Google and Yahoo! Try to Make Facebook's Advances Irrelevant

The battle for the social web has been playing out over the past few months but much of it has been invisible to those that are less technical. Today, Google announced that they will begin providing limited access to an API for an OpenID identity provider. This means Google users will be able to login to sites that support OpenID with their Google accounts.

This is a significant announcement for Google and for the open web. While I could attempt to place some sort of arbitrary divider between the open web and the social web, for discussion purposes, the social web is working to make the internet more open. As such both are substantially integrated and can be used interchangeably.

How Does Google OpenID Compete With Facebook?

The battle over single sign on is a significant one. If you haven’t been following OpenID and the single sign on trend over the past couple years, here’s a brief summary: users are finding it hard to remember the logins for every site they register for. As such, companies are racing to provide services that make it easy for users to login with their regular email address.

Facebook is preparing to launch their Connect service to the masses, making it possible for users to register for a site by using their Facebook account and without disclosing any personally identifiable information.
The Pros
There are some clear benefits from using Facebook Connect. For one, websites get access to a user’s news feed and the ability to virally distribute content and user activities through that feed. Second, companies get the “benefit” of placing a Facebook login button on their site. Why is this beneficial? Well, Facebook is rapidly becoming one of the most recognized brands on the web.

The Cons
For any company, using Facebook Connect doesn’t solve all your problems. The primary downside of using Facebook Connect is that you don’t get access to personally identifiable information of that user. I want to have access to a user’s email address so I can contact them in the future. Unfortunately Facebook prevents that. If you want to read more on this check out my article from July about Facebook Connect as OpenID without email.

There Can Only Be One Login

-Single Login Screenshot-So I really haven’t answered my last question which was: how does Google OpenID compete with Facebook? Well it’s in an abstract sort of way. When you go to register for a site, you are only going to register with one account. Soon enough, you will be able to select from a number of sites that you specify as the center of your identity. As pictured in the image to the left, this is how a theoretical registration form would look.

No longer will you have to enter all your information into fields, instead, that information will come from your identity provider. OpenID is supposed to tie your identity back to a URL but Google has implemented their own version in typical Google style. The point being here is not to debate Google’s implementation of OpenID though, instead to illustrate that your identity can be tied to external accounts.

We’re In it For the Long Haul

In November of last year, I wrote that the email would become the center of social networks. One year later we are seeing this happen as Google and Yahoo! implement new services for the open web. Yesterday Yahoo! announced their open platform which includes a single sign on-like feature and today Google has announced their own version of OpenID support.

This is just the beginning though and ultimately, much of this will require user adoption. I’m guessing that there will soon be a registration “widget” similar to the way that Disqus handles my comments, another party will handle my registrations. As long has I get to have my own database filled with user data, that’s all that matters.

It’s going to take some time but I’d imagine in the next 12 months there will be a huge shift toward a centralized registration system that everybody can use and developers can quickly implement. It’s exciting to see the big players getting involved and while each step can be criticized for its imperfect implementation, I’d assert that this is once again a big step in the right direction.