In economics, an object’s liquidity is described as how easily it can be converted into currency without loss of value. I’d like to propose that the internet makes use of a similar concept in how information is distributed and exchanged. Information liquidity.
Here’s a pre-internet example before we dive in. Say you’ve written a book, and now you want people to read it. To do that, you need a publisher that will first print a large quantity of copies of your book, and then distribute those copies to locations that place your book in close proximity to the largest quantity of people (book stores and libraries). Next, potential readers have to be made aware of the existence of your book, as it’s not enough for it to just be available. There are thousands of books in even the smallest book store, and people aren’t just going to magically know that your book has been added to the collection. Publicity or advertising in one form or another will be necessary.
If the currency of the internet is viewership (and certainly other possibilities exist, but that’s outside the scope here), I would describe the liquidity of a site in terms of how easily it can generate viewership. A site’s liquidity is determined by how accessible it is from the user’s current location. One factor is by the quantity of links that are directed to that location across the internet – the probability of any user ending up at a given site increases as more links to it are made. Another factor of liquidity is through the minimum number of links that must be traversed before reaching a given site.
Described numerically, a site’s liquidity is the inverse of the number of links required to reach it. For users already familiar with a site, its liquidity is naturally 1 because they are likely to have its address memorized. As such, the analysis here is not concerned with content already familiar to users, but that content which is unknown to users. For sites which have no links or have not been indexed, their liquidity would be 0. Everything else falls somewhere in between.
Without search engines indexing much of the internet, a great deal of content would be mostly inaccessible except through very specific avenues. With search engines, however, the minimum distance can be as short as just one or two links, depending on the search criteria and accuracy. Considered like this, Google is the world’s largest information liquidator. Google also provides temporary information liquidity through advertisements, which serve both to increase the number of links to a site while also reducing the average distance from a user’s location to a site. The flip side to this is that search engines aren’t a destination. Ignoring Google’s other services, the probability that a user will traverse from google.com to not-google.com is nearly 100%.
What about social networks, then? They provide information liquidity, but certainly not to the same extent that a search engine does. Social networks aren’t crawling the web for content to index; they’re limited only to what its users are sharing through their service. The liquidity they offer also has a very short expiration date. Content has at most half a day of direct exposure, and long-term accessibility is mostly out of the question (especially given that Facebook and Twitter prevent Google from indexing their content). Still, social networks consistently generate large quantities of content with minimal user input, while also providing a lot of temporary liquidity for off-site content. They manage to be destinations as well as gateways.
The last example I’d like to touch on is video games. It may seem a case of apples and oranges next to social networks and search engines, but they represent one of the purest forms of destination on the internet. Video games decrease the liquidity of nearly all information that exists outside their networks. It’s difficult for two players in different video games to communicate with one another; video games are effectively islands of interaction. Problems exist for intracommunity engagement, as well; few video games support robust linking mechanisms such as might be found in traditional instant messaging – they just aren’t the ideal place to be sharing articles or videos. That said, services like Steam restore some of this liquidity by providing a cross-game mechanism for communication. This may also help to explain the wide appeal of MMOs, as they provide the satisfaction of playing a video game while limiting the information (and social) isolation that is commonly associated with video games.
The next concept in this scheme I’m looking to develop is traversal probability. I recognize that measuring user-link distance is insufficient to describe liquidity, and this is for two reasons. First, not all links are equal. The top result from a search is vastly more accessible than those that appear on the following pages, to provide just one example. Factors such as reputability, clarity, and visual prominence come into play. Second, the identity of the user greatly influences which type of links end up having the greatest appeal. Male users are going to have different behavior than female users. Danish users are going to have different behavior than Brazilian users.
If I’m on the right track, I think this is what I was attempting to measure in my previous graph.