"We should distrust any elaborately
planned, centrally deployed, and carefully developed business system or
process. Successful systems and processes will be agile and dynamically
adaptive; they'll grow and evolve as needed over time."
Here's a quick blueprint for a grassroots music dissemination network
which works on music files that are linkable and freely available on
the Web.
I want:
An aggregated display of recently logged songs by people I've subscribed to.
A nice, big, green "PLAY" button somewhere that lets me play whatever songs are linked to in my aggregator.
An "I LIKE THIS" button that lets me indicate that I like the
song that's currently playing, pushing its URL into my personal
musiclog (the reBlog idea).
A matchmaking engine that uses my published preferences to point me to other musiclogs I might like
Given the above, I could fire up my aggregator in the morning, start
playing the songs it has collected, reblog whatever I like throughout
the day without going out of my way, and once in a while visit the
matchmaker to find good new DJs to subscribe to.
Implementation thoughts, small pieces loosely joined-style:
Single-page aggregators like Blogdigger Groups, Rollup or simply Bloglines are perfectly able to handle display if the syndication feeds' formatting is appropriate.
"PLAY" can be implemented by way of Alf's M3U bookmarklet service, which grabs links to .mp3s from an HTML page and builds a playlist from them. (Try it e.g. here.)
"I LIKE THIS" button - I don't know of any plugins that log preferences into a syndication feed. The audioscrobbler plugin is closest, and logs into RSS
- or anyways it used to - but it logs everything you play. (I don't
like everything I play, though publishing this can be useful too.) But -
if there were a HTML-based player thingy out there that took a playlist
and simply opened songs one after the other in a browser window, a del.icio.us bookmarklet would do the trick and close the publish/subscribe loop.
Matchmaking could be implemented à la Audioscrobbler. (blogmatcher.com similarly recommends weblogs based on links out of a given weblog and seems to work pretty well.)
