On May 10, 2010 PTFS (Progressive Technology Federal Systems, Inc.) announced Harley, a set of bug fixes to Koha, the world's first open source integrated library system (ILS). The fixes were posted on www.koha.org, the corporate website developed by LibLime, a long-term commercial supporter of the Koha open source software that PTFS acquired in March 2010, not on www.koha-community.org, the official website of the loose worldwide consortium of Koha developers and users. The latter was established in February of 2010 as a rift developed between LibLime and other developers of Koha. Users are asking, does Harley constitute a closing of the rift?
Harley does not include all of the development work done by PTFS and LibLime in the past year. It is limited to bug fixes and a few enhancements funded by customers of PTFS between October 2009 and April 2010. PTFS is submitting these for inclusion in the forthcoming community or public version Koha 3.1 and the following Koha 3.4.
Most of the development work done by LibLime since October 2009 is distributed only to subscribers to its Enterprise Koha, a software as a service (SaaS) or hosted solution that it differentiates from Koha Community and KohaExpress, the LibLime offerings that run the official Koha Project release (currently 3.02) of the community.
Some of the enhancements to Koha currently available only in LibLime's Enterprise Koha include an integrated Zebra search engine, enhanced acquisitions, enhanced cataloging, support for MARC21 holdings formats, XML support, 13-digit ISBNs, batch loading of patron records, off-line circulation, check-out slips sent to an email address, hourly loans, course reserves, fine thresholds, enhanced SIP2 capabilities, and consortium support. In 2008 and 2009, 21 participating libraries in WALDO (Westchester Academic Library Directors Organization) contributed almost one million dollars to this development effort.
While the LibLime Koha Community offering is available both as a SaaS or hosted service and for local installation, Enterprise Koha is limited to SaaS because the General Public License, which all those participating in Koha development must accept, requires that any enhancements made to open source software must be "distributed" by free download to any interested party.
Specifically the language says: "You must cause any work that you distribute or publish, that in whole or in part contains or is derived from the Program or any part thereof, to be licensed as a whole at no charge to all third parties under the terms of this License." Hosting the software on LibLime's (now PTFS') servers does not constitute "distribution," therefore, there is no restriction on limiting access to the software enhancements.
LibLime apparently took this course of action so that it would have a competitive advantage over other commercial firms supporting an open source ILS. These include ByWater Solutions in the U.S., BibLibre in France, Turo Technology in the U.K., and Catalyst Limited in New Zealand-all actively contributing all of their current development to the Koha community.
PTFS, and LibLime before it, have said that all development will ultimately be available for installation locally when it has been contributed to the Koha community, but no time frame has been given. While Enterprise Koha subscribers benefit from monthly bug and enhancement releases, Harley represents the first unrestricted contribution made by PTFS, and LibLime before it, in the past year. In an area such as library software, which has only limited development resources, separate public and private versions are unfortunate, but understandable as for-profit companies, including those that support open source software, seek to maximize their profits.
Libraries and consortia that select open source do so for a variety of reasons, but a common thread that runs through the scores of interviews the author has undertaken with adopters in the past 3 years is a "belief" that open source gives libraries greater control over development and reduces cost. What has been described as a "fork" in the Koha development process that creates public and private versions can be expected to reduce control and increase costs.
However, all of the foregoing is a distraction. By focusing on the emergence of two versions of Koha, we lose sight of the fact that libraries and consortia should focus on finding the optimum solution for their integrated library system needs. That requires setting forth a detailed set of functional specifications, support requirements, and performance guarantees and remedies in an RFP that is circulated to a number of vendors, both proprietary and open source. Only when there is a detailed comparison of submissions is it possible to determine which solution is best for that library or consortium.
In more than 32 years of evaluating vendor responses to RFPs-some 3,000 responses-the author has found that a choice among vendor submissions requires evaluating on several criteria, including the following:
- Conformity to the functional specifications
- Flexibility of the software
- Suitability of the proposed or recommended hardware
- Conformity to standards
- Vendor viability
- Vendor performance per existing customers
- Initial cost
- Five-year cost
- Performance guarantees and remedies for poor performance
It is rare that any vendor's offering scores well in each of these categories. Every major vendor of ILS software, both proprietary and open source, has been top ranked at some time in the past 3 years. Every major vendor, including proprietary and open source, has been lowest in 5-year cost at some time in the past 3 years. And each has ranked as low as fifth overall.
Choosing an ILS should be an objective process, not a subjective one.