Johannes Ernst quoted Michael Graves in a recent blog entry: "If it doesn't have a URL, it doesn't exist."
Now, Johannes and Michael are smart guys, so I know there must be some context missing here. However, this comment really sets me off.
There are TONS of things in this world (in fact, almost all things) that don't have URIs that are very real, very important, and are the subject of all sorts of digital transactions. Credit card accounts, my cell phone, my car, my townhome. Nobody has created a URI for them because there hasn't been a need to or because of security issues.
Lets get away from silly sounding statements like this if you really want to make the case that URI-based identity is the way to go (and please stop saying "URL-based identity" because it's ambiguous at best - see "URIs, URLs, and URNs: Clarifications and Recommendations 1.0" ).
In addition, as a "URI-based identity community" have to factor in the cost of getting someone to *create* URIs for all the things we want to be subject to our mechanisms. Most of this cost is comes from users comprehending what a URI is and how it comes to being.
I think geeks think that its really easy, for example, to roll out OpenID authentication and create your own HTTP URIs - but I can definitely tell you almost nobody related to me has the first clue how to do that and doesn't even understand the concept of using a HTTP URI for an identity (yet, perhaps). They understand http uri's for web sites, credit card numbers, phone numbers, IM handles, and email addresses.
In summary: 1) Don't oversell the ubiquity of URIs and 2) recognize that URIs come with costs that this community really needs to own up to and tackle head-on.
Comments