Traversing contexts and Facebook

Jeff Clavier of Software Only has written an excellent post on Facebook's Q&A session last night at Stanford's ETL program.  Beyond the sheer jaw-dropping statistical aspects of Facebook's tremendous growth (summarized below), I would net out a good chunk of their magic as making the process of what I call "traversing contexts" friction-free.

By traversing contexts, what I mean is that I can build a profile of myself -- where I go to school, my former high school, what classes I am taking, what books I like, etc., and within a single click, I have visibility to a bunch of people that have common interests. 

Think about the simplicity and power of that model (relative to search) for social activities.  Or better yet, let me give you a simple use case. 

Let's say that I am taking Abnormal Psychology at my university.  Traversing from my profile, I am one click away from knowing everyone that is taking the same class.  Maybe I am looking for a study partner.  Maybe I just want to see if that girl that I have been looking at knows anyone that I know who can make an introduction.

So I click around some more because one of the profiles that has come up looks "interesting."  Now this click reveals that this woman is "married" to someone of the same sex.  Hmm.  Did not see that coming.  (Sidenote: married is Facebook lingo in the same way that the word "poking" someone is lingo in the Facebook realm -- intentionally ambiguous.)

But it also reveals that this person is a member of the "Derek Zoolander Center for Kids Who Can't Read Good...And Want to Learn to Do Stuff Good Too" Group, which clearly has to be a joke. 

Another click reveals that this is a pseudo group created by lovers of the movie Zoolander, and the profile page of the group has a picture of Ben Stiller, the group has officers, discussion threads, related groups and...1938 members.  Let me repeat that: 1938 members!  Did that many people even SEE Zoolander?

So what have I learned?  In three clicks, I went from the context of my classmates to the context of random but novel student to the context of affilitation with a mindset (by membership in an irreverent group whose purpose is to accessorize my persona).  And I have spent 15-20 minutes traversing these contexts and having a good time in the process.

Before you dismiss this all as just a silly fad, consider the following.  One, the workflow of moving up, down or sideways into different realms of meaning (contexts) is totally friction free in this model. 

Two, it encourages the short attention span generational mindset.  I have 15 minutes, maybe I will see what is new or interesting out there, and post a comment or two. 

Three, it is synchronous with the stumble upon aspect that makes life such an adventure.  I can start anywhere and have no idea where I will end up.  And four, it is fun.  Facebook does not take itself too seriously, all the while providing real utility to its user base.

What is that utility?  Well, it's providing a space for meaningful conversations via a mechanism for "lookup and hookup" according to whatever context jumps your boat.  Increasingly that activity is for coalescing around specific events -- dorm parties, local activities, private get togethers, and now photo sharing. 

No question that the context of a known demographic and filtration around meaningful boundaries like "UC Berkeley" are very powerful, and part of the Facebook's unfair advantage.  But another takeaway is that Facebook proves that you don't need to glut a service with features for features sake or have heavily stylized interfaces to look professional. 

No, the power here is speed, simplicity and friction free traversing of contexts, facilitating conservations of all types.

The raw numbers:

• 5M+ registered users
• Coverage of 45% of US colleges (a total of 2,000 - representing 8M students)
• 80% penetration among students of colleges that are on the platform
• 10th most visited Internet site in the US
• 5.5B page/views a month (230M page/views a day)
• 8.5M unique visitors
• Signing 20,000 new users a day
• Repeat usage: daily 70%, weekly 85%, monthly 93% – can you think of another site that sees 93% of its registered users coming back every month ?

There is much goodness in Facebook to emulate.

Comments

[John "Z-Bo" Zabroski]

I can relate to what you are saying.

Finding ways to quickly "traverse contexts" is what I am trying to convince my computer science friends at college to get interested in. I describe the importance of "traversing contexts" similarly to how you do except I don't use that phrase.

When I read your thoughts on "traversing contexts," I archived back to Chris Anderson's Long Tail web log (which is how I initially ventured to thenetworkgarden): http://www.thelongtail.com/the_long_tail/2005/07/brand_response.html#comments

You had some interesting comments there too and can be combined rather easily to your latest thoughts (or so I feel).

Personally, I believe the biggest change on the Internet is the change from Hierarchical to Hierarchy-less groups of metadata/data/information/knowledge.

Data is not organized into group nestings any longer, but instead organized into interfacing networks which can tap directly into a social conscience (which business owners are beginning to think of ways to monetize).

The power of crawling a network is almost always more powerful than crawling a hierarchy, whether the crawling reference is theological, philosophical or mathematical.

So I think what sites like Facebook are doing, and you may want to see emulated by vSocial, is pinpointing what their interfacing networks are. Just my $0.02

[Brad Webb]

John,

There's no question that this kind of IA is what makes sense. The rigidity of past IAs (Yahoo's old Directory for example) is not only high-friction, but it's also an extremely unnatural way to traverse topics. People learn; they're not taught, so removing the stumble factor because of the belief that you've engineered the best, new way for people to find (search for, really, because taxonomies aren't analagous to Finding) makes the user experience worse. But I think that's one of those "been there, done that" lessons that lets us all move forward and look at mistakes from the past, and correct them with better IA and SA.

Lots of what we've put into vSocial was/is driven off of the "social intelligence" psychology, there's a bunch of very musty, academia-type books on the subject, and a few gems, but something definately worth looking into, because it really ties together information architecture, software design/architecture, user experience and the human experience, which is what we *all* should be shooting for. Well, that, and "jacking in," ala Gibson, but maybe that's the ubergeek in me talking. ;)

[John "Z-Bo" Zabroski]

Brad,

You mentioned so many things: Information Architecture, Software Architecture, the Cyberpunk novella genre, vSocial being a portmanteau of sociology and cognitive psychology, and you even mentioned traversing topics.

It is pretty clear you excel beyond my interests. I know so little about science fiction that when someone recently asked me what character I was playing as in the new Star Wars game, I shrugged and said, "You tell me."

First, let me point out that traversing contexts and traversing topics carries different meanings. So, when you say, "it's an extremely unnatural way to traverse topics," well, of course. I don't believe people traverse topics. I believe that people "traverse contexts" (as Mark has said it).

Isn't improving the speed at which people can traverse contexts sort of how www.EyeQ.tv works? People are trained to read groups of words together. Mathematically, this is analogous to a vector search that builds collocation tables to crawl. Except, here, the collocation tables are temporal and fleeting.

Second, if you want to tie together information architecture, software architecture, user experience and human experience, then I consider there are only three "Doctors" on the subject:

Jef Raskin (The Humane Interface: New Directions for Designing Interactive Systems)

Alistair Cockburn (Agile Software Development)

Henry Petroski (The Evolution of Useful Things: How Everyday Artifacts--From Forks and Pins To Paper Clips and Zippers--Came To Be As They Are)

Just about everything else is a pretender in my eyes. The only other person who interests me is William Whyte, whose book The Organizational Man was recommended by my favorite college professor.

Finally, I do not know for certain that "there's no question that this kind of [Information Architecture] is what makes sense." I question everything, not to be annoying, but because answers come from questions. Lately, I am noticing Google and other search engines producing less relevant searches. I predicted this would happen and described this as the effect of anti-filters (borrowing from the phrase anti-patterns). The problem is pretty straight-forward but no one really has turned it into an area of research, which has disappointed me. (Here's a link to the only effort I have seen so far: http://www.informatik.uni-freiburg.de/~cziegler/papers/A4-Thesis.pdf "Towards Decentralized Recommender Systems") Anyway, just because you do things in a way that makes more sense, does not mean problems won't creep up. Solutions only create more problems, it's got to be the fundamental rule of the universe. No sociology or psychology, just plain physics.

[Brad Webb]

John,

No question there are problems, if I had the definitive answer, I wouldn't be reading up on the subject still. ;) The real problem with my original comment was that I was crossing over between architect and engineer. Topics and contexts from an engineering point of view are one in the same, a datapoint/node. Perhaps my rhetoric could be pulled back a few notches, and be left at "this type of IA is the best we've seen so far" -- obviously if this was the penultimate solution, I'd be out of a job. =)

Also, the Search vs Find debate is one that I take beyond semantics. As a developer, it only makes sense that as Google (or any search engine) has been around for several years now, and has more content than anyone would ever need to *find*, that searching would become less and less effective. That, and how many "Hacks for Google" publications, newsletters and books can there be to lessen the effectiveness of the engine itself.

That being said; yes, my statements were much more absolute than I intended them to be.