There is some irony in John Battelle's account in his book on The Search (2006) of how Google established the success of its search algorithms in and on the Web and how that may have contributed largely to what we are now accustomed to call the Web 2.0.
The irony is that by exactly re-instantiating the good old printing society's principle of producing and dealing with information the computer society could come of age. Reading John Battelle and the literature he is referring to, most importantly perhaps, Jon Kleinberg's paper on Authoritative Sources in a Hyperlinked Environment and Sergey Brin's and Lawrence Page's paper on The Anatomy of a Large-Scale Hypertextual Web Search Engine, both of them dating from 1998, you come to realize that there is no web 2.0 without the preceding construction or reconstruction of authority. Ok, it's a web authority, it's an authority which relies almost (almost!) completely on the self-organising hyperlink structure of the web. Yet, it's an authority nevertheless, which reminds the old battles of the printing society to turn the authority structure of the writing society on its head. The writing society maintained that authority belongs to the sources, holy ones, to be sure. The printing society maintained that authority belongs to the experts, they themselves controlled by further experts, all of them relying on the most recent information, the information best checked according to state of the art methodologies.
And how did search engines take off? They ranked the pages of the web according to links quoting them, to links quoting them coming from pages themselves linked to by others (Kleinberg's authorities), to pages being interlinked as being most often referred to with respect to some issues (Kleinberg's hubs), to anchors describing them, and to the notorious "random surfer" Brin and Page ingeniously introduced to avoid self-circularity.
That's just perfect. Who ever would have started to search the web if only a chaos of results would have been brought to the screen!? So it's the relevance and the reliablity introduced by PageRank and other algorithms which makes us use the search, only thereby producing the clickstreams which Google relies on to make its billions of dollars selling them to advertisers. Of course, the clickstream, once having taken off, does not rely on authority any more but on the surfer's whim. But would that whim have any chance if it could not rely on authority and have its fun circumventing it? There seems to be quite some self-organised authority at the center of the flow architecture of the web, and that's interesting because it resonates with hierarchies constituting some indispensable knots (or nodes) as a gravitational field anchoring the heterarchies (circularities) we are becoming used to live with in the computer age of the society.
Now that the web has had (and passed by) its initial "authority nexus" we seem to be moving rapidly in the direction of being able to make sense of terabytes of heaped information. David Weinberger discusses some aspects of this in "Everything is Miscellaneous" -- that, no longer constrained by the need to categorize and sort physical objects, we are now free to basically throw all information into the heap and pull it out in meaningful clusters as we need. Data structure theories (nodes, hash buckets, etc.) will become increasingly less relevant as the computing resources they were created to conserve rapidly become close enough to infinite that code efficiency considerations no longer matter.
On another note -- is there any work going on to generate some sort of formula that will indicate just exactly how (how much and at what rate) meaning is accreted as a function of the number of links between pieces (bytes, bits, or words) of information?