webponce rants

things less interesting than a pigeon walking in a circle.


OpenID

I'd long wondered what the uptake of OpenID would really be. In principle, the concept of having a 'single sign on' concept across the increasing number of online applications one uses as a modern day web user was ideal. Keep your security credentials managed centrally, and authenticate with each service provider when needed. No longer do you need to keep track of all the different username/password combinations, but just a single username, and a password you can change on a regular basis, increasing security across all of your applications.

My worry about OpenID was take up by less-savvy users. One of the strongest commercial arguments for adopting OpenID was that new users would be able to come to your application with lower barriers to entry, no long painful sign up, just drop in your OpenID account, and you're away - but realistically, is my mum going to be aware of OpenID providers? Probably not. I'd happily implement OpenID in all of our commercial projects at de-construct, but many of the audience members are just not bleeding edge enough to have an existing account, and sending them away to setup an OpenID account seems overkill.

However, with recent news that Yahoo! accounts and Blogger.com blogs will be OpenID accounts also, this barrier is lowered yet again. Suddenly, all those flickr users are OpenID enabled. All those bloggers are OpenID enabled, it can't be long before Google accounts are OpenID enabled, whether Microsoft consider Passport to be worthy of OpenID is yet to be seen (although mooted). The next step is education and getting users to see the benefits. Most of the benefits for those non bleeding users will be immediate, we don't need to shout about them, just implement them, and make the user's life easier. Let's see if we can get our next authentication system using OpenID, and i'll let you know how we get on.