Archives for category: Openweb

Yesterday’s OpenID User Experience Summit (hosted by Facebook) has got together a lot of people interested in building a better user experience for OpenID.

If the process succeeds we’ll be able to have a

world where it won’t matter if you’re on Facebook, MySpace or Microsoft Live – where Yahoo’s content is read by Friendfeed, and your MySpace music preferences are matched with friends in Facebook so that you all can coordinate events on Eventful that you heard about on someone’s TypePad blog and was Twittered about on Seesmic and later indexed and searched by Google.

I’ve been able to follow just a bit of the initial presentations on the UStream but Plaxo’s John McCrea has a detailed live blogging post, moreover all the presentations are online at slideshare (plus here are the UStream recorded sessions)

The first thing I find interesting are the results of the Plaxo/Google hybrid OpenID/OAuth experiment. The stunning 92% rate of click through success shows that the open protocols can be made more effective by implementing a better user experience. Moreover it shows that the users, if given a good enough experience, will not fear to follow a path that’s different from the antipattern they have been (mis)educated to use.

The Plaxo/Google process is awesome if you think about the amount of interaction that is taking place within those two clicks

  • the user is signs up for a new service
  • the service is veryfies his email address
  • the user is grants (scoped) access to the service to a set of his own data
  • the service accesses that data thus importing it and pre-populating the new account

It’s a win-win as this shows how something may be made at the same time

  • safer and more useful for the user (who should not give away his password or let absolute access to his account) and
  • more effective for the service providers (more reliable user data collection, possibly more user signups)

On the downside of this approach is that it will not be easy to scale this beyond the single provider. I hope it doesn’t ends up in a proliferation of single-provider buttons (for somthing that may be implemented with the same undelying technology).

This is a treat without cooperation among the interested parties but looking at the meeting and how the big-cos and many of the smaller players are talking to each other there is hope that it will happen for real.

So… news of the day (yesterday actually) is that Facebook is getting on board in the OpenID foundation (this is following the annoucement a while back about Paypal doing the same).
Commenters are beginning to recognize that OpenID is actually gaining more momentum than most would have tought and the adoption that many big online services are showing is helping to build a larger mindshare also among the less geeky users.

One critique that I hear a lot is that the way the bigger players are approaching OpenID, basically turning every one of their accounts into an OpenID for their users is not really helping the open web since it’s like every site giving out an email address everybody can send to but just accepting email from their own email addresses. This is true, but here lies also a great opportunity for the smaller players. If you are a startup you should be happy, as what’s happening may mean that you can have a lot more potential users with an easier path to try out your service (got a google account? no need to signup!).
Ideally one’s identity online should not be tied to a specific service, but more likely each user will have one main service which represents his own main “self” online. That will be the main service they use online, the one they are spending more time on. It may be a Google account or a Facebook one or a Myspace identity or their wordpress.com hosted blog, but we can be almost sure that it will not be your shiny new little service.

It’s a great thing that now most of these “main” identities have become portable (thanks to OpenID and Facebook Connect) that is usable elsewhere, just being able to tap into the great number of people that will already have one of the hub-provided identities

You should start right now to plan and execute to adopt OpenID as a consumer, adding also Facebook Connect in the mix as long as they are separate (hopefully they’ll be interoperable in the future).

I don’t want provide a new identity just to be able to tell who you are, thus I, for sure, will try to do this with the services I’m working on.

P.S. I specially like this quote from Chris Messina I read on the ReadWriteWeb:

user authentication is like a credit card. You don’t go to a restaurant because they accept credit cards, you go because they have good food. To take that analogy a step further, it is good that every restaurant lets you pay for your food with any of the major credit card vendors