The percloud is my proposal for an easy to use personal cloud, that is for a feasible alternative for the masses to Facebook, Gmail, and all similar centralized Web services and their privacy and data ownership issues. Last week I put online a 10 slides summary of how the percloud should work. This post answers a specific question that I’ve been asked many times by readers of those slides(*):
hey, wait a minute: isn’t this percloud you’re talking about just another Linux distribution? Why?
Here is my answer:
Yes, the percloud is “just another Gnu/Linux distribution” and yes, there already are hundreds such bundles of Free Software around.
The problem, that is the rationale for the percloud, is that as far as I know none of them addresses the needs I’m talking about. Sure, there already are many good distributions for desktops, and many others for servers. But the first category, by definition, doesn’t work as a cloud (that is as something that you can access and use from the Internet, no matter what computer, tablet or smartphone is besides you in any given moment).
The second category, instead, is already used to build and run countless cloud services, but never of the right size. Debian or Centos, to name just the distributions I’m more familiar with, are perfect as very flexible, very general purpose, very scalable Internet platforms for expert administrators.
The percloud, instead, is something different. The average Internet user doesn’t need nor want very flexible, very scalable solutions, only stuff ready for his or her individual activities.
So the percloud will be, or so I hope at least, the first member of a new class of Gnu/Linux distributions: systems that include all and only the Free Software needed to provide email, blogging, social networking and more to one user only. Systems that package, harden, lock and preconfigure all that software so that it is ready to do that job, and nothing else, with the smallest possible amount of initial configuration (which non-expert users could always delegate to a third party they chose and trust!).
So, the percloud is indeed “just another Gnu/Linux distribution” in the sense that it would “just” take an existing distribution and “rearrange” the already existing software it contains in a deeply different way, adding the smallest possible amount of new code. And this is good, because it is exactly what makes the percloud a feasible, realistic proposal.
Being yet another Gnu/Linux distribution means that the percloud cannot solve all the problems that the current Internet has, but at the same time it can give millions of Internet users with only the skills to manage a Gmail or Facebook account something of comparable ease of use, soon, with a relatively little development effort. Without waiting for a whole different Internet, or different politics to manage it.
That’s why it is good that the percloud is “only another Linux distribution”.
(*) and I’m publishing it here instead of the percloud website simply to allow readers to comment it. I’ll add comments to per-cloud.com as soon as possible, but not now.