It is not enough simply to make resources available on the network; more active promotion is required if they are to be discovered.
If I want to know if a particular book exists I may look in Google Book Search or in Amazon, or in a social reading site, in a library aggregation like Worldcat, and so on. My options have multiplied and the breadth of interest of the local gateway [i.e., library] is diminished: it provides access only to a part of what I am potentially interested in.
... the institution is also a producer of a range of information resources
His most recent post is an update on the above (and of course I'm skipping over updates in-between). I think his approach is so important, namely because he's bringing a scholarly, methodical viewpoint to assessing the state of libraries today. There's a statement of the current climate, the problems, and possible solutions. For instance, in speaking about the "decentered library":
Libraries don't have a holistic view of traffic against their entire network presence. The difficulty of compiling statistics across services is well known. Libraries are working across multiple environments and systems, with intermittent availability of good data about usage, no consistent approach across systems, and usually no aggregate view. This means that the library's knowledge of the use of its own services, and of the benefits of particular approaches, is limited in important ways. At the same time, while there is awareness of the benefits of better data, a data-driven approach to engagement, resource allocation, or service development is not yet prevalent.
That last sentence might qualify as understatement of the year (make that the decade) for libraries. While the difficulties of gathering such data are apparent, the project shouldn't be impossible. This is just one of the many areas I wish the ALA were allocating resources toward, instead of pursuing their "Yay, libraries!" campaign. There is an urgent need for analysis of library search and usage data, and if the Pew Research Center can tell us about library usage (albeit in a very different way than that mentioned above), certainly someone in the library community could find the means to gather the data we desperately need.