Once Again: Wrong OpenID/Facebook Connect Comparisons

Admittedly, user experience hasn’t been OpenID‘s real strength so far. Usually the redirects between the OpenID Provider and the Relying Party are criticized as being confusing to users. This might be true, but the problem is identified now and will be solved soon, hopefully.

While I understand the criticism, it’s strange to see comparisons between OpenID and Facebook Connect like the one on the Kwiqq blog. By the way, Kwiqq is a British company building customized social networks.

The author shows some screenshots of logins via OpenID (the Relying Party is Dopplr) and Facebook Connect (the Relying Party is Brightkite). He concludes that there are four steps required until he is logged in with OpenID while there are only two steps required with Facebook Connect. Oddly, he has to log in to his OpenID Provider while he is already logged in to Facebook so he omitts that step in the second scenario.

Comparing apples and oranges? First, Dopplr’s login screen could be different. So there was no need for step one, an extra click deciding between username/password and OpenID login. Let’s compare Dopplr with UserVoice.

Dopplr:

dopplr-login

UserVoice:

uservoice-login

As you can see there is no extra step required! It just depends on the website’s implementation.

If I am logged in to my OpenID Provider already, like the author is to Facebook, there is also no additional step required. I am logged in to UserVoice with just two steps like he his to Brightkite using Facebook Connect.

Of course, companies can manage logins the way they want. If they prefer Facebook Connect they should use it. But blaming bad user experience completely on the protocol – OpenID – is wrong. Relying Parties are responsible for it as well.

Tags: , , , , , , ,

View Comments

  1. Raj Anand’s avatar

    Hi Carsten, thanks for your thoughts on the blog and referring to the Kwiqq Blog :) I can see how Facebook Connect and OpenID is like comparing apple and oranges. Firstly I should clarify I’m not anti-OpenID, quiet the contrary really.

    The blog post is based on my observations. I specifically pointed out about the 2 (Facebook Connect) to 4 screens (OpenID) comment:

    (I’m assuming that I’m Logged into Facebook, reasonable as most users tend to be logged into Facebook)

    As you rightly mentioned it varies from user to user !

    1. Carsten Pötter’s avatar

      Raj, I guess many users will be logged in to Facebook before they enter a Facebook Connect enabled site. Though the same probably applies to Google and Yahoo! customers who are able to use their accounts as OpenIDs.

      Also logging in to the OpenID Provider first before starting surfing is a good practice anyway: the risk of phishing is reduced and it provides a better user experience.

      That’s why I thought it was wrong comparing both systems the way you did it.

      1. Raj Anand’s avatar

        “Also logging in to the OpenID Provider….”

        That’s a really good point. I agree with you !

  2. Cody’s avatar

    I am the founder of a new social networking website http://OurConvo.com In short, it gives users a public platform to discuss topics in an simple manner with the rest of the world and gives groups a one page platform to meet. I think it will also be the first large scale site to launch with Facebook Connect. I welcome your feedback on the site and if you like it, I would be thrilled for you to run an article about it.

    Here is our launch press release: http://ourconvo.com/about/pr/07april09/

    Please let me know if you have any suggestions or questions about the site.

    1. Carsten Pötter’s avatar

      Cody, I just had a first look at the site. Though, I don’t think I will cover it. It just doesn’t seem to fit the content of this blog. Sorry!

      1. Cody’s avatar

        Okay, well thanks for checking it out. Love any of your feedback.

Comments are now closed.

blog comments powered by Disqus

Page optimized by WP Minify WordPress Plugin