One of the smartest data guys I know, Equinox’s Bob Molyneux, is in New Jersey as I write this talking to the New Jersey VALE group about Evergreen and consortial ILS deployment, a topic that we at Equinox know a bit about — we manage a State-wide consortium in the US and a growing Province-wide system in Canada, and there are several other installations well under way.
Why are large consortia forming around and with Evergreen? To answer that question we must first define what a consortium is. There are different types of consortia, and each consortium of each type has its own subtle variations from the broad definition, but basically they can be broken down into three main groups
- ILL consortia like GIL — These consortia build union listings for broad search. Some ILL consortia will attempt to smooth the way for interlibrary loan by mediating requests through a central system, though require that all participating libraries use the same ILS for seamless ILL and are notoriously difficult to configure and maintain.
- Purchasing consortia like WALDO — These are generally meant to provide leverage when negotiating with vendors. Direct resource aggregation is not their primary concern. As such, these consortia do not directly leverage economies of scale for implementation, and they generally end up being as expensive as taking the ILL consortia route to resource sharing.
- Consolidating consortia like PINES — These consortia work to realize the full benefits of interdependent cooperation through broad resource sharing initiatives, such as a single shared ILS implementation.
Evergreen directly addresses the needs and concerns of Consolidating consortia by providing a way to consolidate hardware, software, support and administrative services, while still allowing local control over policies and procedures. This works, and works well, and is inspiring the creation of other Consolidating consortia throughout the US and internationally.
(As an aside: yes, I know you’ve heard that Evergreen requires adopting uniform policies, but that’s simply not true — it never was and never will be. Every participating library in an Evergreen installation can implement their own local policies, though there are benefits, especially for patron experience and staff training, to adopting similar policies.)
But what do all of these types of consortia have in common, in terms of what they attempt to address?
Markets. It’s all about markets.
Things
First, there is the internal market of resource acquisition within a set of cooperating libraries. Resources are scarce and must be allocated in the most efficient and effective manner possible.
Because ILL consortia do not attempt to consolidate, but rather duplicate ILS installations, this market is not addressed in any direct way.
The issues here are partially addressed in the Purchasing consortium model, though only on the procurement end and not on the resource requirement end.
However, this is one of the core goals of Consolidating consortia, and one of the driving forces behind the initial development of Evergreen. The reduction in direct costs that come with an Evergreen powered consortium is undeniable and amazing. What were once repeated, per-system costs for licensing, hardware and support becomes a single shared investment which can be 90% smaller, a full order of magnitude, than the aggregate for existing individual systems. This economy of scale is unprecedented in library-run software, and presents wholly new opportunities for self-direction and resource distribution.
People
Next, we have the external market for patrons. Libraries, for better or worse, are traditionally silos. In practice, they build local collections, limited by available resources, based upon what they believe will most benefit their patrons. Every community is different, and so every collection is different. Patrons, however, are not concerned about local library politics or the intricacies of collection development. Patrons want to retrieve the items they need, and they want to do so as efficiently as possible. They vote with their feet, and will follow the path of least resistance to the items they desire.
ILL consortia attempt to address this issue by building a new system around and on top of existing standalone ILS installations. These union listings, though, are not authoritative and are generally not updated in real- or even near-real time.
Purchasing consortia do not address the issues of the patron market in any direct way, as their constituent libraries do not (within and for the purposes of the purchasing consortium) aggregate collections.
Consolidating consortia, on the other hand, address the issue by being the authoritative source for information. They can provide a seamless experience to the patron and drive circulation of the “long tail,” helping patrons discover resources they never knew were available. In PINES we have seen a trend whereby patrons from areas served by non-participating libraries will travel out of their way to a PINES library because PINES has a larger, broader collection — nine million items on 1.8 million titles — than any one participating library could hope to amass. Put another way, patrons will seek out the larger, less convenient collection in preference to the smaller local collection, period.
So we have shown that one way to positively position libraries in both of these markets is to consolidate services in a single large, scalable system. To date, Evergreen as implemented at PINES is the largest single-instance public library ILS deployment in production in the United States with 275 outlets in 49 distinct library systems. To the best of our knowledge it is also the largest single-instance public library ILS deployment in the world. Evergreen is also the first, and currently only, ILS built to deal with the complexity of such a consortium, and is proof positive that Consolidating consortia work and that patrons and staff reap the benefits of addressing the issues in both of these markets.
TIMTOWTDI
There Is More Than One Way To Do IT — the Perl motto
Consolidating consortia, even implemented with Evergreen, are not without costs and great effort, however. It takes strong leadership and great political will to build a large Consolidating consortium. It requires buy-in from all parties of interest, and a willingness to work together and compromise. Even with all this, timing and existing contracts can make building such a consortium nigh on impossible. So what’s to be done?
Enter: FulfILLment™
FulfILLment is a new Open Source product being designed and developed by Equinox Software to help address cost and patron issues from the point of view of an ILL consortium by applying the lessons learned building Evergreen for the Consolidating consortium model. At the same time, FulfILLment aims to provide enhanced services to both patrons and staff, lowering the barrier to entry for all users.
FulfILLment leverages the underlying architecture of Evergreen and many of its concepts and algorithms. Some of the greatest strengths of Evergreen — its hold and circulation policy flexibility, its open and extensible OpenSRF architecture, and its easy integration with external services — are at the heart of FulfILLment and can be applied directly to the problem of inter-system ILL.
FulfILLment does this by redefining the problem space. Conventional wisdom says that ILL is conceptually distinct from circulation. The problem of automated ILL has thus far been addressed mainly through the use of union catalogs. This eases the search and discovery portion of ILL, assuming that the union catalog is up to date, but many union-catalog based systems do not address automated mediation well, if at all. In fact, many systems are little more than cross-lending agreements on paper that require full staff mediation. The reasons for this are clear if you consider these points of traditional ILL:
- Little or no a priori knowledge of item or patron status
- Little or no surrounding context for the ILL request
- Either: forced ILS uniformity to provide seamless ILL; or spotty interoperability of NCIP and related protocol implementations, or in the worst case, so-called “paper ILL”
All of which lead to complicated data gathering, retention, aggregation and reporting.
However this conceptual distinction between ILL and circulation is not necessarily correct, given the proper circumstances. Consider these counterpoints from the view point of a circulation system:
- Complete global state knowledge
- Complete context of hold requests
- Implementation based on “best and most appropriate” protocols
FulfILLment brings the benefits of a circulation and hold system based on the core algorithms in Evergreen to the ILL problem space. By encouraging and facilitating the participating institutions to collect and enter all relevant information about ILL policy and system definition, FulfILLment can provide not only truly automated mediation of ILL request (holds) but also full ILL transaction (circulation) management and automated transit management.
If the infrastructure and algorithms of Evergreen are the heart of FulfILLment, the FulfILLment Next Generation Discovery Interface (NGID) is its public face. The FulfILLment NGDI is a hybrid physical/virtual union catalog which automatically loads and deduplicates bibliographic records from all participating institutions for central search. Records can be pulled in using standard protocols such as OAI-PMH, Z39.50 or SRU, or can be automatically pushed into the NGDI from the local systems by whatever means are available.
FulfILLment leverages the hold targeting and capture algorithms from Evergreen — arguably the most advanced in the world in terms of policy and process modeling — and uses them to find the best item to fulfill the request of the patron based on all available information.
FulfILLment’s hands and eyes consist of the Local Automation Integrator, or LAI. This system uses the best and most appropriate protocol for each participating institution’s ILS, be it NCIP, SIP2 or a custom connector, to query each local ILS that advertises items on the requested record in real-time in order to provide status information to the hold processing mechanism. Data concerning the requesting patron is also pulled in real-time, and can be automatically obfuscated or expunged if required as soon as all open ILL requests and transactions for that patron are completed.
Once an ILL request has been accepted by the lending library and the item captured for remote circulation, FulfILLment tracks the complete life-cycle of this transaction. Both ends of the ILL, the lending and borrowing libraries, know the current status of items on loan and can generate reports based on this information using the Evergreen Reporter, one of the most advanced and flexible reporting engines available for an ILS.
The road ahead
This true end-to-end management and automation of ILL, particularly in an ILS-agnostic fashion, is something that has not yet been achieved in the library world. The potential benefit to staff in terms of reduced workload when fulfilling ILL requests might be enough on its own to make FulfILLment worthwhile, and when the cost reduction benefits of Open Source software, the advanced Next Generation Discovery Interface and the capabilities of Evergreen are brought to bear on problem of ILL, the return on investment will be immense.
We at Equinox are truly excited about the possibilities that will open up for building and growing ILL consortia on a scale not yet seen. You can expect to hear more in the near future and we hope you will all join us on this path to FulfILLment.
–miker
