Mitch Kapor has two posts about Microsoft’s purchase of Groove Networks. Mitch was founder of Lotus and more recently the Open Source Foundation and was also the first outside investor in Groove, so he has several good insights into the software and how / whether Groove would be as Open Source if it were done today. The quote that got my attention the most was this one though:
With the prospect of open source-based server capabilities of all kinds becoming more like the electrical power and distribution system, universally available on demand in whatever amount is needed, a whole class of objections to client-server architectures such as dependence on non-local, unreliable and inconvenient infrastructure diminishes. Groove’s peer-to-peer architecture performs uniquely well in areas where the telecom infrastructure is weak, such as conflict-ridden areas of the Middle East and Asia where both military and humanitarian aid groups have deployed it successfully, but this alone is a niche application.
I like peer-to-peer technology for a whole host of reasons, but I think he’s right that the infrastructure arguments for P2P are (and always have been, IMO) weak except in niche applications (bandwidth saving via BitTorrent being the notable exception). But the driver for P2P technology hasn’t been about limiting the effects of technical infrastructure failure — it’s been about limiting active efforts by an adversary to stop communications. The adversary might be an opposing army, an oppressive government or the RIAA, but the goal is the same — and that’s a need hasn’t changed in the past 10 years.