bookstore by Евгений Малолетка (cyberpunk)) on
bookstore by Евгений Малолетка (cyberpunk)) on

Reading Ted Striphas’ blog entry from last February, The Shannon and Weaver Model brings to mind another one of Shannon’s contributions to information theory, and one that I find applies directly to the business of bookselling.

As I have written before, I think discovery is destined to become once again a core service of independent bookstores. The easiest way to explain why is to use Shannon’s concept of signal to noise ratio.

Claude Shannon researched cryptography for the military during the Second World War and then he researched telecommunications for Bell Telephone Labs. One of his research concerns was how to maximise the value of a telephone line for communicating information (improving phone calls). His key insight (illustrated in the formula below) was that an information channels capacity for conveying information depended on two things.
Shannon–Hartley theorem
First was the bandwidth: how big the channel of information was. Metaphorically this can be compared to a highway or a drain pipe. The more lanes on the highway, the more traffic it can carry; the bigger the drain, the more water can get through. Anyone trapped in traffic on a busy highway knows that having an extra lane would make all the difference (and of course some drivers create that extra lane for themselves). I live in North Wales and during the past week we’ve had so much rain that the bandwidth of the drains has proven woefully inadequate.

Lack of bandwidth
Lack of bandwidth (source: Nibiru – Planet X)

The second was the nature of what went through the channel. Shannon divided it neatly into signal and noise. Signal was the message meant to be put across; noise was pretty much everything else.

The formula is technical and many see it as the foundation of the information revolution (the log2 in the formula is where binary maths get involved). It is also a powerful metaphor for a growing problem in the world of publishing. With the rising levels of book production (328,259 new books published in 2010 alone, plus 3,806,360 new POD and reprints). Publishers act as an access channel to books, but the capacity of that channel is growing astronomically, making it increasingly difficult to differentiate the book a reader wants from all the books that are available. If the wanted book can be seen as the signal, it will only be noticed if the noise can be filtered down to a level where the reader can make an informed choice. Signal to noise. The goal then becomes to identify or develop the best filtering systems we can. Software and information engineers are doing their best using rationalized algorithmic techniques; the media (book clubs, book reviewers, film makers, publisher marketing officers) help some more. Neither can (yet) provide the trust that comes from social connections and community combined with knowledge and experience of a subject, the books, and the reader. For that, I would have to look at the independent bookseller. – particularly the bookseller dealing in second-hand books.

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.