Brad Fitzpatrick has been dropping some interesting mind bombs since starting at Google. First it was the Social Graph API recently followed by PubSubHubbub (which I need to write about one of these days) and most recently the WebFinger protocol. The underlying theme in all of these ideas is creating an open infrastructure for simplifying the tasks that are common to social networking media sites and thus improving the user experience.

The core idea behind WebFinger is excerpted below from the project site

If I give you my email address today, you can't do anything with it except email me. I can't attach public metadata to my email address to give you more information. WebFinger is about making email addresses more valuable, by letting people attach public metadata to them. That metadata might include:

  • public profile data
  • pointer to identity provider (e.g. OpenID server)
  • a public key
  • other services used by that email address (e.g. Flickr, Picasa, Smugmug, Twitter, Facebook, and usernames for each)
  • a URL to an avatar
  • profile data (nickname, full name, etc)
  • whether the email address is also a JID, or explicitly declare that it's NOT an email, and ONLY a JID, or any combination to disambiguate all the addresses that look like
  • or even a public declaration that the email address doesn't have public metadata, but has a pointer to an endpoint that, provided authentication, will tell you some protected metadata, depending on who you authenticate as.

... but rather than fight about the exact contents

The way this is written makes it sound like this would be a useful service for end users but I think that is misleading. If you want to find out about someone you’re best of plugging their name into a search decision engine like Bing or the people search of a site like Facebook which should give you a similar or better experience today without deploying any new infrastructure on the Web.

Where I find WebFinger to be interesting is in simplifying a lot of the common workflows that exist on the Social Web today. For example, I’ve often criticized Twitter for using the hand picked Suggested User’s List as the primary way of suggesting who you should follow instead of your social graph from a social networking site like Facebook or MySpace. However when you look at their Find People on Other Networks page it is clear that this would end up being an intimidating user experience if they listed all of the potential sources of social graphs on that page (i.e. IM services, email address books, social networking sites, etc) then asked the user to pick which ones they use.

On the other hand, if there was a way for Twitter to know which sites I belong to just from the email address I used to signup, then there is a much smoother user experience that is possible.   

This is a fairly boring and mundane piece of Social Web plumbing when you think about it but the ramifications if it takes off could be very powerful. Imagine what direction Twitter would have taken if it used your real social graph to suggest friends to you instead of the S.U.L. as one example. 


Saturday, August 15, 2009 9:14:10 PM (GMT Daylight Time, UTC+01:00)
Could using using i-names (XRIs) accomplish the same thing?
Sunday, August 16, 2009 3:25:24 PM (GMT Daylight Time, UTC+01:00)
If you want to find out about someone you’re best off plugging their name into a search decision engine like Bing or the people search of a site like Facebook...

Says the man whose name is rare enough that he can count on showing up on the first page of such a search. We're not all so lucky.
David Young
Monday, August 17, 2009 1:34:24 PM (GMT Daylight Time, UTC+01:00)
So the search approach to naming is broken, as David points out. Search barely works for people with unique names who are widely published on the internet. For less famous people, search is useless. I've tried to find friends on Facebook, and given up in frustration after paging through thousands of "John Smith"s. Facebook recently introduced profile URLs, and it wasn't by accident.

In fact, we already use WebFinger on a localised basis - when I add someone of Flickr or Facebook or Twitter, more often than not, it's by searching for their email address. Sometimes it's done behind the scenes, with email address import tools, and sometimes it's done by hand, but it's a well-established practice.

The more interesting side effect of WebFinger is that it enables federation. If I want to add someone to Twitter, Bob can tell me that he's @examplebob, and I can go to and add him. Now, if I want to add someone from Identica or Facebook to my Twitter stream, I'm stuck. Obviously there are a whole raft of problems preventing this use case, but WebFinger gives us a clear way forward: I can ask someone for their email address, and away we go: @examplebob becomes, and now I don't have to worry about which site my friends use to post photos or little text updates.
Tuesday, August 18, 2009 8:11:22 PM (GMT Daylight Time, UTC+01:00)
If there's one thing I appreciate about Facebook, LinkedIn and their kind, it's that they shield people from my email. I don't want to ever give anyone my email, because I want to be able to turf it if I need to, at which point, people can still find me on Facebook etc. But, it's true, having an uncommon name is a mixed blessing. WebFinger seems like a good idea, but it also sounds kind of like it's grooming people for the password anti-pattern. We should be telling people "Don't give away your email, don't give away your email password..."
Wednesday, August 19, 2009 4:33:33 PM (GMT Daylight Time, UTC+01:00)
Hmm...interesting, but at the same time complex topic
Comments are closed.