Posts Tagged ‘echo nest’


Mobile collaborative playlisting (a prototype)

March 19, 2010

One of my colleagues at Olin College, Mark Chang, teaches a course on Mobile Applications Development. Instead of a midterm exam, he runs a design contest. We invited Paul Lamere of Echo Nest to campus to talk about their APIs, and Mark’s students had ten days to build an app for the Android that used those tools – kind of like Music Hack Day. Mark was kind enough to ask me to be a judge, so I got to see all the great prototypes they came up with.

While all of the apps were excellent, the one that I thought had the most interesting concept was the DJMixr (by students Miguel Bejar, Rhan Kim, Hyeontaek Oh, and Poorva Singal), an app to make a collaborative playlist for a party. The app would allow guests to add songs to the playlist directly, and also would scrobble song information from the phone’s music player (it would be backed with a streaming on-demand music player, so it would only transfer the song information, not the songs themselves). The Echo Nest back end would be used to interpolate recommendations based on the seed songs, in order to allow smooth transitions between songs to avoid musical whiplash.

While this app is just a prototype, of course, one of the things I find intriguing about this is that it’s a different social model for the music at a party. Rather than having one person be responsible, or at best having people wandering over to a laptop between drinks to maybe add a song or two or change the Pandora station, an application like this would make it possible to have a truly collaboratively generated soundtrack to an event. It seems like there would be some interesting emergent behaviors, like maybe a metastable equilibrium between music for the people who want to dance and those dedicated to indie rock. What do you think?


SanFran MusicTech Summit roundup

May 21, 2009


This year’s SanFran MusicTech Summit had a lot going on. Somewhere north of 600 people turned up, including tech/development people, marketers, and business development people, and I’d say about half of the crowd put up a hand when asked if they were a musician (I suspect there’s considerable overlap between that category and the others).

Unfortunately, with three concurrent tracks, it’s easy to miss a lot of the conference. If you’re interested in knowing more about what went on than what’s in this post, here are some other summaries: CNet, NY Times, Washington Post. And if you think Twitter, and not journalism, is the first rough draft of history,  you can read what people (including me) said about it in the heat of the moment by searching for #sfmusictech.

Some highlights:

Biggest tease: Paul Lamere of Echo Nest demo’ed their extension to Spotify that uses their musical analysis-based technology to create extended Spotify playlists, based on a seed song, for example. It looked great, and you can read more about it at Paul’s blog. Unfortunately, because of the byzantine licensing arrangements (another theme of the conference), Spotify is not yet available in the US. It’s gonna be the future soon, right?

Second-biggest tease: We wrote about Band Metrics, a music analytics service that aggregates data from across the web, a few months ago. The site was demo’ed at the conference. While it’s still in private beta, conference attendees were given an invite code. Unfortunately, it turned out to be only valid for the day, so my attempt to register after the conference was unsuccessful. Stay tuned for more info about the service in the coming weeks (I hope!).

Biggest app north of the border: I talked to Darryl Ballantyne, of Toronto-based LyricFind, and he mentioned that their free iPhone app is one of the most-downloaded in Canada. The company does all the back-end work to make licensed lyrics available (their engine powers, for example) and they’ve recently branched out into consumer products. Worth checking out (and I’m not saying that just because they brought kegs of Canadian beer for the reception).

Biggest fight: By far the most, uh, lively panel I went to was the Monetization: Idealism in Practice panel. Jim Griffin talked about Choruss, which is a proposal to charge colleges a flat per-student fee for access to music, in whatever form (torrents, streaming, purchases, whatever). The school would collect data on how students acquired and listened to music, and use that data to disburse the collected funds to artists (by means which were not fully explained). It was worth going to the panel just to see the Electronic Frontier Foundation (in the person of Fred Von Lohmann, one of their senior staff attorneys) on the same side as the major record labels (Choruss was started by Warner and is backed by several other majors). However, this is still a pretty contentious proposal, especially since the colleges are kind of paying protection money (‘you promise not to sue us, right?’), because it’s not clear how the indies or unsigned musicians will be represented, and because it seems incompatible with other business models. I know that many z=z readers are musicians, associated with a college, or both, and I’d be really interested in what you have to say – does the Choruss model make sense to you? Why or why not?

MP3: Jonathan Coulton – The Future Soon [buy/download]

Image:  San Francisco Music Tech #4 by Flickr user .schill, reposted here under its Creative Commons license.