Not any time soon, I think — but it ought to. How many phone numbers do you actually know? Beyond my own, I know maybe half a dozen, the rest are stored in my phone. A quick survey of my friends tells me that I’m not alone in only actually remembering a small handful of numbers. But really, why should we bother?

Back when I was first logging on to the Internet, ICQ was the protocol of choice, another communication system wherein you’re assigned an arbitrary number that you use to uniquely identify yourself and others. ICQ was pretty quickly overtaken by AIM and MSN, and is today essentially dead (though it lives on in some capacity as an addition to the AIM network). MSN and AIM were better than ICQ in two important ways: they let users pick their unique usernames, and they had contact lists that were stored in the cloud, so that users could log in from anywhere and have immediate access to all of their buddies. (I believe that ICQ later implemented this second feature, but it was already too late.)

So what’s preventing phone numbers from being replaced in a similar fashion? Are there any reasons why we would want to avoid such a transition?

Broadband penetration rates are rising, but aren’t yet universal. Once they are, it seems to make a lot of sense to have phone service be another data service that disappears into the data you pay for for internet connectivity. As soon as this happens, the phone is just an application using connection, it just happens to be a physical application. Once the phone is just an application, it becomes pretty trivial to “sign in” to a phone address, much like an instant messaging client. Visiting a friend who has a spare phone in their guest room? Just sign in to their phone with your account, and have your calls routed to your physical location. By the same token, you could sign in to your phone address on your laptop, or on your mobile phone.

Such a transformation is already underway with mobile:

A few weeks ago, a friend told me about a service called iSkoot. iSkoot claims to bring Skype to the mobile phone; essentially, to let users use their data plan to make calls rather than their talk plan (which would be nice to drop) they do this by bridging your Skype account through their dial-in number. It works… kinda. I only have a few friends on Skype, and calling them took a few tries, each of them taking significantly longer to make than a normal phone call. Once I got a connection, the sound was delayed and of pretty low quality. This probably has more to do with the quality of the Bell/Telus data network — all the same, I think iSkoot is on to something.

I can, unfortunately, think of a few barriers that stand in the way of a change like this one. First off is money. If telcos can charge you twice, once for phone service and once for Internet access, they’re naturally going to be resistant to simply dropping the former from their revenue streams.

Second is safety. Whenever there’s a power failure, the phones still work, the same is far from true for internet connectivity. Even if all houses had power-on-lan, it isn’t much help for any devices that access the net through a wireless router, which likely will be powered by the electrical grid.

Third is annoying segmentation. In Canada, everyone uses MSN. In the states, it’s a mix of MSN and AIM. If we were to move to a speech as an application model, there would need to be a single standard. The instant messaging giants have not made any consumer friendly moves to use one standard protocol (though Google gave it the old college try with the Jabber-based GTalk protocol, but MSN and AIM already reigned supreme), I imagine the same problem would happen with phones-as-applications. Given that a phone is only as useful as the number of people you can use it to call, having betamax vs vhs battle every time you tried to call someone would become very tiresome very quickly.

So, for all of the great strides that we could enjoy by disconnecting the telephone system from the telephone system, there seems to be too much in the way of hurdles. Given that Skype can already interface with existing phone systems easily enough, I’m sure that some sort of hack-solution will come about that gives us  legacy support with the added benefits of the fully digital new generation. Unforuntately, even a solution like that seems painfully far away.