Gothenburg University Library: episodic digital transformation through open source

Koha is an international community and open source library system that was first put into operation in New Zealand in 2000. The Koha library system is distributed as free software under the GNU v3 (LGPL-3.0) licence and in Sweden there are over 70 organisations that are part of the community and use the system. Gothenburg University Library (GUL) joined the Koha community in 2016 and put the system into operation in 2018 after an evaluation between purchasing library systems on the market or using an open source solution. I have interviewed Camilla Gillén, who is library manager for Digital Services at GUL, about the organisation’s motives and decision to use open source code instead of procuring a system solution on the market.

Analysis

GUL is its own area of ​​operation with over 200 employees whose task is to support research and education at the University of Gothenburg. The library system is a business-critical system that handles all physical lending activities such as deposits and loans, renewals, requirements and the acquisition of books and printed journals. The reason for acquiring a new library system was above all that the old system had been in use for a long time and according to procurement rules needed to be exposed to competition. Prior to the acquisition of the new library system, a feasibility study was carried out to evaluate the advantages and disadvantages of different library system solutions. One factor that was considered in the evaluation was that GUL already uses open source for other systems and is used to participating in open source projects. Experience from previous procurements was that the library systems were unfinished and not adapted to Swedish conditions and that the system vendors did not carry out the promised adaptations in time or that the required functionality was never implemented. Gillén also describes that the procured system solution that GUL previously had was difficult to adapt to the organisation’s processes, which meant that the organisation had to adapt to the system instead of the other way around.

The feasibility study identified several advantages of the open source library system compared to procuring from system vendors. The advantages Gillén lists include that the open source solution was the most cost-effective alternative. In addition to the fact that the library system from the system vendors (which was included in the evaluation) was expensive, GUB had to pay for functionalities that were not needed. In addition, there were no possibility to adapt the system to internal work processes, which had generated large costs that did not bring any added value for end users or the organisation. Although cost was an important factor, the ability to adapt and further develop the open source solution to the needs of employees and the business was what decided in its favour. This brought control over the system development and enabled operational and strategic flexibility by developing the system according to the needs of the business.

“Now we have greater control and can quickly re-prioritise functionality we need in the system.”

Before the system was put into operation, GUL helped to incorporate a modern search engine into the international version of Koha that all users of the system got access to. To participate in the development, the international Koha community has documented quality assurance processes that everyone must follow to contribute with new or expand the functionality of the system. Quality assurance processes are part of the self-regulatory mechanisms that, together with source code management systems and discussion forums, form the basis of many open source projects and are important for maintaining routines and building trust between developers. During the implementation period and the first years after the system was put into use, some of the system developers spent a lot of time communicating and participating in the network’s discussions to influence the system’s functionality. According to Gillén, it took time to build trust within the international community before they included functionality and adaptations from the GUL code base.

GUL is today an active member of the Swedish and international Koha community who exchange ideas and collaborate in the development of the open source solution. The Swedish Koha community is not a registered organisation by itself, the management of the resources is instead handled by the member organisations, which take turns leading the cooperation. This is to minimise administrative costs so that all funds can go in full to benefit the community by building skills, giving lectures and courses to spread knowledge about the library system. System development at GUL uses a well-established scrum methodology with sprints of three weeks which enables the users – the library staff, to participate in the development and submit ideas for functionality to Koha which the product owner can then plan into future sprints. GUL’s iterative development process from proposal to implementation has meant that the system is developed seamlessly with the organisational requirements, where ideas can quickly be tested and incorporated. Gillén admits that the adaptation of Koha in some respects has gone too far because they have a well-functioning development process to take advantage of employees’ needs and ideas.

“Now that we have our own developers … and well-functioning processes, we may be making more adjustments than are financially justifiable.”

It is difficult to place a financial value on the opportunity to re-prioritise and develop an open source solution according to the needs of the organisation. An employee survey of the digital work environment at GUL shows that employees gave Koha the highest rating for user-friendliness among purchased and in-house developed systems. On average, an employee at GUL uses between ten to twenty different systems in his work role. According to Gillén, the exchange of knowledge, ideas and solutions has increased significantly between national universities after the change to the open source solution. The community collaborative development has been strengthened over time and much of the exchange of ideas and knowledge takes place via a Slack channel (cloud-based collaboration platform and discussion forum) where members can ask questions and help each other.

Over time, GUL has become a leading participant in the Swedish Koha community and answers questions from other members about the system. In this way, new members receive support and help, which makes the network grow and creates trust between participants. Previously, when GUL procured library systems, it was mainly librarians at various universities who were active and collaborated. Gillén describes that the choice of open source solution has created a collaborative community where resources and knowledge are exchanged across organisational boundaries to jointly develop Koha. If GUB had procured a proprietary system solution from a vendor, this dimension of value creation would not have been possible because they would not have had control over the development and access to the source code. Gillén does not see the fact that GUL’s staff spend time helping others in the network as a cost, but instead as an indirect gain when several people participate in the community with new ideas and perspectives, which makes the Koha system more mature and adaptable.

“We have staff who are very knowledgeable, who are generous with their knowledge and answer questions when other universities have to make local adjustments.”

From the perspective of a public sector, Gillén believes that it is important to share knowledge and resources with others who have the same need for similar system solutions. Thanks to the fact that several of the employees are included in the iterative development process, from users and product owners to system developers, the system can be adapted and designed according to the needs of the organisation. Had the organisation not had development capacity, it would have had to rely on consultants, which in the long run can create problems. As an example, Gillén says that Gothenburg University has centrally developed several of its own system solutions with external consultants, where they later became impossible to further develop and make security updates because consultants disappear or leave the companies they contracted.

“Gothenburg University has centrally developed systems with external consultants. This creates problems for the business to maintain and update because the consultants who participated can no longer be obtained ”.

Open source is not always the best solution because it requires knowledge and capacity to manage and develop the system solution. Therefore, GUL has chosen to procure a specific solution for indexing large amounts of metadata and free text search on content in articles and e-resources. This is because they themselves do not have the financial resources or the skills required to maintain and develop the technology on which the system solution is based. If the business itself does not possess the technical competence needed, procurement, according to Gillén, is the only real alternative. Then the business may instead try to set requirements for the functionality needed with the risk that the business needs to adapt its structures to the standardised system solution that is procured.

Discussion

GUL’s decision to use an open source solution was strictly a commercial decision to enable operational and strategic flexibility by taking control of the development of a business-critical system. Experience from previous procurements that GUL has carried out highlights the problems with dependence on system vendors by having to adapt the business’s processes and structures to a proprietary system solution. A standardised proprietary system solution can lead to unforeseen costs that are not detected in the procurement procedure. These may be limitations in the structure and design of the system solution that are impossible to review because the system is closed. The library systems that GUL previously procured had structures and designs adapted for the American market, which created unnecessary complexity and problems for adapting the system to the needs of Swedish universities. Another example of the introduction of proprietary system solutions that cause large costs due to limitations in system architecture and design is the West Coast region of Sweden (Västra Götaland), procurement and introduction of a new health record system where the system provider Cerner, can not deliver on time. And where the implementation of the system overdrafting the initial budget of three billion (SEK) and the system provider can not comply with Swedish patient data law regulation. This is an experience that GUL also describes regarding the acquisition of a previously proprietary system solution from a vendor that did not deliver the promised adaptations on time or that the demanded functionality never was implemented.

Dependence on system vendors may also entail risks that are beyond the control of the organisation, as the system solution may use software components that are not known at the time of acquisition. Examples of this are Coop’s (Swedish supermarket chain) cash register system that was hijacked by ransomware that the supplier Visma EssCom is developing; which in turn uses third-party components from Kaseya, a software component for remote control and updating of IT systems. Had Coop had the same opportunity to scrutinise the system’s source code as GUL did prior to the introduction of the Koha system, similar security problems would probably have been averted.

Another perspective between proprietary systems and open source solutions is the added value, which in the latter alternative is created through the exchange of ideas and knowledge that is shared through collaborative development in the community. GUL describes a significant increase of knowledge exchange, ideas and solutions between universities after the switch to the Koha system and participation in the open source project. This exchange became possible when GUL themselves took control of the development and became part of the Koha community by eliminating the limiting factor that the system vendor opposes. Collaborative refers from an economic perspective, partly to cooperative – collective ownership and cooperative use of means for production and distribution. And partly to synergistic – cooperation in a creative, innovative and productive approach for a common goal. That is, the collaborative development and investment of resources that take place in the Koha community is to collectively benefit the members and users of the system.

Summary

Freeing oneself from restrictive structures, which often entails closed and proprietary systems, enables better control over data, possibility of iterative development methodology and cultivation of internal system development capacity, all of which are important factors in enabling digitization and digital transformation of businesses. A literature review of the current research highlights several capabilities, structures and digital assets required to enable digitization and digital transformation. One of these factors that is crucial for performing digital transformation is participation and positioning in the digital ecosystem. In the interview with Gillén, this is manifested through participation in the self-regulating international Koha community, where GUL contributes to the international code base. In addition to participation in the digital ecosystem, the organisation needs the ability to collaborate and manage partnerships with a heterogeneous group of actors in the digital ecosystem. This is important to achieve digital transformation as it cultivates the ability to exchange knowledge and ideas across organisational boundaries to jointly ensure adaptability, quality, security and interoperability for system solutions that are collaboratively developed. In the case of GUL, the heterogeneity is manifested by the fact that in Sweden alone there are around 70 members consisting of universities, municipalities and law firms with different needs that must be coordinated in collaboration with other users and developers in the international community.

In the Koha community, GUL supports other members to implement and use the Koha system, which helps to cultivate the capabilities of their own operation by understanding the needs and perspectives of other users for a common system solution. In the digitalisation literature, this is referred to as the ability to build competence, cross-fertilize and empower external developers and partners outside the organisation. Another ability that benefits from the exchange of knowledge and collaborative development across organisational boundaries is the ability to coordinate various development initiatives and the ability to orchestrate several different IT-supported business processes that contribute to creating added value for end users. None of these dimensions of value creation are possible with proprietary software where system vendors have control and ownership of all digital assets and resources.

If the business does not have access to its own system development capacity or is not involved in the open source ecosystem, the business must rely on contractual relations and outsourcing to an external vendor. Few studies have identified successful examples of outsourcing of strategic IT innovation between client and supplier. The few cases of successful outsourcing of strategic IT innovation are based on long-term partnerships that develop into a benevolent relationship as a result of common goals and goodwill between the parties. The problem of achieving strategic IT innovation through contractual relationships is referred to in the research literature as the “innovation outsourcing paradox”1 2 and is a clear indicator to organisations that tries to achieve digital transformation solely through procurement of proprietary system solutions and contractual relationships. The insight into GUL’s use of open source solution is an example of an episodic initiative to perform digitisation and digital transformation by regaining control over the development of a business-critical system solution.

The content is created by Clear Byte and is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.

  1. Oshri, Ilan, Julia Kotlarsky, and Alexandra Gerbasi. ‘Strategic Innovation through Outsourcing: The Role of Relational and Contractual Governance’. Journal of Strategic Information Systems 24, no. 3 (2015): 203–16. https://doi.org/10.1016/j.jsis.2015.08.001.
  2. Aubert, Benoit A., Rajiv Kishore, and Akie Iriyama. ‘Exploring and Managing the “Innovation through Outsourcing” Paradox’. Journal of Strategic Information Systems 24, no. 4 (2015): 255–69. https://doi.org/10.1016/j.jsis.2015.10.003.