This post is the fifth of a five part series on ‘using the web for documentaries‘, addressing the following points: the embeddedness of society in the internet, the political in the web, the politics of engines, the politics of tools, and the web as an anticipatory medium.

So tools have politics too. Society and internet are closely intertwined; massive amounts of data are put online each day, so Internet is often quite up-to-date. This brings us to the final part: the web as an anticipatory medium.

Let me start with two quick examples in connection with the American elections. When time had come for the presidential nominees to come forward with their candidates for the vice-presidency, those who had been monitoring Wikipedia turned out to know the names of the candidates before they were announced. This was true for Palin as well for Biden.

Another entanglement of the virtual with the real can be found in a study carried out with the Issue Crawler, network location software. A set of web pages concerning a specific issue are entered as starting points. Then all those pages are being crawled, retrieving all their links. Then a method called co-link analysis is applied which, in subsequent iterations of crawling, only retains websites with at least two different sites linking to it. Just like Google, the Issue Crawler sees links as acts of association. This allowed us to track the websites of all democratic and republican presidential candidates since June 2007. The co-link map of all democratic presidential candidates attracted particular attention. It is common knowledge that Obama had a brilliant social networking strategy, where a lot of work was done locally, ‘from the ground’. When you look at subsequent Issue Crawler maps, you will clearly notice Obama had the same tactics online where Obama and web 2.0 websites take over the democratic candidates’ space, long before Obama got chosen to be the democratic nominee for the presidential elections.

The first maps show a strong presence in the liberal blogosphere (the place of gossip, activism, etcetera),

then, as the primaries are approaching, the different candidates become visible on the map,

and then very soon Obama’s strong presence, with sub sites, as well as a number of web 2.0 devices are noticeable.

This trend continues until March and April, when we see that the network is entirely dominated by Obama and Web 2.0 (at that time, Obama was not yet sure of his nomination).

These maps clearly demonstrate Obama’s online tactics: web 2.0 sites, connecting and reflecting the off-line behavior of good social networking.

Remember that in all maps shown the starting points for the crawls, and thus the maps, were the same: the homepage of each democratic candidate. In subsequent crawls, other moments in time, you clearly see obama taking over the network; even before the primaries – before Obama was chosen by the democrats as their presidential nominee!

This simple method, only allowing sites on the map if at least two other sites link to it, unveiled Obama’s tactics at a rather early stage . This method can be seen as a rhetorical tool. It is our angle, or cadrage, of the documentation of online strategies: it is our way of framing the story. Since the Issue Crawler was conceived in 1999 this method has proven solid in finding and representing different actors in network and debates. For instance: NGOs associate readily with companies, but those in turn hardly ever associate with NGOs.

So, the Issue Crawler tells a story from a certain viewpoint, and so has its built in politics, giving it its story telling power.

Journalists also recognized the anticipatory power of the web by regularly looking at Google’s Zeitgeist – an ever updating chart of the terms most searched for by Google’s users. According to an article by NRC handelsblad (November 8, 2008) political journalists regularly checked Google Trends to see which were the most popular terms looked for. Obama turned out to be more popular than McCain (http://www.google.com/trends?q=obama%2C+mccain). Nowadays all journalists have a blog and want to be read, so obviously they wrote more about Obama. He was what people were looking for, and journalists writing about Obama, mentioning him in their title would be read more (in this way having more impact, becoming more polpular, and often also earning more because of increased viewing of their embedded page ads). So tools can influence politics as well. Writing about Obama and Google trends became a vicious circle.

Another great example of the web as an anticipatory medium is Google Flu. Google noticed that when people are feeling sick they often query the web before going to the doctor. Google transformed this data into a prediction of flu epidemics at http://google.org/flutrends where past flu outbreaks are overlaid with this year’s prediction. Google claims they can detect flu outbreaks two weeks before the regular channels (e.g. hospital emergency rooms).

At the start of this presentation I claimed the internet is embedded in society. I then stated that the internet is more than one single medium. We have seen there is now more information available which can be accessed faster, but we should not loose sight of the political in and the politics of engines and tools when doing research with or about the web: all tools, engines and spheres need a specific approach.

Of course not everything is reflected on the web. Probably the internet penetration rates are the clearest example of this. Below, the first infographic depict the countries of the world scaled by the percentage of internet penetration in that country. The second depicts the same but with the percentage, and thus scale, inverted.

To conclude this series: different questions can be asked and answered with different media; they may be indexed by different engines and through different kinds of tools. Just think back at Cassian Harrison’s ‘Britain from Above‘, told with GPS. He used this particular medium for his specific problem: how to show patterns of movement in big geographical areas?