Spotify Is the Coolest Music Service You Can't Use
wired.com
wired.com
Seriously, though, Spotify is great and I've been a subscriber for a while. Its coverage isn't 100% but is close enough for anything within a wide gamut of "popular." On the negative side, Spotify has caused me to hardly buy any music in the last year and is saving me a ton of money... which isn't going to the artists whose music I'd have bought. Could that model survive long-term?
On the positive side, Spotify has caused me to not pirate any music in the last year and introduced me to lot of bands who I have went on to see live.
* it's not "any song by any artist" (although it does pretty well). Recent releases are generally missing; certain artists (like Bob Dylan) and labels have opted out entirely.
* Spotify has negotiated a deal with the collection agencies to get its music catalog. If the agencies decide to renegotiate, they could essentially kill Spotify at will (like what nearly happened to Pandora, and forced them to block non-US listeners).
* the service doesn't "support itself" - last year it lost £16 million [1]. While I'm sure they have plans to reach breakeven, until they do they're dependent in principle on their ability to raise new capital.
* the "big four" major European record labels are investors and shareholders in Spotify, between them holding nearly 20% of the company according to some sources [2]. They probably (speculation) have board seats. In other words, the labels have significant control over Spotify's direction and decisions.
* the popularity of Spotify is a highly visible PR win for the music industry's political campaign. The EU and its member states are currently debating things like copyright reform and how to stimulate an IP-centric economy. The record labels, of course, lobby to keep the law exactly how it is, but with stronger rights enforcement measures (preferably in civil as well as criminal cases) and longer copyright terms. Entrepreneurs are calling for things like a compulsory music license for online broadcasting, meaning music startups could experiment with new business models with predictable costs, and without the massive existential risk of having to individually negotiate terms with the collection agencies. Spotify means the music industry gets to say "look, music startups are doing great!", hiding the fact that most prospective music startups won't ever have the chance to get the kind of licensing deal Spotify got by handing over equity to the labels.
I'm not meaning to bash on Spotify as a service - it's brilliant. But it's not an indicator that the music industry "finally gets it". The music industry is just as screwed and stupid as it's always been, and their backing of Spotify is so politicised it's hard to see it as a good sign.
[1] http://eu.techcrunch.com/2010/11/22/spotify-had-a-16-66m-los...
[2] http://techcrunch.com/2009/08/07/this-is-quite-possibly-the-...
I understand MOG doesn't have a free version and that's the utopian dream of Spotify (so says Wired) --- but Spotify bled money out their noses last year. They were hundreds of millions of dollars in the red. I can't speak to MOG's profitability (I don't have info for them), but I guess as someone who lives in the States where people get sued over music, I look at Spotify and then at competing services like MOG and just sort of shrug my shoulders over $5/mo to get unlimited music onto all of my devices.
Sure if grooveshark wasn't for profit then maybe it'd be justifiable, but they profit off a product they don't own or have the permission to, it's lame and entirely immoral. It's limewire but with the content stored in one place... they must have some damn good lawyers.
And it has almost everything I've ever searched for on it. These days it is starting to become more likely for obscure, independently-distributed music to be there than not to be there.
The data structure it puts your local music metadata index into is performant enough to search terabytes of mp3s without getting bogged down. No other music software I've tried can do this.
And it streams in ogg. What more do you want.
(Or if it does, then the UI needs fixing to make it discoverable)
This data is expert-compiled and fairly comprehensive, but there is no statistically compiled data a-la last.fm. I imagine they are working on this functionality.
You can also use the last.fm integration to get recommendations off last.fm then listen to them on spotify I guess.
Smart recommendations, however, are not the experience they're looking for. Spotify seem to be aiming for actual human recommendations. They're focusing much more on the social stuff, both in marketing and engineering hours, than they are on trying to tell you what music to listen to.
Oh right. That's not available in the free version or listed in the quick version comparison (though it is listed in the full comparison) which is why I'd not heard of it.
>Smart recommendations, however, are not the experience they're looking for. Spotify seem to be aiming for actual human recommendations. They're focusing much more on the social stuff, both in marketing and engineering hours, than they are on trying to tell you what music to listen to.
If I want that stuff, I don't get it using a computer; I talk to people.
I'm coming to the conclusion that here - as with damn near everything - I'm "Not The Target Market(TM)" :'(. However it does seem that their target market is vast, so good luck to them.
The idea that Young(TM) people who post a lot of stuff on each others facebook profiles and the like don't talk to each other is a myth.
They post the link, then next time they see the person in real life they have a conversation about the link. Having both given the link their full attention independently. When you're talking to someone IRL there is unlikely to be a suitable computer to hand, especially if the link points to some music.
YMMV. I use Spotify a fair bit, despite having a fairly crappy UI (not that it's too bad, just annoying in places), but there's no way in hell I'd pay for it unless at least one of the following things were true:
1) The premium version cost just £5 a month - The £10 it actually costs would cover a new album each month, which is a permanent asset worth more to me than being able to listen to their selection for a limited period. There is a £5 a month subscription, but it doesn't have mobile or offline mode, so it's near useless. Ironically, I might not feel that way if the premium option simply didn't exist, but putting them side-by-side looks like naked profiteering.
2) I could search for something and expect at least an 85% chance of finding it. The current 50%-ish is nowhere near good enough to pay that much for. I listen to a fair bit of European techno/rock/metal/industrial and those seem to be particularly weak areas. If you can have a conversation with a random person in a pub about a band, there's a good chance they have it I guess, but even then there are a lot of surprising gaps. For a random example, my last search was for something (anything) by Arcade Fire, but no joy.
As you say, it is technically very well designed - very responsive, and even without offline mode it buffers well enough that the frequent dropouts in my crappy ADSL connection haven't caused any problems I can recall. I just feel that it's overpriced for such a limited selection.
I think the rave/electronica scene must be hotter than average on uploading their stuff to aggregators. It is starting to become standard for the artist-owned labels to be on there. I even see a good number of 'bedroom' producers without any record deal turning up now.
My real message, which was obfuscated by the snark in my original post, was that it's worth checking out for the user experience alone -- in an age where we're all out building web apps, it's pretty cool to see a new thick desktop client that is just so good.