Lucas Gonze’s analysis of client-side remixing is spot on. Summary: client-side remixing is to precise syncrhonization as HTML is to precise layout. If you don’t need precision, enjoy.

I see three limits to client-side remixing. All can be raised:

• Bad client software. It either doesn’t work or barely works and you need a very keen eye to find a gratis download amongst enticements to buy a super-premium subscription version (cf RealPlayer).
• Lack of expressivity. Remixers don’t just overlay source segments, they also apply various effects to the same.
• Streaming-like experience. In order to obtain a smooth client-side remix playback you (actually your client, this is a subset of “bad client software”) will have to download most of the needed source content first. I often have a bad experience with playing-while-downloading of individual songs and videos over the net, nevermind many coordinated sources.

I suspect that with excellent client software the client-side remix experience could be very good. Lack of expressivity seems like the toughest hurdle to me. However, if said excellent client software can download and run code safely … effectlets?

Video games seem like a highly constrained example of what client-side remixing could do. They pull off co-ordinating lots of different source media (sometimes all local, but that’s beside the point) with code quite well, versus hardcoding different sources into a single stream at the point of production.

However, anytime in the near future using client-side remixing to evade those who would prevent distribution of The Grey Album and the like is pointless. Client-side remixing isn’t up to the task, and you can still download the album from the web after weeks of brouhaha, nevermind P2P networks.

Memory augmentation: cc-metadata client-side remixing [1] [2]