Community governance

Community governance is a shared responsibility across the academic community. Designated libraries host the CLOCKSS archive node servers. These libraries also are represented on the CLOCKSS Board of Directors, matched by an equal number of publishers. So, there is a 50-50 split between libraries and publishers on the Board, each with one vote, and it has very international representation.2 There is currently one publisher vacancy on the Board.

Decentralized preservation

CLOCKSS looks to academic libraries to provide stewardship and preservation in accordance with the principle of decentralized preservation. In this way, libraries are reinforcing their established social value as memory organizations by being engaged in preservation. They are also ensuring against social and geophysical risks because they are spread around the world in regions that cover Australia, Hong Kong, Japan, Canada, the US, Scotland, Italy and Germany. As well as distributing the risk geographically, the CLOCKSS Archive has actively involved universities from different political and social backgrounds.

Open source technology

The deployment of proven, open source technology by CLOCKSS is based on the use of the LOCKSS software, which has been securely and safely preserving web-published content for well over 15 years. The LOCKSS software continues to evolve to handle web advances to preserve new content types and the LOCKSS technology continues to be adapted for the dark archive functionality used by CLOCKSS. One of the key features is the strictly limited access control, which eliminates any ability of outsiders to access the content on the archive nodes. The existing 12 preservation servers talk to each other, and only to each other, to compare and update any content. Additional functionality has been added to the LOCKSS software to enable a copy to be extracted at the time content is triggered. In the broadest sense, the CLOCKSS Archive is a Private LOCKSS Network (PLN), and there are several other PLNs preserving different types of content.

Commitment to open access

The CLOCKSS Archive is committed to open access. While we are a dark archive that permits no access at the time of the trigger event due to the non-availability of archived content, the CLOCKSS Archive Board will intervene with a vote to trigger the content, which may be set in motion by a request from the community. With the Board’s approval, the content is made available OA to those who want to host the content under a Creative Commons license. There are two main hosts: Stanford University and the University of Edinburgh, though anyone may host the triggered content.

The CLOCKSS community

The CLOCKSS Archive‘s designated community can be identified in three parts. The first includes the scholars, students and readers of electronic academic content (the end users). The second is made up of the libraries that purchase and manage this content on behalf of those end users. And finally, of course, there are the publishers of the content.

CLOCKSS provides indirect, independent services to the community served by its Archive. Essentially, this is content insurance for the libraries. They know if content is in our system and they are subscribing to it, when the content is no longer available for those reasons previously noted, CLOCKSS will be able to provide it. Libraries have peace of mind about the content from participating publishers, because publishers have provided the content for preservation to the CLOCKSS Archive.

In February 2011, the author was hired as the first full-time Executive Director. By 2014, the CLOCKSS Archive had nearly 200 participating publishers in 29 countries, including Egypt, Greece and Romania. As of today (the beginning of 2015), it has 722 supporting libraries in 42 countries, and the CLOCKSS network is expanding. In January 2015, Brazil was in the process of completing its application process to become the 13th node, and the Board looks forward to welcoming Brazil to its network. As the CLOCKSS Board has authorized 15 nodes in total, there are opportunities for growth into other regions such as Africa, other parts of Asia and Europe.

CLOCKSS was audited by the Center for Research Libraries (CRL) in the US using the Trusted Repository Audit Checklist (TRAC). The audit was performed from September 2013 through June 2014 and included an on-site visit and audit of the system. The CLOCKSS Archive was certified as a ‘Trusted Digital Repository’, scoring 13 out of a possible 15, tied on performance with one other archive. It scored a perfect five in the Technology and Security category – the only archive to score perfectly in that category. This certification is a very important accomplishment for the Archive and involved a strong commitment of resources from a small team to prepare all the materials for CRL to review.

Database content preservation

Academic content in databases is a growing area, some of which is not peer reviewed. Among the various items that are being collected in databases are data sets, figures, single entries and ongoing revisable entries. Databases can be broken down into three categories. The first is a closed database, where all the individual records are already contained in the database table. This type is relatively straightforward to preserve because the content is fixed and does not change over time. The second type of database is open, where entries can be made continuously through time. One type of open database is the appended-style database where every new record is added to the end of the database. The data content can be separated into time intervals, and preserved in distinct blocks of data based on a date interval, for example. The other type of open database is the continually updated database. This particular type of database is nearly impossible to preserve and one can only gather a snapshot at a given point in time. To the best of this author’s knowledge, there is currently no known way to capture a continuously updated database.

A further challenge with databases is their status among academic libraries regarding their worthiness to be preserved. Large data sets and data files also present a size problem, not only in collecting the data, but in distributing the data. At the moment, there is a working group on the Board of the CLOCKSS Archive developing new guidelines for databases. When CLOCKSS began, its focus was on journals and then expanded to e-books. The challenge of databases is one not only for CLOCKSS, but for every preservation organization.

Changing standards and formats

Another challenge to digital preservation is the problem of ever-changing internet standards and formats. New programming languages like Ajax and HTML5 have to be addressed as content moves in that direction. Each of those formatting tools allows for much more dynamic interactive content. Once again, the challenges are in the changing nature of the content. In many cases, the actual content of articles and chapters has not really changed, but the issues are with screen interactions introduced to provide additional context or direction. These tools just allow you to access supplemental or related materials. The LOCKSS team at Stanford University, which supports the LOCKSS software, is working under a grant from The Andrew W Mellon Foundation to develop adaptive tools for the LOCKSS harvesting process. Since Stanford University and the CLOCKSS Archive are both non-profit in nature, they continue to rely on outside funding grants in different initiatives to help keep the software up to date. The LOCKSS software is an open source software, and there are a number of developers at various PLNs working on additions and changes to the basic software to solve different problems of parsing, organizing and retrieving content. This model of adapting the software has worked for well over 15 years, and we fully expect it to continue with our various partners around the globe.

Collection development policy

One of the organizational challenges for CLOCKSS is to come up with a method to develop a consensus of what content needs preservation. The advent of the OA publisher and the continued growth of commercial publishers pose significant challenges. Commercial content is on an exponential path and has been for more than two decades. From the CLOCKSS point of view, the support we receive from libraries helps us to develop and maintain the archived node network and the managing infrastructure. The fees that we receive from publishers, both annual fees and transaction fees (ingest fees), are applied to the technical services needed to preserve content. We consider the growth of supporting libraries as key to our funding and to the growth of content under preservation. As our Board works on developing updated guidelines for the selection of publishers, we face a balancing act. In most models of services provided to the community, the funding is based on fee tiers, where the largest pay the most and the smallest pay the least. In short, the largest publishers subsidize the work it takes to preserve a small publisher. The work for either size publisher is essentially the same. Although the time spent gathering the content will vary from small to large, the inherent cost is in the initial set-up, and this effort is generally equal.