Libraries traditional collected various materials, especially books, journals, printed material of all kinds, and sometimes art pieces or other physical items. As time went on the list expanded to sound recordings and videos, whatever the device (LP records, CDs, DVDs, etc.)
At the same time, libraries embraced cooperation: the "inter-library loan office" (resource sharing) at Sacred Heart University Library is a vital service. "Loan" is now a misnomer: more than half of our resource sharing transactions obtain a copy (digital or photocopy) which is not to be returned --a service, not a loan.
Indeed, with such cooperation, with e-books, with digital media of all kinds and better players (iPads, Kindles, iPods, and all their kin), why bother trying to gather or collect anything? Why not simply depend upon the network?
This is the crux of the pundits' argument that libraries will simply disappear in the brave new information world where everything is available freely and easily. Unfortunately, that world doesn't yet exist, and is unlikely to do so. Information will not be available freely and easily for the simple reason that a great deal of it will continue --at least for the foreseeable future-- to be controlled by for-profit corporations (e.g. Ebsco, Elsevier, ProQuest, and their kin).
Of course, such information can be easily accessed via web browsers --but only with the right authorizations and passwords. The proxy server makes some of this invisible, especially on campus, but those authorizations are still there. The great information data-banks that corporations have built can exchange data but are no more likely to merge than will the great and regional banks suddenly coalesce into one financial institution.Information will remain divided into corporate silos, some openly accessible, some not.
The first thing that the library does is manage access to much of this environment of information. All of this environment? no --think of Google. Even Google, however, has numerous aspects (Google Books, Google Scholar, Google Patents) which are little known to most users. Librarians --libraries are primarily a service, not a thing-- are critical navigators in this environment.
At great length --with all this! why collect anything locally? The answer is: here, now, and in person. When students and faculty are in "inquiry" mode, they want resources: the old amusing saying, "librarians like to search, everyone else likes to find" is true in a way. There is no finding without searching. Finding can mean materials online, and materials at hand. We still live in an environment of mixed media. Some people prefer printed resources for a variety of reasons: layout, familiarity, ease of use, portability. As mentioned before, paper is (all said and done) not a bad storage medium.
So the question: why collect? is answered: to serve students and faculty here, now, and in person. The question can't be "how many print resources do we have?," but "how good are the print resources we have?" Who now would bother to build a library to hold all the books available? Who is really ready --now-- to walk into a library with no printed books at all?
The ease with which one private secondary school in Massachusetts disposed of its print collection is belied by its dependence upon other schools' print collections --resource sharing. They simply outsourced their print needs to other schools --pleasant news to the budget officer, but with the an effect similar to living on take-out food rather than cooking your own in the kitchen. It's not a bad thing to do for one meal or another. But all the time? What happens in the long run? What happens when they lose access to critical intellectual resources because they have become unavailable due to contractual disputes between database providers?