Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Incident report: Spotify outage on March 8

engineering.atspotify.com

Post reply on HN
Print view View original

121–130 of 132 posts

Re: Incident report: Spotify outage on March 8

#121

Post by ge96 »

ge96 wrote: post #119
DougMellon wrote: post #96

Earlier quoted context omitted.

I currently own the Sennheiser hd 650's[1] with a Schiit amp[2] and absolutely love the combo. I also have them on a harman curve. You can get the HD 6xx from Massdrop for cheaper and it's the exact same driver[3]. My partner uses the Sennheiser 58x's[4] as well as she doesn't like to use an amp and they are probably the greatest bang for your buck I've come across. [1] https://en-us.sennheiser.com/high-quality-headp…

Thanks that'll probably be a next model up, it's up there in price for me. Will keep it in mind I'll compare the 58x to DT 77 although open backs limit usage for me I think (not have other people hear my music).

Looks like the 58x is better than DT 77 though without using eq

This one of those things drives you insane so many choices

Re: Incident report: Spotify outage on March 8

#122

Post by trutannus »

carstenhag wrote: post #84
trutannus wrote: post #66

Earlier quoted context omitted.

That and the accounts appearing locked, with password reset links not working. For the first few minutes of the outage, I thought I my accounts were what was compromised. Caused genuine concern for a while there since it appeared I had lost control over my Spotify account, and there's credit card data in there.

My better half thought that her account was hacked. So yeah, the logout thing is even worse than not being able to log in. Doesn't help that the error message they have is not very useful (there should be a "yeah it's our fault" message)

For sure. The "invalid password reset link" one really caused worry. It took until I hit the support page to figure out there was an account issue. Probably should have issued an email notice so I could have seen it when I tried to reset my password.

Re: Incident report: Spotify outage on March 8

#123

Post by justinhj »

belval wrote: post #71
justinhj wrote: post #64

Earlier quoted context omitted.

Because of outages and or censorship of content

I probably listen to music on Spotify ~5h a day while I work and I haven't had any downtime ever since I subscribed. As for censorship there was that thing with Joe Rogan, but as far as I know Spotify sided with him so I don't see where the censorship is?

You are in a thread about a Spotify outage. The White House press secretary called for censorship of Joe Rogan and I expect it will happen soon as they find a reason. Personally I cant stand the user interface and many times I am away from a network connection and something I downloaded is not available or Spotify is just stalled trying to connect. Its probably just not a good fit for my use case

Re: Incident report: Spotify outage on March 8

#124

Post by wwweston »

yunohn wrote: post #118
wwweston wrote: post #91

Earlier quoted context omitted.

> I think Napster/Limewire/Kazaa were far greater influences on price point expectations for Music. Unauthorized copies have been a fact of life not only through the dawn of the P2P era but back through the cassette era. And you don't even have to get into the ethics of that in order to understand that none of them ever set price point expectations. Everyone involved in that activity ultimately understood that they w…

> Spotify adopted an economic model marginally different from piracy but with the veneer of a legitimate economic transaction, the pretense of some kind of proper accounting. How does your critique of Spotify not apply to Apple? They lack a free/ads tier, and claim to pay a bit more to artists. Does this fundamentally change the fact that they’re still /streaming/?

/Streaming/ itself isn't the problem. The problem is the /payouts/. Retail revenues were replaced with fractional revenues more suitable for broadcast. That makes life economically more difficult for musicians.

So my critique could apply to Apple in approximately the ratio between their payouts and Spotify's (averaging around $.01 to $.004).

I would also point out that Apple's scheme starts to get into territory that looks reasonable-ish as a replacement for recording retail. Ask yourself how many plays you're likely to get out of a single you buy. If my recall of my iTunes history is any indication, it probably averages out to around 50. $1.00 a track, 70% to artist, that's right around a penny per play.

Of course, there's a short peak (opposite of long tail) of more individually popular tracks that people will play hundreds of times. Those the economics probably works out down in the tenth-of-a-penny range. But even then, retail incentives were better -- front-loading payouts creates a situation where new-music-creation is incentivized more heavily, leading you to be more likely to have more from your favorite artists if there's more to make.

Re: Incident report: Spotify outage on March 8

#125

Post by hk__2 »

rightbyte wrote: post #69
hk__2 wrote: post #43

Earlier quoted context omitted.

According to Last.fm I’ve listened to 156k songs from 16k artists (35k albums) over a little less than 10 years. Last time I bought music it was on iTunes and the lowest price was €0.79 per song. That would be a total of €123k. Over the same period, I gave exactly €1200.93 to Spotify [1]. [1] note: it’s a pure coincidence that this is ~ €10 * 12 months * 10 years; the price has not always been the same.

How do you find so many different songs? Shuffle? You are essentially listening to 40 new songs per day.

Spotify "radios" (playlists that are generated on-the-fly once the thing you were listening to ended), recommendations (Spotify’s weekly playlists) and top 50 playlists, browsing the "similar artists" tab on artists I like, browsing the "discovered on" tab to find playlists. Then, when I like a song from an artist, I usually listen to at least their most recent album.

Last week I found a Russian artist in a playlist I follow and I listened to all their albums. I don’t know how to pronounce their name nor the title of any of their songs, but it was a nice background music for coding since I don’t understand a thing of the language.

Re: Incident report: Spotify outage on March 8

#126

Post by BuildTheRobots »

hiptobecubic wrote: post #107
peterburkimsher wrote: post #106

Earlier quoted context omitted.

It surprises many colleagues and friends that I still use an iPod Video from 2005-6. Life without adverts, week-long battery life, instant track skipping, ratings, playlists (and OTG playlists) make the music experience better than any streaming services, which are really just radio replacements. Embedded systems might not be "cool" any more, but they're repairable, and have other features too (iPod notes, games, con…

You can also do this with your phone. You don't have to stream everything. Just add your files and use an app that can play music locally.

Or something like plex/embly/jellyfin and I can stream any of my home media to my phone/tablet that won't let me have removable storage.

Re: Incident report: Spotify outage on March 8

#127

Post by nemothekid »

wwweston wrote: post #91
nemothekid wrote: post #70

Earlier quoted context omitted.

> No one has done more than they have to create the price point expectations for products like theirs. I think Napster/Limewire/Kazaa were far greater influences on price point expectations for Music. This reads like complete erasure of music distribution from 1999-2011. Not to mention you already have companies like Rdio and Grooveshark that were pretty much squeezed to death by the labels. Yes, Advertising is a poo…

> I think Napster/Limewire/Kazaa were far greater influences on price point expectations for Music. Unauthorized copies have been a fact of life not only through the dawn of the P2P era but back through the cassette era. And you don't even have to get into the ethics of that in order to understand that none of them ever set price point expectations. Everyone involved in that activity ultimately understood that they w…

>Everyone involved in that activity ultimately understood that they were not participating in economically supporting the artist (or indeed, in any economic transaction at all).

I don't believe you can present this decision as independent of piracy and business model. The reason for Spotify's abysmal payouts compared to Apple's has to do with the _business model_. Advertising is an awful way to pay to content when it comes to music especially when you have to placate labels. You are making it seem Spotify just kept the money for themselves, when there simply wasn't that much money to go around.

>Rdio and Grooveshark are weird examples to pull in

I bring them up (and include Pandora) because they had the functionally the same business model; to appease labels by giving them advertising revenue. It didn't work because the revenue wasn't there.

>The primary reason you couldn't have pulled that off in 2008 was economic contraction, not piracy.

Spotify was founded in 2006; and launched in the US in 2011 (the same year Limewire shutdown). You could not have launched a successful, subscription only streaming service the year limewire shutdown. Your chart is a great resource - look how tiny the "On Demand Streaming (Ad supported)" revenue bar is; that is the "money pool" most artists are drawing from when they get paid from Spotify.

To summarize, Spotify's payouts are historically terrible and will continue to be terrible for as long as they continue to support their free product. The advertising market just isn't there, and all those free users depress the pay per streams that Spotify provides. Spotify's "price expectation" was driven by the fact they likely needed to launch with a free ad supported product or they would have never succeeded in the competition with piracy. Apple now enjoys only having to compete with Spotify, and not with Limewire, and doesn't have to offer a free product. With a much larger revenue pool to draw from, per stream, their numbers naturally look better than Spotifys.

The effects of piracy cannot be discounted; even Jobs practically built iTunes and the iPod on the backs of piracy.

Re: Incident report: Spotify outage on March 8

#128

Post by TobyTheDog123 »

eertami wrote: post #49
TobyTheDog123 wrote: post #31

Earlier quoted context omitted.

It is. In my personal experience: - On Mac, Apple Music will just consistently stop playing audio. Force quitting & restarting fixes it. - On iOS, the app is prone to crashing more than any iOS app I've seen. It's not horrible (I very rarely see crashing period), but it's far more than 3rd party apps or even Apple's own set. - On iOS and on a weak connection, Music will not play even if it is downloaded. I suspect it…

>I think way more people can tell the difference between Spotify's audio quality and Apple Music's Lossless/Atmos. I sure could, and it was obvious. But have you blind A/B tested this with a third party? I don't mean to be combative, but the audiophile community is ripe with this kind of thing. People claiming that a speaker cable (or sometimes even a power cable) gives them better quality audio. And then once you bl…

I mean it's obviously purely anecdotal, but I would by no means consider myself an audiophile. I use Airpods Max at my desk and Airpods Pros during my run, and that's about it.

I would love to see someone do this experiment though, maybe I am an audiophile and didn't even know it.

Re: Incident report: Spotify outage on March 8

#129

Post by kactus »

xibo9 wrote: post #47
kactus wrote: post #10

Earlier quoted context omitted.

I switched to Apple Music from Spotify just because I don't like podcasts mixed in with my music, and UI that constantly advertises their podcasts side. The improved sound quality on Apple Music is a major benefit, and I was able to add all my music that isn't in their library (and sync across devices). I also love that upcoming albums that drip feed new tracks aren't categorized as EPs like Spotify does it. I can vi…

Might I recommend you check out the Cider client? I've been using it for a while and its pretty decent.

Oh neat! Been looking for a better Windows client. I'll check it out.

Re: Incident report: Spotify outage on March 8

#130

Post by benatkin »

sususu wrote: post #108
benatkin wrote: post #95

Earlier quoted context omitted.

Logging out on a 401 is pretty common in authentication, because it gives the user an opportunity to sign in again. Maybe the problem resulted in a 401 being returned, when it would have been better to return a 500.

In the web at least it was returning 500, I guess that the desktop/mobile apps make some check in intervals. When the backend returned 500 the app thought that the current login could not be authenticated and logged out everybody

I think some will aggressively log out even on 500s. If you log out 500s that's not the same as logging out failed network requests. The server has to actually send a response, so while it will erroneously log out sometimes, in a reliable api, or one that sends errors in 200s but is otherwise reliable, it won't be so often that a whole lot of users give up. Plus some API middlewares will default to hiding error response codes for security and always send 500 no matter what the upstream response code (this assumes the error codes aren't needed by the devs because they can look at the logs).

I've seen logging out on failed network requests, but that is highly annoying and frequently occurs and typically just a coding mistake.

Post reply on HN