How Personal Clouds are Different — Subject-oriented, not service-oriented

This is obviously heresy. For years, there has been broad agreement across the entire IT industry that “services” (in the sense of software-as-a-service, APIs etc) are the fundamental building blocks to create things, and that every piece of software that isn’t already service-oriented will either die or have to be re-architected immediately to avoid irrelevance.

The trouble with service orientation is that it leads to data fragmentation. For example, on the consumer web, I might use the Flickr service for photo sharing, the blogger service for blogging, the gmail service for e-mail and so forth. How many usernames and passwords to such services am I supposed to remember? And worse, how am I supposed to remember where I posted the picture from my friend’s wedding last year? Was it flickr or Facebook or my personal blog?

Service-orientation is great for those people who build services. (Hmm, that’s kind of obvious.) It is not so great for those people who have to use them because of fragmentation: of data, of login information, of user experience.

Enter personal clouds. Personal clouds are the opposite, in that they are subject-oriented, i.e. they are built around the subject they deal with. My personal cloud has everything in it that relates to me. My pool‘s personal cloud has everything in it that relates to my pool. My community’s personal cloud has everything related to my community. This is a structure that makes sense to me, the user.

Sorry, techies, but it is more important that the structure makes sense to the user than that it is easier for the techies. I’m sure one can bring service-oriented services into subject-centric personal clouds without going back to fragmentation. Because subject-orientation is so much better for the users, and that’s another reason why we need Personal Clouds.