Joshua Kim started the year off with a challenge to examine our assumptions . Here are some of mine about the purpose and nature of academic libraries, in no particular order.
Learning how (and why) to explore ideas independently is an important goal of higher education, and libraries enable and inspire that kind of learning. A corollary to that assumption is that when we explore ideas, we are willing to have the information we encounter change our minds. An enigma wrapped inside this corollary to an assumption is the notion that people want to have their minds changed. Wayne Bivens-Tatum questions that notion usefully, suggesting information literacy is an "unnatural" activity. Yet, as Dana Longley points out in a comment, this suggests education itself is an unnatural act. If it is, I'm all for it. I'm pretty sure Wayne is, too.
Sharing knowledge advances knowledge. Libraries enable sharing and should help build and fund infrastructures for sharing while resisting every effort to disable it, whether those efforts are encoded in copyright law, business models, or academic cultures. This can be tricky. In the effort to enable maximum sharing for our local communities we willingly disable it for those outside the community. We should, at the very least, insist on the right to share works among libraries as we have always done. When we traded ownership for greater and more convenient access, we inadvertently supported withholding knowledge in order to fund the current system, one that institutionalizes inequity. It also includes enormous inefficiencies such as the generation of large profit margins for a handful of large corporations who now own the legal rights to the contemporary record of an enormous proportion of academic knowledge. In most cases, these rights are not exclusive, but academics have so far shown little willingness to add to their busy agendas the task of uploading their publications in draft form to the web so that they can be freely read. It's extra work and the rewards are not obvious. Libraries need to take more responsibility for fixing this problem which is, in large part, our doing because of the trade-offs we made.
Libraries are not, or at least should not be, engines of productivity. If anything, they should slow people down and seduce them with the unexpected, the irrelevant, the odd and the unexplainable. Productivity is a destructive way to justify the individual's value in a system that is naturally communal, not an individualistic or entrepreneurial zero-sum game to be won by the most industrious (with the slightly-less-industrious thrown under the bus). Taking the measure of academics by counting publications is harmful. I don't know when the word "productivity" became so entangled with the work academics do to advance knowledge, but that was a bad day. Productivity is the enemy of open-mindedness and reflection. We should run it out of town on a rail.
The order libraries create must invite disorder. This is something that is particularly important when it comes to helping students learn how to use libraries. Our systems, which were made that way, are broken by definition. (In case this sounds convoluted, I'm riffing off "the system isn't broken, it was made that way.") If we truly thought knowledge could be nailed down in a system, there would be little use for libraries. Students could simply be taught the right answers and that would be that. As I've said elsewhere, we put books on a shelf together only so they can have a good brawl. Or orderliness conceals anarchist tendencies. Likewise, we need to recognize that the order we create can be oppressive. We need to ensure that the library is open to all, that we recognize publicly that what rules we have are flawed and negotiable. We operate within cultural, social, and economic systems that are unjust, so we need to be alert to ways in which libraries may appear (and may even be) exclusive and inhospitable - so that we can change. (Nina de Jesus has some interesting things  to say about this.)
Libraries are for the common good. Period. This is what sets us apart from other popular institutions that provide information. When we wish our libraries could be more like Google and Amazon, we are doing it wrong. Google and Amazon have two things they want from you: your money and your life. Granted, Google's multiple search engines and Amazon's ability to connect scholars to a wealth of books quickly are extraordinarily useful, and libraries have learned some helpful lessons about how to make catalogs and indexes better, or at least less awful. But the fact is, we will always be a disadvantage to these behemoths because we do not gather and use personal information to improve the search experience (and make gobs of money). That's because we feel, like Edward Snowden, that privacy matters, that it is necessary for freedom. Luckily, libraries are not really about search. They are not an information shopping platform hobbled by inferior technology. Rather, libraries are common ground where knowledge and culture can be shared and nurtured. Protecting the commons while inviting people to use and contribute to them is what librarians are for.
Or so I assume.