thoughts on e-books and the ILS

I've been quieter about library systems and technology on this blog than on synthesize-specialize-mobilize, but I'm going to return to that realm for this post. I've had the privilege of joining the Orbis Cascade Alliance's Shared Integrated Library System Team this past month. This team is charged with analyzing the market for a shared ILS for the Alliance and investigating options for its procurement. The team read over RFI responses submitted by several vendors the other week giving us a general picture of the maturity of the market.

I'm not going to comment on those RFIs--we've promised to keep them within the Alliance. But I will say that this process and our library's recent exploration of the e book marketplace has got me thinking about where the next generation of library management systems should go.

I've been a champion of the WorldCat-based ILS for a long time now. Putting the world's largest cooperatively created catalog at the center of a library management system is a logical way to leverage network effects. It also could enable libraries to see data about their activities in the aggregate. OCLC clearly has a strategic advantage with WorldCat data as well as their resource sharing network. One of the main drivers of the Alliance effort is greater collaboration on traditional technical services activities like acquisition and cataloging of print materials and the shared WorldCat database might enable some deep collaboration here.

But if we really are moving into a world where e content becomes prevalent, or even dominant, the central nature of WorldCat bibliographic data is up in the air. When you have a scenario where libraries purchase large packages of electronic books, it doesn't necessarily make sense to do the cataloging for these in the distributed sort of way that WorldCat is built around where libraries do original cataloging as they get items in hand. It makes more sense to perform that activity at a higher level on the chain: the publisher, the book vendor, a national library. Likewise, if we load thousands of records for e books into our discovery system for demand driven purchasing, the records need to be created ahead of time in the absence of physical items.

Furthermore, the expectation for full text searching of books is growing, something the WorldCat database can't help with at this point. Fundamentally, WorldCat is based on records whose data was intended for cards in a print catalog. In a search engine world, an ILS built around managing a catalog, even a web-scale next generation one, may not make as much sense as it used to.

The ability to easily assimilate and manage large pools of vendor supplied data and metadata for licensed and purchased e content may be the most essential aspect of the next generation integrated library system. Serials Solutions seems to shine in this area historically with its e journal products and more currently with Summon while I've been less impressed with OCLC's record here.

It also seems logical to include digital collections management as a next generation activity of the ILS, though its also possible that the tools for this activity (DSpace, ContentDM, Omeka, Shared Shelf, etc.) are evolving too independently from the ILS market to try and bundle them in.

On another note, our library looked at purchasing some e book packages recently. Before looking closely, I had the impression that it was possible to license large aggregations of high quality e books. But what we're finding is that many e books are only available for sale, and at very little discount from the print even if purchased in aggregate packages. This situation really makes the demand driven acquisitions model seem attractive, though I'm leery of buying e books on one platform should that platform prove to be funky or worse, extinct, in the future.

Ebrary's 'Academic Complete' package seems to be a good deal. Offered for a modest annual fee, it provides access to some 50K e books. I get the impression its sort of like the NetFlix Instant library: it offers a wide array of quality content, but is by no means comprehensive and many of publishers' newer, popular titles are not included.

libraries, sciences, and spiders

At the NITLE Summit session on library collections earlier this month, I advanced the idea that on liberal arts campuses, libraries might reorient themselves to provide more support for the humanities and social sciences while consciously putting less resources into support for the sciences. Ithaka studies have shown that though scientists value the online resources that libraries provide, they are less attached to the library as place and as an intermediary in the research process. Perhaps we should concede the scientists aren't as interested in the library, and concentrate on our role as a laboratory for the humanities and social sciences.

This idea was met with some consternation by many of the library directors in the room that have productive, valuable programs in place with the sciences on their campuses.

Well, I'm happy to report on an example from our campus that perhaps somewhat counters the notion that we should scale down our connection to the sciences. Last year, as we explored ways that our digital initiatives program could grow and support more academic areas, we talked to a few scientists on campus. We were suspecting that they might need help with data management. In a couple of the conversations, however, we heard about the need to educate broader audiences as an aspect of NSF grant projects. There was interest in the library's ability to get some of their research data out on the web and make it accessible in new ways.

Our first foray into this type of project involved Biologist Greta Binford, a scholar famous for her work with venomous spiders. This fall, in her introductory "Perspectives in Biology" course, Greta had her students collect, identify, and photograph spiders in the immediate vicinity of our campus. Our digital initiatives crew, chiefly Jeremy McWilliams, worked with the students to upload images of spiders, descriptive metadata, and genetic sequences to Flickr. Jeremy and Anneliese Dehner then did some web programming and design magic to pull together the work of the students into a website, The Spiders of Lewis & Clark. It even pulls in data and images from the Encylclopedia of Life API.

This project is a great example of a digital project that really involves students in its creation and pulls in data from the participatory web. It was gratifying to see the Encyclopedia of Life list us on their homepage after we contacted them about our use of their data. I think its clearly a first for us to make the homepage of one of the reference sources that we list in our research guides. And its particularly fitting because Encyclopedia of Life is a modern reference source built in an open, participatory, decentralized way from numerous information sources.

I'll bring this back to my previous post, and cite this as an example of students getting experience doing work, specifically scientific inquiry, that prepares them to work at the edge of knowledge creation. In this case the library brings the expertise required to organize and present this knowledge. In a sense the project goes beyond referring to information sources to create an information source of its own.

pulling the library out to the edge

One of the concepts in the Power of Pull that intrigues me is this notion of the core vs. the edge. The core has the resources and the authority; the edge is weaker but more innovative and experimental. We want our students to be able to work at the edge when they graduate.

Academic libraries by there very nature might be viewed as organizations that fit into the core of their institutions and for that matter, the core the broader higher education system. Much of what we do is organize, preserve and provide access to authoritative, validated information. Our information literacy initiatives often focus on teaching students to find and identify "appropriate," vetted sources of information, that is information backed by established authorities.

Could it be that this notion of information literacy is misguided or perhaps just half baked? That its basic premise is based on trying to teach students the old, "correct" way of finding information, when in fact this will not serve them well in an age of constant change. It's not that students don't need to know how to find reliable information from within stocks of knowledge. But knowing how to do this is only going to get them so far. If they are going to be at the edge of innovation in a particular field, they are going to need to know how to tap into knowledge flows.

How do we do this? How do we develop an information environment that challenges students for a world of constant change? We need to present them with problems that they can only solve by tapping into tacit knowledge of experts and engaging in current knowledge flows.

Think of a project where a student is tasked with producing original historical research from archival materials. She would likely enter an archive with the guidance of an archivist and perhaps work alongside other students and researchers, picking up tacit knowledge about historical research (a field that is in flux) the way. To make sense of what she finds, she might very well need to connect with other students/scholars that also are knowledgeable about the topic and this connection could be evolving and dynamic with multiple feedback loops.

Or think of a project where a class is charged with developing a digital archive of something. I'll use grafitti art in NYC to take an example from one of our recent digital initiatives. Given the ephemeral nature of this art form, students would need to tap into knowledge flows about where it is happening through both personal and online networks. As they capture images of the art, they might need those networks to understand how to classify the art and capture its context.

We need to be giving students the opportunity to make new discoveries and create new knowledge. We can do this by providing raw materials, expertise, virtual and physical creation spaces, and personal networks.

liberal arts futures

This IHE article captured some of the mood at the NITLE Summit in Virginia the week before last, which I attended. I find often at technology conferences there is a sense of religious reformation in the air and it this vibe was definitely apparent at this one. As is often the case there was a sentiment that educational and financial models at liberal arts colleges must change drastically.

The mood on my campus heading into the end of the semester, is vastly different. This past week, I attended a teacher-of-the year ceremony, in which students gave their own testaments about their favorite teachers. These accounts were inspiring, most commonly emphasized how effectively faculty convey their passions to students, and got me thinking that our model is certainly not entirely broken. Following on that came announcements of a bevy of Fulbright awards to students as well as a ceremony honoring undergraduate researchers.

Many of the remarkable things I hear about on campus involve student research, whether it be in the science laboratory or in the archives. And I think there is a connection here with the reform zeal of the NITLE Summit. John Seely Brown's talk emphasized the connection between thinking and doing and the importance of transferring tacit knowledge. Student research experience that give students the chance to tinker and put themselves at the cutting edge of knowledge gets at this mode of learning.

But is experiential learning of the kind advocated by JSB really all that new? I guess not. I think his point is that this kind of learning becomes more important as we navigate an rapidly evolving socioeconomic system with almost unlimited access to explicit knowledge. I have trouble buying the idea that change is as rapid as he says it is in all sectors of society, but I think he is on to something.

We were also interviewing candidates for our dean of the college job this week. One of the more interesting conversations that came up at one of the faculty candidate talks was around academic rigor. Many faculty believe that there is a deficiency here, that we need to develop a culture that asks more of our students academically. I think there is a tie-in here with the liberal arts value proposition: our expectations should be very high in selective liberal arts colleges and this can be something that sets us apart from other players on the education landscape.

checking in from CLIR/NITLE Summit

I flew to Washington DC yesterday to attend the CLIR Symposium and NITLE Summit. I'm looking forward especially to hearing the keynote by John Seely Brown tonight. I am participating in the session on Library Collections: Strategies for Access and Preservation. Ahead of that discussion, I'll put down one idea that I have about a "liberal arts college digital library."

The Orbis Cascade Alliance is working on a demand-driven shared e book collection. Lewis & Clark is an enthusiastic participant in this endeavor as are our partner Alliance Libraries. I am concerned, however, that this endeavor might run up against some limitations because of the heterogeneity of Alliance Libraries, both in terms of size and the academic programs they support.

I'm wondering if there might be bigger opportunities for collaborating on purchased/licensed digital resources if a large group of libraries with similar budgets and academic programs pooled resources together. Say 100 liberal arts colleges came together and built a "liberal arts college digital library" stocked with e journals, research databases, electronic books and a top-notch shared web presence. The goal would be to cover say 70%+ of the average library's information needs with this shared digital library, which would be priced at a certain cost per FTE. The remaining 30% would be reserved for print books and journals, electronic resources not included in the deal, special collections, etc. There would be savings because of collective bargaining power but also through lots of staff time saved because selection, evaluation, and cataloging of resources and perhaps even the web portal would be done centrally.

A few potential gotchas: vendors might not give dramatic enough price breaks to make the electronic resources included in this package greater than what the average library could get for their money with current ala-carte consortial pricing; it might be hard to get libraries to pull out of state or regional consortial deals already in place; choosing a one-size-fits-all set of resources might be hard given that there are real differences in academic programs; how would previously purchased e resources fit in?; also, if we did use a shared web portal, integrating local-only resources could be a challenge. Some kind of study would need to be done to figure out how big these obstacles really are.

There is a rightful wariness on the part of libraries towards "the big deal," a sense that each library should be free to buy only what it needs. This would be the mother of all "big deals," but the package of resources selected would be done presumably by a a nonprofit collective with input from member libraries, not a vendor (though one could imagine a vendor inserting itself into this position at some point).

In a networked environment, why should 100s of libraries that basically serve a similar set of users all be devoting staff time to selecting, evaluating, purchasing, and organizing a similar set of resources? In a sense, I think the Tri-Colleges already do something similar to this proposal, but only on the scale of three institutions.