Cloud Classification

Interesting post from Luis Villa over on the rather wonderful Freedom to Tinker concerning cloud computing, and how it may not be all that the forecasts predict. Luis helpfully distinguishes at least four different sorts of cloud here – hosted applications, big data sets (a nice library analogy here), collaborative creation and hosted computing power or storage.

Luis starts though with a link to Richard Stallman‘s interview with the Guardian (that I missed entirely), and some of the comment it generated. One of Stallman’s concerns is the lock-in that outsourcing technology provision will produce, potentially worse than the lock-in associated with proprietary code.

With his commitment to free computing, RMS’s concern is a predictable one, but the warning is worth hearing, with not only students making increasing use of web 2 applications, but schools outsourcing much of their information infrastructure, either as managed services as per BSF or to google via google apps for education, which now, significantly, can provide security and ‘compliance’ tools at a generous discount.

In an ideal world I think we’d love to host all these services ourselves, and the e-safety and data security arguments are strong. However, when google (or even, in threory at least, a BSF provider) can provide these more reliably, with better functionality and for less money, it’s jolly difficult to take a stand and say that in-house solutions are best. The waters get murkier still when one allows for the fact that students are now routinely trusting their data and creativity to the cloud. That said, hosting things in-house, with open source wherever possible, probably results in a deeper understanding of the tech, and the chance to really tailor the solution to the school in a way otherwise impossible; these are things wholly compatible with educational aims.

It’s in Luis’ third cloud type, “services that make creation of new data technically or economically feasible” that the most significant benefits of externally hosted services are obtained, and these are, of course, manifold. By encouraging collaboration outside the institution’s boundaries, the opportunities for students to engage in activities that are most meaningful to them and, crucially, to learn from and alongside those with shared or complimenrary interests, talents or enthusiasms are far, far greater than even within the largest school: something just as relevant for teachers and school leaders as for our learners.

I don’t think it any accident that some significant parts of this collaborative use of the cloud is built on open source software: mediawiki, wordpress, elgg, lamp – and whilst schools don’t get admin priviledges they can impact the development of these tools. It’s my hope that as young people get a taste for the collaborative and creative side of computing, they will come to see that open source development is a clear expression of the same ethos, as it was for Stallman at the start of the GNU project; perhaps he’s worrying unnecesarily.