Re: The Structural Problems of the Internet for Cultural Policy

Christian Sandvig is a well known professor at the University of Illinois and dabbles in, among many things, systems architectures and structural impacts on culture (and vice versa). Or, in not so many words, the digital divide. I had the chance to read a draft of his paper The Structural Problems of the Internet for Cultural Policy on the train ride back in from Chicago this break. Written originally in 2003 the article points out a commonly overlooked issue in the way the internet is setup. While it enables many people to logon and create content, rich corporate and or government agencies are better able to push this content for users. Though speeds are always getting faster, if Joe Shmoe in California invents a device that defies gravity and then posts his work up on the internet he could easily get overwhelmed by the sheer number of requests and visitors to his home server. But if the New York Times places up a news story about his anti-gravity machine they can distribute it to web cache servers all across the country for localized (faster) access. The concept is called multicasting, and while the web ought to be knit with it, its complicated and confused roots don’t place it at the center of architecture.

Or that is, until now. After giving the article some thought – I’m fairly certain this isn’t true anymore. I mean sure the poor grandmother might still not have any more access to technologies to get her latest knitting strategies on the internet than before, but on the whole I think the masses of America have found an increased ability to be content producers. And what’s more is that they can be immune to the Slashdot effect. How? Oh well you already know my answer. Facebook. Okay and many more sites to be honest. Social Networking Services, YouTube, Wikipedia, blogs, Google Pages, and more all offer powerful places you can post your inventions, insights, and material on the web and not have to worry very much about issues of access and web hosts being overwhelmed. Furthermore, companies and schools often times have web cache servers, devices solely dedicated to hosting the most popular files people on the network request from the web. And beyond this methods of data sharing like Torrenting use multicasting as a fundamental backbone. In the span of only 2, maybe 3 years, Christian’s insight was rendered irrelevant. This is not to say that structural features don’t influence culture and access to information, as they most certainly still do, but that norms are changing pretty fast. In any other field something published in 2003 would be pretty darned recent and yet for us Internet scholars that date lies on the transition to Web 2.0 and the material becomes contested (and otherwise mostly invalid outside of examining history).

And to top it off I just got an email from my boss over at RSC (the University Housing technology division) explaining that the campus will be taking down the web cache servers soon. Why? Our bandwidth is so much greater than before that they actually degrade performance more than they assist it. So much technology is placed into the complicated exit routers that analysis has determined it would be faster for nearly everyone on campus to just let people stream directly to the web.

I’m sure Christian has updated his perspective in light of this, given places like Facebook and YouTube are essentially predicated on viral propagation. Perhaps there are issues in regards to what people are able to post here or what is recognized as popular. More to think about, I just thought it would be good to toss up my impressions. Give me your thoughts.