The Last.fm team writes about the music service's engineering, business announcements, outages, and data tools like a MapReduce bash script.
HN links it for the BashReduce script, which one commenter favors over Hadoop, and for company announcements such as radio subscriptions and outages.
I have some suggestions. Now, I'm a bit of an audiophile and also I'm willing to deal with amounts of jank that others find intolerable. YMMV.
I use Deezer as a music service. It's $15/month for HiFi comparable to Tidal's HiFi subscription. Tidal, however, has DRM that does not allow it to play on Linux as a webapp nor does it have a Linux desktop app. Tidal's out. Now, here's the interesting thing: I never use Deezer to listen to music.
- I use Deemix Server (https://download.deemix.app/server/) to serve a webapp on my media server. This plugs into my Deezer account and I use it to download FLAC quality audio into my media server.
- My media server, of course, runs Plex (https://www.plex.tv/). I have a lifetime pass and it's totally worth it, but keep in mind it is required for this setup.
- Locally on my (phone|laptop|desktop) I use Plexamp (https://plexamp.com/). Plexamp is a beautiful high quality player with no bells and whistles but does one thing really well and that's play my music.
- Since this is all offline playing downloaded files, I use last.fm which plugs into both Deezer (https://blog.last.fm/2012/01/06/scrobble-with-deezer) and Plex (https://www.plex.tv/blog/plex-media-server-v0-9-12-5-a-tasty-update/) for recommendations, tracking listening, and connecting with friends. Honorable mention to MusicButler (musicbutler.io) which used to be free and simply notifies you of new releases for a "Release Radar" experience.
All this combines to be a less seamless…
Some more info:
This is from Last.fm developers, esp. RJ (Richard Jones), mxcl (Max Howell; the guy behind Homebrew) and others.
The current source is written in Erlang. https://github.com/RJ/playdar-core
There is some old abandoned source written in C++:
https://github.com/RJ/playdar
I found this while searching for the Boffin source code. Something about Boffin:
* http://blog.last.fm/2009/03/11/introducing-boffin-lastfms-music-knowledge-meets-your-mp3-collection
* http://www.last.fm/group/Audioscrobbler+Beta/forum/30705/_/510180
And in the abandoned C++ repo, there is something about Boffin (haven't really looked much into it yet):
https://github.com/RJ/playdar/tree/master/resolvers/boffin
A good list of hadoop alternatives: http://www.quora.com/What-are-some-promising-open-source-alternatives-to-Hadoop-MapReduce-for-map-reduce
my personal favorite is BashReduce (~120 lines shell script vs ~600k lines of java code in hadoop): http://blog.last.fm/2009/04/06/mapreduce-bash-script
If you're in bioinformatics you might be interested in this talk on handling ridiculous amounts of data (PyCon 2011): http://blip.tv/pycon-us-videos-2009-2010-2011/pycon-2011-handling-ridiculous-amounts-of-data-with-probabilistic-data-structures-4899047
They published a story they got from a source, which turned out to be (mostly) wrong.
References:
http://www.techcrunch.com/2009/02/20/did-lastfm-just-hand-over-user-listening-data-to-the-riaa/
http://blog.last.fm/2009/02/23/techcrunch-are-full-of-shit
http://techcrunch.com/2009/05/22/deny-this-lastfm/
I don't see how that is Arrington's "need to interject himself into the stories"
It's the former, sort of... See the comment on the post titled "Joost, HodgeStar, et al": (I can't find a permalink)
"These are the countries in which we have the most resources to support an ad sales organization, which is how we earn money to pay artists and labels for their music."
Edit: Found the permalink: http://blog.last.fm/2009/03/24/lastfm-radio-announcement#c007521
Since the originally submitted link was deleted from last.fm's forums, here's the replacement, and a link to their blog:
http://blog.last.fm/2011/07/18/library-and-streaming-services-outage
http://www.last.fm/forum/21713/_/756231