I recently read a thought provoking article on Anil Dash’s blog titled “The Web We Lost”


A quote that gets at the crux of the post:

This isn’t some standard polemic about “those stupid walled-garden networks are bad!” I know that Facebook and Twitter and Pinterest and LinkedIn and the rest are great sites, and they give their users a lot of value. They’re amazing achievements, from a pure software perspective. But they’re based on a few assumptions that aren’t necessarily correct. The primary fallacy that underpins many of their mistakes is that user flexibility and control necessarily lead to a user experience complexity that hurts growth. And the second, more grave fallacy, is the thinking that exerting extreme control over users is the best way to maximize the profitability and sustainability of their networks.

I was somewhat taken aback by this article. I’ve been on the internet since 1993, and I think of it basically as follows:

  1. The internet is the set of all machines that are interconnected and communicating via TCP/IP
  2. The web is a set of socket applications that communicate via HTTP, generally on socket port 80.
  3. Social media sites like Facebook, Twitter, etc are web applications. Clever web applications, with cool Javascript-powered interfaces, but they’re essentially glorified BBSs.

So the idea that “the internet” has devolved into a few “walled garden” sites with ever-increasing control over their user base is jarring to me. But then I thought about it from the perspective of a perhaps-techologically-naive user who just got online a few years ago, whose experience of the intenet has been mostly through interacting with Facebook, Twitter, and mobile apps. That world looks a lot different from the one I see.

The third point (“social media sites are just web applications”) )is what makes me hopeful that the current trend toward “walled gardens” is just an unfortunate and self-limiting trend. That is:

  1. Facebook, Twitter, et al are wildly successful, but they are successful because they have large user bases. Actually, scratch that — “user base” is marketing-speak. They have communities of users.
  2. They didn’t always have communities; they started very small, and grew organically.
  3. If they get sufficiently awful, they will be replaced.
  4. They may be replaced simply because they fall out of favor. I can’t recall any great crime that MySpace committed, but it’s a ghost town now.

Basically, the way the internet is implemented means that (almost) anyone can put up a new web site, and the accessibility of free development tools and low-cost hosting services means that the barriers to standing up a full-fledged server are quite low.

So while I do think the current trend toward Balkanization of internet sites is unfortunate, I don’t think it’s permanent. Thinking back, it seems that the communication patterns have oscillated back and forth between centralization and distribution:

  • In the beginning, there were dial-up BBSs.These were the primitive ancestors of today’s bulletin boards. They required that the user dial up to a specific phone number and establish a direct modem-to-modem connection to log on. Very centralized.
  • Once the internet came along and people were able to exchange email, they began to set up mailing lists for discussion of special-interest issues. Very distributed, peer to peer communication.
  • At some point, USENET came along. (This is where I came in, back in 1993). USENET was (and remains) a collection of newsgroups and servers that use the NNTP protocol. USENET is basically a grand unified BBS containing a set of hierarchical “newsgroups”, each one devoted to a specific topic. For example, “comp.lang.programming.java” is for discussion of the Java programming language. This seemed like the end-point of the evolutionary process at the time. In the early days, the signal-to-noise ratio was very high, and if you were interested in a programming language, chances were that the author(s) of the language would be active participants in the newsgroup. Though the USENET system is a distributed one, logically it’s centralized.
  • As the internet grew, USENET became a victim of its own success. Groups got noisier, and eventually the core participants began to start their own mailing lists, so that they could carry on discussions without being subject to random noise and trolls. About this time, Web BBS sites began to catch on, for similar reasons. Back to peer-to-peer and distributed.
  • The next step of evolution was the “Web 2.0” generation of rich apps like Facebook, Twitter, etc. Which brings us to today.

So, while the trend right now is toward centralization, I don’t think that this is the end of history, either. Underneath it all, TCP/IP is still there, and HTTP is still there, and hyperlinks work just the same as they ever did. It’s Google’s counter-argument to accusations of monopoly. “If you don’t want to use Google, just type another URL into your browser.”

“The web we lost” isn’t lost at all; it’s still down there, working the same as it always has. All that needs to be done is to reclaim it.