CircleID

I was reading David Weinberger's reports on how the New York Times is planning on tackling its "link rot" problem where articles slip behind the pay-wall. Part of their solution appears to be to replace articles with their summaries.

As usual, this got me thinking about telephony. Why don't phone calls and callers have URIs or URLs? (Side note: URL is something you can actually fetch over the Net, URI is just a unique name.)

Let's take addressing the endpoints first.

Obviously, ENUM is one way of "Internetising" the phone number address space. Furthermore, RFC 2806 "URLs for Telephone Calls" has a backways-looking way of creating URLs for PSTN end points, as "tel:", "fax:" and "modem:".

But does ENUM really go far enough? Does it have the right balance between the "DNS" part of URIs (e.g. "foo.com") with the "path" part? In other words, why do we need to put all the transient stuff into a format-constrained and centralised ENUM/DNS database? Why should I have to gain anyone else's blessing as to what records I store in the directory? Why don't you fax me at enum://www.martingeddes.com/fax://? (Your syntax may vary.) Why bundle the core identities, delegation, access control and personas into a single database?

It's not a criticism; just an observation. I don't have an answer.

ENUM also is rather "telco-centric" in that it tries to record the connectivity protocols to reach me. Although not necessarily technically constrained, in practice it doesn't seem to play well with the bigger identity picture of federation, or meta-data about who I am and who is willing to back those assertions as being true. On the other hand, not being all-things-to-all-people may be a feature and not a bug; that said, I'm not sure if it's a good solution to a badly-framed problem.

In other words, there's probably still work to do in the establishment of URIs for identities and personas.

The "conversation" itself also lacks references. I can't link to the phone call we had last week—neither the metadata nor the media; the dump of chat conversations from my Skype client doesn't include URLs. No URLs for voicemails, or for emails.

URLs are the network extension of a basic Unix philosophy. Loosely speaking, everything in Unix is a file or a sequence of characters. That makes for simple building blocks to create new things. Call it "stupid computing", if you like. Complex IBM operating systems with enforced journaling, sharing and transactionality (think: QoS, smart networks) lost out. URLs do the same for distributed network resources as Unix files do for single computers.

Solutions based on URLs (or plaintext files) tend to be less "efficient" than custom-designed solutions, but more resilient in the face of change. (Sound familiar?) I casually note the rapid rise of URL-friendly blogs and wikis, and the relative obscurity of vertically integrated collaboration tools like Groove.

I saw a good example of "URLization" from Joi Ito the other day. Product managers at various companies are being encouraged to use Technorati to track commentary on their products. If your product doesn't have a clear, unique URI, then you're in trouble. Will Peerio objects have URLs? Is GNUP really the DNS for distributed persistent objects?

This all brings me to my big question: on the Internet, if something doesn't have a URL, does it really exist? Or has it just disappeared into the analog memory hole, only existing as a memory in the brains of the humans it passed through?

We need to crack the URL problem for real-time communications like IM and phone calls. Until we do, the IP revolution will be stalled re-inventing solutions to elemental problems within endless vertical silos.

Written by Martin Geddes, Consultant, Chief Analyst. Visit the blog maintained by Martin Geddes here.

Related topics: DNS, Enum, P2P, Telecom