Not Quite a Blog

(cross-posted to "Challenges of P2P" here)

In the P2P debate, it's often easy to forget that there are important uses of P2P technology other than file-sharing -- uses that do not implicate copyright. While file-sharing focuses on the consumption of the network, other technologies focus on production utilizing the vast computing and innovative resources available at the network's edges. Our own Space Science Lab here at UC Berkeley has made a valuable contribution in this area through the development of distributed computing.

Harnessing the power of tens of thousands of CPUs, the SETI@home project pioneered this technique to analyze vast quantities of radio telescope data for signs of extraterrestrial life. In an effort to allow other researchers to harness the power of distributed computing, those behind SETI@home developed a general-purpose version of their distributed computing software, the Berkeley Open Infrastructure for Network Computing (BOINC). BOINC projects now include climate modeling, protein-folding calculations, gravitational wave detection and improving particle accelerator design in addition to searching for ET.

One such project, climateprediction.net, reported their first results in last week's Nature. As reported by Science's wonderful ScienceNow service in a story called, "Climate Models Heat Up":

Now modeler David Stainforth of the University of Oxford, U.K., and 15 colleagues have pumped-up this approach by utilizing the processing capacity of 26,000 personal computers. Volunteers from the general public contributed the power of their machines to climateprediction.net, allowing the researchers to vary six parameters, several at one time for a whopping 2578 simulations. Most of the simulations predicted a global temperature rise of around 3.4°C, but some ran as hot as 11°C, 2°C warmer than any kind of study before it. Stainforth, whose group publishes its work in the 27 January issue of Nature ("Uncertainty in predictions of the climate response to rising levels of greenhouse gases"), says he can't say how likely the 11°C heat-up is because "we can't yet give a probability for our results."

One can imagine a consumption-based P2P filesharing service that is similarly unhindered by issues of copyright. For example, a filesharing service for only public domain or Creative Commons-licensed works. Of course, there are the open questions of whether or not such a service could be built and whether or not it would sustain. If we suspend reality for a second and imagine that all (or a significant number of) public domain works were available on such a system, I think we can all agree that such a service would be culturally influential and and have a broad impact on our growing information society.