On the Project VRM telecon today, we had an excellent discussion regarding “PDS” terms and my blog post last Sunday (Out with “Personal Data Store”, In with “Personal Data Service”). Iain Henderson of Mydex made the point that the key advantage of the term “personal data store” is that it describes a place where an individual can keep the data that does NOT exist anywhere else on the net, but which must exist someplace under the user’s control in order for them to be able to share it the same way they can control the sharing of data which does NOT reside in their personal data store.
So my previous blog post needs to be retitled. It’s not “out” with the term “personal data store” at all. Rather it is “in” (as in “integration”) with the terms “personal data service”, “personal data server”, and “personal data ecosystem”.
I took the action item to coordinate the efforts of volunteers on the call (Iain, Paul Trevithick of the Higgins Project, Doc Searls of ProjectVRM, and anyone else who wants to volunteer via the ProjectVRM mailing list) to formalize the terminology and turn it into Wikipedia entries. We’re going to start by writing up draft Wikipedia entries on the Project VRM wiki. To kick that off, I’m starting here with a first cut on the relationships between these terms:
- A personal data service is an online service that enables individuals to store and share of data over which they personally have control. Note that this does not say anything about where this data is located, i.e., where it physically resides on the network. (It is intentially agnostic about this.)
- A personal data server is the server responsible for providing personal data service (for a single individual, or a group, or a whole community). Once again, this term does not imply anything about who operates this server, or where the data resides. Like an email server, a PDS server could be operated by the individual his/herself, by a third-party service provider, by an community, by a government, etc. It also does not imply anything about the technology or language it uses, nor the protocols it speaks (though it does suggest there will be standardized protocols, just like email servers or web servers).
- A personal data store is a physical repository for data over which an individual exerts access control. Again, the term itself does not imply where such a store lives on the network (i.e., in the cloud, on a local device, in a smart card, on a SIM, etc.). It also does not imply that the only way to access a personal data store is through a personal data service or a personal data server. However it does imply a natural relationship between them, i.e., an obvious interface for a personal data store (wherever it is located) is through a personal data service (wherever it is located), and one obvious location for a personal data store is at/behind/under/inside (choose your location metaphor) a personal data server.
- The personal data ecosystem is the universe of personal data services, servers, and stores together with the applications, networks, and services that rely on them to deliver their value proposition. See the Personal Data Ecosystem Focus Area of Identity Commons for more about how this ecosystem might evolve.
What do you think? While you should feel free to comment here, even better would be to join the ProjectVRM mailing list and discuss these terms there, then help us refine the draft Wikipedia entries for which there are now placeholder pages on the ProjectVRM wiki.