One of the people on campus I talk to fairly regularly works at Greenback's main library. I'm not sure whether there's a specific reason for it or not, but the library staff seems to be more enthusiastic about sustainability than are the employees of just about any other department.
As a result, we've looked into implementing the same sorts of building improvements as I'd like to do to just about any building on campus: resetting heating and cooling points, replacing inefficient lighting, making the HVAC system more efficient, evaluating the building envelope and resolving any problem areas, modeling building occupancy, improving recycling facilities, etc.
Of course, because it's a library, we don't have as much flexibility as we might in a different sort of building. Temperature and humidity need to stay within fairly narrow ranges (narrower than would apply in an office building, for instance). Lighting levels in the stacks need to be bright enough to allow the book spines to be read, without being so bright that those same spines will get faded out. That sort of thing.
But one topic that came up in our conversations was both obvious to everyone at the table and apparently uncomfortable to those with "librarian" in their titles -- most of the space in the library building gets far less frequent use than it did a couple of decades back. Library facilities utilization has grown along with Greenback, but library building/space utilization has actually dropped off, at least for most of that space.
Maybe 75% (I'm guessing) of the usable square footage of our main library building is used for storing books. Open stacks, where students, faculty, staff, even the general public can peruse shelves from floor level up to almost ceiling height. Shelves not 100% crammed full of books, but probably 97-98% full so that if they got any fuller it would be hard to slide books out and back in again. As full as is practical, and maybe a little fuller than that.
But times have changed, and information technology has changed, and information retrieval patterns have changed, and so library utilization patterns have changed. I sometimes go into the stacks armed with the catalogue reference of a single book and end up spending 15 minutes or more just browsing the volumes stored adjacent to the one I knew I wanted. I occasionally (very occasionally) see some other person seemingly engaged in a similar effort. I've begun to notice that the other person is usually over 40 (by appearances). Somehow, I don't think stack-browsing is a skill they teach in elementary school any more.
And I can certainly understand if that's the case. Research for the purpose of coursework is far easier than it used to be, and (at least if volume is any indication) far more productive. Hours spent poring over volumes of the Reader's Guide to Periodical Literature have been replaced by split-second searches of library databases or Google Scholar or amazon.com. Wikipedia (sorry, profs) provides a starting point for just about any literature search, even if it's just used to locate vocabulary useful for constructing a Google search string. Full-text PDFs have virtually eliminated copying of old-fashioned print journals. Bound books are sometimes useful, but they're so 1960's. (Actually, in Greenback's library, a lot of them really are from the 1960's. Or earlier.)
My one-on-one conversations with folks from the library are more open, and so more productive, than anything said in an open meeting. Most of them, while honoring their deepset bibiophile roots, will admit that information technology has changed, and that shelves full of books are no longer an efficient (or for many students, an effective) means of information delivery. If a discussion in Library Journal a year back is any indication, this sort of subject gets repeated and in-depth attention in professional circles.
But the question that intrigues me actually goes at least one step beyond "what is the best sort of library for today's university?" My sense is that while the static information (library) side of the academic equation is beginning to get engaged with 21st-century technologies and their implications, the dynamic information (instruction and/or learning) side is still largely in denial (at least at Greenback). Lectures with Powerpoint slides are better than ones without, sure. But they're still lectures. And the Powerpoints I've seen have few embedded links. And a good proportion of those links are a year or more old and so probably obsolete if not technically broken.
My take, as a consumer of higher education as well as a provider and a facilitator (that's what you call someone who provides facilities, right?) is that pedagogical technology at large and mid-sized universities needs to change and that when it does, whatever it changes to will have a profound impact on future university library utilization demands.
If that's true, I want Greenback to make a limited investment in making our existing library infrastructure as efficient (economically, ecologically or however measured) as practical. Evolutionary improvements. Marginal improvements.
Let's not invest all our assets in solving today's problem if it's about to get overtaken by tomorrow's. Instead, let's perform a thought experiment -- based on your concept of the pedagogy of 2019, what will students and faculty demand of libraries ten years hence? And (specific information technologies aside, since no one knows what they will be), how should libraries organize to meet those needs?