Dewey death-imal System

What is abundantly clear to me after finishing The Long Tail is that the only way to truly have a long tail is to take your "business" online. Any permutation of a brick and mortar location, be it store or library, is going to severely truncate the tail. You need to have "room" for unlimited inventory and you need to make that inventory accessible, searchable, and rateable in a thousand different ways or you are really just shoe-horning your services into a long-tailed model, not really extending your tail.

Libraries, even taking into consideration cooperatives which serve to extend the tail, are limited not only by a physical brick and mortar building but also by the classification system that defines pretty much everything within its walls. The Dewey decimal system is antithetical to an essential concept of the long tail: giving people mutiple and varied ways to organize and sort through the massive amount of stuff that populates the tail. Surely our library databases circumnavigate some of that, but bottom line is the library system is based around Dewey and as a result even our filters are limited. A lot of work is being done now on different ways of classifying and searching digitized material (this is one that I find interesting: http://www.cs.umass.edu/~mimno/papers/f129-mimno.pdf), and this is where I think libraries should be placing their focus. How can they build digital materials into their collections? Saturating your collection with digital materials will insure that your tail is legitmately, and not artificially, extended. Undoubtedly there are question marks out there as to who will be supplying the content, how much content can reasonably be expected to be digitized, and will people respond to digitized material, but the reality is that preparation made now will not only extend the library's tail but may well determine their future viability.

Philosophically libraries are aligned with the long tail: universal access to almost anything you need. Can libraries truly be living this philosophy without serious exploration of digital collections and innovative ways to search them?