In preparation for a panel at the MIT Enterprise Forum I summarized my thoughts about walled-garden social networks. It came back around the other day in conversation with Avi Bryant, who — like a great many people — is appreciating the refinements Facebook brings to various protocols that work less smoothly (if at all) on today’s open Net. He reflected our exchange into a blog post where he also asks: “How do we Facebookize the open Web?”

As Avi notes in his posting, I think Gary McGraw is right when he says: “People keep asking me to join the LinkedIn network, but I’m already part of a network, it’s called the Internet.” It’s also obviously true that the walled gardens are petri dishes growing cultures that will merge with or (some believe) recreate the open Web. So although I don’t spend a lot of time in the walled gardens, I do visit them in order to learn about the cultures growing there.

Last night, I noticed that I’m one of four 1974 graduates of my high school who appear in Facebook. That’s fewer than 1% of the 600 grads. This year, 302 grads appear in Facebook. That’s 50% of the graduating class. Here’s a nice picture of the bimodal distribution over time:

In 2012, what percentage of that year’s class do we think will be participating in Facebook?

While we ponder our answers, I’d like to digress for a moment on the data supporting the chart, and on the chart itself.

Although I thought there were about 600 students in my graduating class, I couldn’t remember for sure. And I had no idea how the student population might have changed over the years. How can you check these facts? Here’s the Google query I ran:

“cheltenham high school” “student population” 100..5000

The third term exploits a powerful but little-known and rarely-used feature called numeric range search. In this case I picked a lower limit that I was sure was smaller than the population of any one grade, and an upper limit that I was sure was higher than the population of the whole school. The query finds pages with both of the exact phrases (the quotation of which is another underutilized feature of search) plus a number in the given range. From the result set, I culled two documents.

First was the curriculum vitae of Joseph Cifelli, which confirmed that in my era “the high school student population was divided into three houses of about 600 students each.” That’s 1800 total, and there were three grades — 10, 11, 12 — so 600 per grade sounds right.

Second was a page on a real estate information site which, though frustratingly undated, I presume to be reasonably current. It gives the student population as 1706. That’s close enough to 1800 for my purposes.

The chart itself is the first I’ve attempted using Excel 2007. The process feels very different from earlier versions, but I’m no expert in this area so I can’t make a detailed comparison. I do have a question for Excel wizards, though. In following Edward Tufte’s recommendation to subtract ink wherever possible, I was able to get Excel to remove almost all of the unnecessary cruft: grid lines, tick marks. But there were still a bunch of zeros that were unnecessarily reported out on the long tail of the data labels. I wound up taking a picture of the chart and then using Paint to remove those, in order to achieve what I think is an admirably clean and spare result. Is there a way to get Excel to do that directly?

Anyway, back to our question: In 2011 will Facebook’s penetration among CHS grads approach totality? Beats me. That’ll depend on the Facebookization of the Web, but also on the Webification of Facebook.

Avi’s post has examples of the former:

  • “a smart feed, which aggregates and filters all of my subscriptions in a holistic way”
  • “an API which allows access both to your blogroll data and to your smart feed…so that if I post and tag a photo of A on my blog, anyone subscribed to A’s blog – even if they have no idea who I am and aren’t subscribed to me – will get an item in their feed about it.”

Here’s an example of the latter. Facebook invites me to manage streams of photos and events. But I have other ways to manage streams of photos (e.g., Flickr) and events (e.g., Eventful). Why not enable me to hook into them?

The process of diffusion can flow in both directions, I guess, and I hope that it will.