Dailykos published a useful short essay by a former harvard librarian, reflecting on the Harvard reorg/layoff news.
I see a couple interesting points here.
Harvard has a famously byzantine library system comprising over forty libraries, and administratively divided into two separate library systems (confusingly called the Harvard University Library or HUL, and the Harvard College Library, or HCL) has changed very little in terms of organizational structure since the late 19th century.
Harvard is not alone here. In fact, I’d suggest that the oldest academic libraries, and ironically especially the old ones that really excelled 80+ years ago, are most likely to have completely dysfunctional organizational structures and organizational behavior today.
Libraries today aren’t the same as libraries 80+ years ago, especially with regard to electronic content we purchase, which has different workflows to manage and different economies to purchase; and in terms of metadata maintenance as well, something which the blog author rightly points out libraries realized the benefits of cooperating/coordinating/sharing many years ago — but sharing cards (or data to print cards) through LC is a different beast than than modern metadata control needs.
I also generally agree with the blogger’s conclusion — but with less optimism:
But second, the importance of catalogers, and more broadly speaking, librarians is not necessarily diminishing into nothingness. The environment has changed radically, and there are sure to be plenty of future “massacre-like” events that will painfully remind us of these changes. But librarians do have a future, and I think it may even be a bright one: they just need to accept that it won’t be quite the same as the past.
I fully agree that there is still as much of a need for the tasks librarians have always done as ever — most definitely and even especially including cataloging/metadata control.
However, despite agreeing with that, I am actually not optimistic, like that blogger is. We are running out of time to demonstrate that our profession, community, and industry is capable of meeting the metadata control needs of the 21st century. We are not doing a good job of it. We do not seem to be capable of changing our priorities, expertise, organizational structures, and inter-organizational collaborative infrastructures, to deal with it.
The traditional goals of libraries have traditionally are still useful and needed just as much as ever, but with different ways of accomplishing them. There is still a great need for an organization specializing in information management on behalf of a user community, and without trying to make a profit off that user community. But I am, sadly, no longer particularly optimistic that libraries as they are are actually capable of accomplishing those goals. However, even in the best of cases, trying will result in some painful organization reorgs — nobody likes change. (It’s of course also possible for painful reorgs to end up entirely useless or even counter-productive, or simply admissions of defeat as libraries slowly die).
Hint: If you or your organization thinks if we can just put all our metadata into RDF as quickly as possible and therefore be “doing linked data”, that this is necessary and sufficient to handle modern metadata control needs — you have not only missed the boat, you are on the wrong boat. I have lately been seeing a worrying increase of people suggesting “oh, we just need linked data to solve that problem”, with “linked data” meaning “the data we’ve already got expressed in RDF”, with a worrying ignorance/disregard for what good data actually entails in the 21st century systems environment.
Bibliographic Wilderness 
A small bit of hope: What you say is true, but what that means is that we haven’t shown that LIBRARIES can do these things.
That’s a different thing from saying that LIBRARIANS can do them. I, for one, am intentionally gearing my library-school teaching to help students find information jobs outside libraries. Embedded librarianship in action!
So …. what does good data actually entail in the 21st Century systems environment?
It is interesting that Mr. Dempsey asks the question, ‘what does good data entail in the 21st Century?’ It might be as well to ask the question ‘what does good data and network organization entail in the 21st Century?’ To what extent should libraries outsource their systems to large conglomeration style databases that carry all the worlds metadata vs distributed microgrid like systems (library and otherwise) that are networked to each other and capable of communicating and interacting but also retain complete autonomy in their own local spheres of influence? From an organizational perspective the problem of library systems large and small is the same sustainability issue as relying souly on large hydro dams vs small microgrids to generate energy.
In a sustainable world it should be possible for a home owner to generate enough power to supply their own needs as well as contribute power to sustain the larger systems of a community. In that kind of world an old house can be just as good as a new one if it is renovated to meet the needs of the people who are going to live there. Dysfunction can exist in a large institution such as OCLC just as easily as it can in any smaller library system, such as Harvard, and we should not rely completely on either.
The question may seem like a chicken vs egg argument, in which one wonders where and from whom does change need to come first. In my mind the answer lies with the money, both in the government and the private sector, and the direction that our respective economies and underlying technologies take us. Certainly there are commercial entities that are rushing in to fill a perceived void in information organization and service that libraries are apparently not able to economically meet. At the same time the core services that all libraries provide is something that in my opinion we have yet to totally comprehend let alone appreciate, in the same manner that we are still grasping to understand the changes that social networks are creating in our societies.
Libraries are currently an important part of the economic cycle and as such do continue to contribute in very powerful ways to the economic viability of any given community. Given the right circumstances we should not only be able to continue to co-exist with Google and other search engines and information vendors, but also make important, even vital contributions to these other entities. However if politicians and the general public perceive that their information needs can be adequately addressed without having a library in their respective community, sure some libraries inevitably will disappear. That’s economics.
Social networks are a means for people to reach out to each other to build strong communities. However social networking and intellectual networking are two different things. Libraries are as much an intellectual network as anything else. A positive healthy social network and environment will not sustain itself if the people in it have no means of knowing if the intellectual expertise and knowledge they rely on and is made easily available to them is valid and truthful. We all need to be aware that without good information (and metadata) and the reliable systems of accessing it, there is the real potential for social unrest. In this respect one piece of good data that people should be able to rely on, and Mr. Lorcan should be aware of, is that the librarian who lives in a given community has a vested interest in that community and is the best person to insure the community’s access to useful and valid information. My bet is that the metadata created within individual communities and the systems that promote the autonomy of communities will at the end of the day be the ones most likely to continue to exist.