The procurement of public sector information systems ought to be based on the principle of how to provide the best possible service for the users in a cost-effective way. Furthermore, the public sector is bound by the procurement law in their procurement process – the procurement needs to be open, all vendors need to be treated equally and no discrimination should take place. The procurement law of Finland is based on the EU directives and aiming at enforcing a effective utilization of public money.

The information systems procured by various public agencies range from simple systems for common tasks to enormously complex unique systems for some very specific and very unique tasks. The different information systems can be classified into four broad categories: non-unique simple systems, unique simple systems, non-unique complex systems and unique complex systems. In addition, the information systems differ also in another dimension, related to the need for evolution during their life-cycle. Lehman has proposed a qualification of systems in his famous paper from

1980 – an S-program is written according to an exact specification, a P-program is written to implement certain procedures that completely determine what the program can do, whereas an Eprogram is written to perform some real-world activity and how it should behave is strongly linked to the environment in which it runs. In this three-dimensional categorization, the dimensions are thus complexity, uniqueness and evolvability. The specification and procurement process of any information system should address the specific characteristics of a system of a certain category, strongly addressing also the effect of the varying need for evolution.

Intuitively, for a simple systems, especially for the non-unique ones of S-program type , one should aim to find a commercial product to fulfil the need. At the other end, for a unique complex system of the E-program type, one should definitely approach to specification and design in an iterative and incremental way. For other categories, the determination of the specification and procurement process is more difficult and there is a high risk of either over-specifying a system and making it more complex than actually needed, or, on the other hand, simplifying a system to conform to the features of a product while actually the complexity and uniqueness of the system would require more thorough and adaptive approach.

Based on my experiences on and background with public procurement of public sector information systems of varying size and complexity, my understanding is that when an agency is preparing for a tendering process, the type of the system to be procured is not taken into account in an appropriate manner. System specifications are simplified or made unnecessary complex due to the lack of understanding the actual needs to be fulfilled and the level of evolvability required.

By this talk, I’d like to encourage the academia to look into this issue and through research provide the public sector better understanding of the status quo and concrete ideas for improvement.