so that you pay up, but that only works in a world without much choice; ;now you have YouTube Music, Apple Music, Amazon Music, and Deezer. Spotify, I'll readily pay up and switch if you improve those awful ads.
I'd rather just not use spotify at all.
> Thank your obsessions with Michael Jackson and Doom (Original Game Soundtrack) for that.
> But wait, it gets worse:
You listen to these tracks too much:
Enter the Cipher by Follow The Cipher Afterlife by Metalite Winterfall by Follow The Cipher Valkyria by Follow The Cipher My Soldier by Follow The Cipher You stan these artists to an uncomfortable extent:
Amaranthe Mick Gordon Nightwish Celldweller Little V. You are 1% basic. Have you considered there's a reason nobody listens to monomer?.
You're stuck in the early 2010s. For you, music's been all downhill since Amaranthe made Leave Everything Behind.
Well, that was really something.
I need to go recalibrate my taste levels.
Shutting down.```
well shit
But when it starts telling us the stuff we think is high-brow stuff is actually crap, that's when people really start feeling insecure and second-guessing our personal taste. You start thinking wait, what's wrong with Nine Inch Nails and Haim? I thought they're supposed to be good...
Tangential: I wish one-off applications like this also offered a one-click de-authorization step at the end to remove access to my account. It’s a pain to go figure out how to do it for certain accounts, especially if you’re mobile and there’s a decent chance you’ll just forget to do it later. I do not like leaving access tokens around in other people’s databases unless I have to. I understand this would require the service to provide such functionality to the application developer, which I’m sure isn’t always available..
Spotify has done an excellent job of recommendations lately. It took several years of liking songs and making playlists, but I've been very impressed with Discover Weekly.
The shuffle function is still broken, however. It only "shuffles" from like the top 50 songs.
UPDATE: It told me JERRY MARTIN is basic?! The composer for Sims and Sim City? Bruh...
I revoked access to the Spotify API shortly after use, and it didn't seem to do any harm to my account.
False. Time to throw this into the death pit.
I definitely don't use my Spotify account enough for it to do meaningful recommendations. Too much time on the Discover Weekly and it is more and more diverging from tracks I actually like. The algorithm feed itself and lose precision.
I miss the recommendation capabilities of rdio.
If you do radio for a decently sized playlist, it is basically just akin to using shuffle. What is the point in that? Absolutely useless for music discovery.
Log in with Spotify"
I never used Spotify, so he was doing this assessment from my cookies...from where? Youtube? I don't login there either, I just browse it. Anyway, that's my test on this.
I believe it focuses quite a lot on artists or tracks listened to a lot, and judging from the covers it shows it selects some top N songs for its analysis. Fun stuff though!
Sincerely, AI
This tool is also awesome -- `Finding a lot of Above & Beyond.` Yes, you are and there should be even more!
I often find myself playing songs for others I myself wouldn't want to listen to and as a result my Discover Weekly feed is unusable with no apparent fix.
Truth...
> Thank your obsessions with Ed Sheeran, mellow gold, and Wonderful Tonight for that.
> Based on your listening habits, I can also tell you your spotify was...
> mainstream-country bad
> like-country-but-smoke-weed-with-your-gay-friends bad
> DMB-die-hard bad
> wear-carhartts-to-your-desk-job bad
> requests-bohemian-rhapsody-at-weddings bad
> Here's what else is going on in your aural trash fire:
You listen to these tracks too much:
Highwomen by The Highwomen
WAP by Cardi B
Classic by Cam
Forgetting You by Cam
Bridge Over Troubled Water by Willie Nelson
Ed Sheeran
George Strait
Sturgill Simpson
Cam
Dolly Parton
You're stuck in the early 2010s. You only listen to Obama-era jams like Poison & Wine by The Civil Wars and Emmylou - Spotify Sessions by First Aid Kit.
I can start over with a new account, but I will lose all my bookmarked songs.
QA did not bother to check this simple on boarding process for long time users.
Personally, I never use anything that asks me to sign in with something else. My data is my business and IMO there are plenty of ways to get a laugh without compromising privacy.
> Like... a LOT.
> Your spotify was moms-spaghetti-movie-soundtrack-baby-shark-prison bad.
> Thank your obsessions with Eminem, soundtrack (e.g., Hans Zimmer and Two Steps from Hell), and Baby Shark for that.
LOL my kids used my spotify a lot...
In terms of recs, it does ok. There are weeks when every discover Weekly track is bizarre ambient music or very quiet music, like I'm supposed to be a spa owner. Then sometimes it has delivered great tracks. But as a musician, it's very rare for me to find new gems, I've heard of so many via friends, class, YT, etc.
Overall, spotify is great, but the app and algorithm could use improvement.
Loading your music library...Spotify limiting how many people
can use this app at once, so you'll need to wait or try again
later.Spotify by comparison is a crime against good taste.
Really well written. Unique responses to many (I thought) relatively unknown songs and artists.
The final score, a rambling "BAD" rating, was very funny.
Great job!
When we judge whether music is "good" or "bad", we're evaluating it as a work of art. Being novel, smart, well performed etc are all naturally important considerations.
But when you listen to music for hours while working, studying, workout out, either alone or through headphones ... music is no longer art exactly. Music becomes a tool which helps us guide our own moods or quality of attention. Music becomes a cognitive aid.
My desk is bad as sculpture. My documentation is bad as literature. My thinking-music is generally bad as art-music.
Isn’t this like one of those personality harvesting Apps on Facebook?
Can’t we permission The Pudding to see our place instead of logging in as us?
You've been listening to a lot of Yo La Tengo lately.
u okay?
You stan these artists to an uncomfortable extent:
The Beatles
Cream
Led Zeppelin
Rush
Yes
Whoops sorry, I'm a big prog fan :p
I resent this comment
As a 35 y/o with a toddler I'd say they nailed it.
I'm so offended now. I mean, not surprised...
I envy those people
Pandora was OK. The issue I had with it is that "Liking" or "Disliking" a song seemed to poison the playlist.
YouTube is what I use to discover music today. It can take me down a rabbit hole, but going down one never seems to adversely impact song recommendations the next time I go searching for music.
Newer stuff I tend to like usually ends up being on Bandcamp.
Hah.
This is just awesome, man.
I guess it would be even worse, except that most of the Tzadik Records catalog isn't on Spotify.
> Your spotify was music-to-stalk-boys-to-hail-satan-heavy-eyeliner-post-punk bad.
> allegations-of-abuse-have-never-been-proven-in-court bad
> gallagher-brother bad
> instagram-your-tent-at-sunset bad
Thanks, AI.
... LMAO. love it.
u okay?
Awesome project
Mine was:
You've been listening to a lot of Monolink lately.
u okay?
uh ... Spotify?
Super curious though.. what data did you use to get the model to write super accurate slang
My current approach at least makes me really happy. The problem is I'm likely radically different.
The general problem is so are many people. But all in different, incompatible ways.
For instance, my (current) approach is fundamentally label based. Things are organized by label and I "rate" releases as I listen to them. This leads to a generalized "label rating" which can be partially extended to (1) a rating of the artists who released on the label along with (2) albums and eps which may see a release under another label. Then there is a similarity index to other labels based on a percentage of artists who release on both labels and this leads to something going into a big "consider" pile which then gets treated randomly. It's not really a "recommendation" as much as it is a "might be related? i dunno" system.
For people who want familiar songs they can sing along with, this is a terrible no-good system. For people who like to listen to the same song multiple times a day, this is equally useless. For those who want steadfast genre uniformity and things to sound more or less the same, this is also useless.
This is designed to be exclusively long-tail and stay firmly within the confines way out at the end of the tail. But that's exactly where I want to be.
In the pareto principle, everyone in the 80% looks more or less the same but everyone in the 20% is strange in a different kind of way.
There is no 1 recommendation system that works for all customers. Users have to be segmented. It's utterly inescapable. That's why these systems try to just service the 80% and just not put any effort to the other 20%.
The problem with this approach is the 80% is no longer 80%. More people are becoming snowflake 20%ers and this trend is likely to continue. The mainstream orientation engines of radio and cable television aren't influencing like they used to and the 'service the 80' approach just won't work any more. It's over.
The single recommendation system approach is being tweaked to try to service this coalition that no longer exists. So what happens?
Everything starts to suck. Google results, youtube recommendations, spotify, they all start to be unfocused, scattered messes that just discard all the signal, amplify the noise and return piles of nearly identical derivative garbage.
Socially we've moved beyond the assumption that simply everyone in society will be in some conventional nuclear family with the wife vacuuming at home and the daughter playing with dolls but technologically we still seem to model humans using utterly conventional and outdated single purpose models that simply do not apply and are increasingly a poor fit for everyone.
The new normal is that there isn't one and it's time for tech to play catch up.
Too much beebop and nothing past 2015.
A lot of people here revelling in their 'not basic' results, the thing is, that's maybe something interesting, but the vast majority of arcane music is straight up bad. Just because it's 'different' doesn't make it good. Most artists who are actually 'pretty good' have some kind of following - they get picked up, at least marginally. Of course there's nothing wrong with 'liking that band with 2K views'. One reason this happens if if they are 'genre' meaning, a newer band, playing into a past sound or genre, people may not pick up. A local band here in Montreal is really good, but they play a kind of specific Cajun country/rock, so their appeal is 'respectable' but 'quaint'.
Since 2013 when I've started paying for Spotify it has turned from one of my favourite apps to one of those I hate most. On my previous Android phone it would freeze every other day when there was no internet. Why? Who knows. Could be related to me using SD card, or could be something else. The phone wasn't very old, but still on many days it was basically unusable. Somehow I was gifted a new phone and it kind of works. I say kind of, because sometimes when I click on a downloaded album it doesn't work - no tracks display. Why? Who knows. And few days ago something bugged out and tracks suddenly started setting volume to max. Apparently it's known issue that appears occasionally. Why? Who knows. Happily I wasn't wearing headphones so I did not get hearing damage. But who knows, maybe this will happen again.
There is also desktop app which severely crippled local file support few years ago. That really sucked for me, because I record a lot of my own stuff and I want to listen to music in one place. I don't know if this was fixed since then because I pretty much switched to using web app, which usually works. There is also iPad app which seems to routinely delete stuff I have downloaded do my iPad; and sometimes it also freeze in offline mode when browsing my library. At this point I didn't even try filing a ticket with support, I lost hope of them fixing this. I suppose soon I will simply stop paying them and make switch to something else.
EDIT: Why post this here? I hope people at Spotify see that submission about Spotify is on top of HN, and maybe, hopefully, someone sees my comments and thinks twice about setting 2021 OKRs and put some capacity into improving the app reliability. I know, this is optimistic, but who knows.
Having said that, it's wonderful. I can download all my music. It has some truly wonderful playlists for when I'm coding. It's recommendation engine seems to recommend really nice tunes week after week. The yearly reviews, the On Repeat etc. all this stuff is really nice to look back at.
Then, I FINALLY have a podcast search which can search for podcast episodes and then, when I click on the search result, take me to tho god-damned episode instead of the full podcast with 2500 fucking episodes to hunt through. It actually has a sane podcast interface that isn't a cluttered mess.
I was so sceptical of it because I loved Pandora to the absolute hilt, but now it's the only Saas I pay monthly for.
Spotify has been a breath of fresh air, I switched a couple years ago. It is missing some stuff here & there, but I haven't hit too many bugs like others have (on Windows or Android), and their catalog is much bigger than Pandora's was when I was using it heavily. Turns out, what I needed was more content. What once seemed like Pandora's best strength - that 'special sauce' recommendation algorithm - is now clearly being done to a similar or greater level of competence by damn near everyone else in the space. And like you say, the podcast stuff has been solid & gotten better, too. These things never turn out how I expect.
I suspect recommendation is more steered by contract agreements than showing you new bands you might like.
They have over 1k employees, surely many of them experience the same easily fixable flaws? My fav theory is, once a company becomes successful and big enough, noone wants to own problems, everyone wants to be part of new exciting projects with a purely positive outlook. And if business is going well, upper management has no reason nor (informal) mandate to get tough and make strong demands.
The fun thing is, my company has jumped on the agile train a while back (nothing wrong with that per se in my opinion), and Spotify was always that cool example "where agile works". Well, no idea what they're doing, but if their KPIs mainly consist of how bad of a UX they get away with, they for sure are very successfully improving their product.
It constantly forgets which episodes of a podcast I've already listened to, or where exactly I've stopped when in the middle of an episode. Their support for downloaded songs at this point seems to be almost non-existent (I stopped using spotify offline quite some time ago as it became too much of an annoyance). If a song I liked is not available anymore but is then available again, I have to like it again. It still can't differentiate between studio and live versions of songs which I would like to filter out (I hope someone corrects me that I'm just too dumb to find the right button). I once got an email that one of the bands I liked released a new album - but it was a band with an identical name (and never heard of before!); apparently, they send out mails by matching band names instead of some unique IDs (no idea how that passed any kind of review; must've been some mashed-together 11-hour release).
... the list could go on and on. The thing is; I really want Spotify to be good, it's what made me stop pirating music, it was just way more convenient. It still is, but it could be so much more, especially given all the cash they're rolling in.
I have encountered situations where I hit PLAY and nothing happens, or I hit PLAY and it says it's playing, but there's no audio, or I hit PLAY and the app says it's unable to play the track. Restarting the app seems to fix most of these issues.
There are also scroll bugs when browsing podcast episodes. The scroll position will just jump to the same spot as you try to scroll down the page, making it impossible to view the episodes in a series.
https://community.spotify.com/t5/Ongoing-Issues/Spotify-won-...