Social music



[Some semi-intelligent notes will go in here eventually ... t.b.c. ...]

:: Technologies ::

  • COllaborative Filtering Engine
    “Collaborative filtering is the process of computing personalized recommendations by finding users with similar taste. Such techniques have been successfully used to recommend newsgroup articles, movies, music, and e-commerce products. Collaborative filtering is most known for its use on popular e-commerce sites such as Amazon.com or NetFlix.com. … This free server (including source code) allows anyone to easily set up a recommendation system. It should run on any platform that supports Java 1.4.”
    » http://eecs.oregonstate.edu/iis/CoFE/
  • Open Fingerprint Architecture
    “MusicDNS and the Open Fingerprint Architecture provide a system for identifying a piece of music with nothing more than the sound of the piece itself. MusicDNS is the largest single dataset of acoustic fingerprints in the world with more than 16 million individual tracks identified and growing DAILY! With the Open Fingerprint client-code, tracks can be identified consistently against the MusicDNS dataset, and new tracks are easily added. The Open Fingerprint is deployed world-wide–and now the client source code is available under Open Source license.” Underpinning technology for MusicIP (see below).
    » http://www.musicdns.org
  • foafing the music
    “a music recommender system, based on user’s profile. This means that, depending on what you like, what you listen to, where you live, etc, you get personalized music recommendations.”
    » http://foafing-the-music.iua.upf.edu
  • XML Shareable Playlist Format (XSPF)
    An open source XML-based data format for sharing playlists that can be played on a personal computer or portable device. In the same way that any user on any computer can open any web page, XSPF is intended to provide portability for playlists. Underpinning technology for WebJay (see below).
    » http://xspf.org
  • Audioscrobbler
    “The Audioscrobbler system is a database that tracks listening habits and calculates relationships and recommendations based on the music people listen to. After installing an Audioscrobbler plugin for your media player (eg iTunes, Winamp, XMMS) the name of every song you listen to is sent to the Audioscrobbler server and added to your music profile. The Audioscrobbler system powers our main site, Last.fm, as well as exposing data via webservices so other projects can make interesting things from the data and recommendations we provide. A massive part of sucessfully recording listening habits is maintaining a healthy set of metadata describing albums and song information. We are working with the MusicBrainz.org dataset, and in future hope to offer more data based on their music GUIDs.” The system behind last.fm (see below).
    » http://www.audioscrobbler.net
  • Musicmatch Music Discovery Engine
    “The Musicmatch Music Discovery Engine is fueled by the music tastes of more than 30 million Musicmatch Jukebox users who voluntarily share their listening habits with us. These music fans play more than 1 billion songs each month and we use this information to understand the relationships between thousands of artists around the world as well as the daily popularity level of hundreds of thousands of songs in the music universe. Think of it as the library of all the world’s music organized by 30 million librarians.” Underpinning technology for Musicmatch Jukebox (see below).
    » http://www.musicmatch.com
  • tunA
    “tunA is a mobile wireless application that allows users to share their music locally through handheld devices. Users can “tune in” to other nearby tunA music players and listen to what someone else is listening to. Developed on iPaqs and connected via 802.11b in ad-hoc mode, the application displays a list of people using tunA that are in range, gives access to their profile and playlist information, and enables synchronized peer-to-peer audio streaming.”

:: Internet-based music discovery & recommendation apps ::

  • iTunes
    No apologies for putting this one first, not only because so many of the apps listed below integrate with iTunes. [Mac; Windows; mobile]
  • last.fm
    The best. “Last.fm feeds each user an individual stream from their vast mp3 collection, and the user is allowed to fast forward or listen to tracks, or press the ‘I love this’ button. These three simple choices, over time form a musical profile for each listener. These profiles are then associated and cross-referenced to build up maps of affiliated tastes, and to some extent predict what music users might like, although they may never have heard of it – but (like the amazon feature) people who liked listening to Bad Manners also listen to these other four obscure punk ska bands you’ve never heard of.” (quoted from a review at The University of Openness) API profile at ProgrammableWeb. [Linux; Mac; Windows]
  • Imeem
    [Linux; Mac; Windows]
    » http://www.imeem.com
  • Indy
    Pure genius. “Indy uses an advanced collaborative filtering system to predict what kind of music you’ll enjoy hearing. As you rate songs, Indy finds out what you do and don’t like. It compares your preferences with the ratings of all the other Indy users. For example, if you rate a song highly, and another user also likes the same song, Indy guesses that you’d probably like other music that they enjoyed. As you rate more songs, Indy will gets better and better at picking songs that you’ll really enjoy.” [Linux; Mac; Windows]
    » http://indy.tv
  • Pandora
    Though not strictly a ’social music’ app, this is ingenious: an online juke box that, from an initial user specification of a song or an artist, will build your personal playlist by proposing (thanks to the ‘Music Genome Project’) similar songs or artists that you either approve or disapprove. The downside at the present time (24th March 2006) is that, for legal reasons, it’s currently available only to US residents: “At this time we are only licensed to offer Pandora music services to residents of the United States. Audio streaming regulations differ from country to country, and we are working on acquiring the proper licenses so we can legally offer Pandora outside of the United States. We require your zip code to confirm that you are a resident of the United States.” [Mac; Windows]
    » http://www.pandora.com
  • MediaMaster
    “MediaMaster is revolutionizing the way music fans experience their digital music. By hosting and storing consumers’ music collections online, MediaMaster makes it easy to stream your music to any computer, a broad range of mobile phones, and internet-connected living room devices. The idea sprang from our frustration at being unable to have full access to our large music collections away from our computers. Since we’re techno-heads who had already launched 3 companies, we decided to build a solution. We combined with some industry insiders to create a world class team to pull it all together. Now, we–and you–can listen anytime, anywhere.” [Browser-based; Facebook integration]
    » http://www.mediamaster.com
  • Audiri
    “Our goal is to help you find music you like, easily. This means organizing music conveniently into genres, advanced search functionality, and a radio suggest system that can optionally play songs based on what our computers think you will like. Our charts are easily sortable by popularity (believe it or not a feature lacking on some audio sites) We also track plays in several different ways.” Music (streaming radio and some downloads) and video. Open source and free. [Linux; Mac; Windows. Requires Javascript and Flash]
  • GenieLab
    “Think of your favorite musical artist or band and search for them using the search tool … Give your favorite artists a rating of “5″. Or, click on Most Popular Artists and let the system guide you through what other users have said they like. Rating popular artists allows the system to quickly ascertain your listening tastes. Once you’ve rated enough artists the system will automatically start making predictions. When you’re ready to discover some totally new music, click on “Get Recommendations”!” [Web-based -- Linux; Mac; Windows]
  • MusicMobs / Mobster
    API profile at ProgrammableWeb. [Mac; Windows]
  • MyStrands
    “MyStrands?�� mission is to understand musical taste in order to help people discover new music and get greater enjoyment out of the music they already love. MyStrands helps us discover new music that we might never think of on our own, such as linking one artist in a genre to another, somebody famous to someone less so but equally deserving. By interconnecting listeners around the world, we build lists that connect us to one another. And in so doing, we not only expand our musical horizons but obtain greater enjoyment from the music we already love.” API profile at ProgrammableWeb. [Mac; Windows]
  • MusicIP
    [Linux; Mac; Windows]
  • WebJay
    Yahoo-owned web-based playlist service. API profile at ProgrammableWeb. [Linux; Mac; Windows; mobile devices]
  • Soundflavor
    “… the world’s most advanced song-level search, navigation, and recommendation technology. The Soundflavor engine is designed to help consumers discover, sample, and purchase music as playlists (the digital music equivalent of mix tapes) by suggesting individual songs?��or playlists?�� which match their unique tastes in music. The technology marries objective (and unbiased) metadata, artificial intelligence, and listener ratings to yield unparalleled accuracy.” [Web-based -- Linux; Mac; Windows]
  • MOG
    “Creating a MOG lets you share your musical side like nothing else. Our ‘MOG-O-MATIC’ application creates a basic MOG page for you, so your friends can see your music collection and the artists, albums and songs you actually listen to. As you collect and listen to music, MOG-O-MATIC keeps score for the world to see. Customizing and making your page an extension of your musical soul is made easy. And soon, you will even be able to add MOG to your blog or MySpace page. Just register for MOG, then download MOG-O-MATIC and get your page up and running. Then explore what some other moggers are into. MOG is FREE.” [Mac; Windows]
  • liveplasma
    [Mac; Windows]
  • Yahoo LaunchCast
    Personalised streaming radio service based on initial user preferences, collaborative filtering and on-going ratings. [Mac; Windows]
  • Musicmatch Jukebox
    “Input any artist name and the Music Discovery Engine will display the 50-100 most closely related artists – far more than you’ll find anywhere else. You’ll also get a list of each artist’s tracks in popularity order, so when you discover an artist you’ve never heard of we’ll take you right to the best music from that artist. And if you choose, Musicmatch will even give you daily recommendations for music so you don’t have to work to find great music that suits you.” [Windows]
  • Mercora IMRadio
    Peer-to-peer streaming of user radio stations. [Windows]
  • The Echo Nest
    Not yet live, but one can sign up as a beta tester [Mac; Windows; mobile]

:: Music production ::