In recent years, election technology vendors have started to provide more and more services to EMBs around the world. By deploying their solutions in more contexts and elections than a single EMB would usually be responsible for, vendors have developed a great amount of technical, operational and sales experience. EMBs are often confronted with many offers by vendors to evaluate and eventually purchase new types of election technology. The products and services on offer can be of great help in organizing an election, but there are also several challenges related to this increased vendor involvement in elections.
Needs-driven Approach
While vendors have an interest in the smooth conduct of elections they are involved in, their primary objective is naturally to expand their business rather than to improve the electoral process. Vendors will therefore shape their proposals in order to maximize the use of their products rather than look at the electoral outcome of using their technology. The EMB’s role in selecting electoral technology needs to be the opposite: its primary objective must be improving the electoral process, which may or may not be facilitated by the different technological options available.
Therefore an EMB’s starting point for evaluating technology should not be which technology to choose or how to implement it. An EMB should first and foremost define which electoral problem needs to be addressed and which are the best ways to do so.
Procurement
Elections are unique in each country, and elections technology needs to provide unique solutions for the country in question. Electoral procurement is expected to be conducted with a high degree of integrity, transparency and competitiveness. Yet with high-value contracts at stake, lengthy appeals procedures need to be included in the envisaged timeframe. Technology procurement is a complex process, and tends to take longer than initially expected. With a set election date, this can come at the expense of implementing the chosen solution. Therefore, tender exercises need to be initiated well in advance of the election in which the technology will be used.
The procurement process needs to carefully determine the appropriate systems to purchase. While a full specification would ideally be developed without vendor participation, an EMB may not be able to fully specify all needs and be aware of all technical options and possibilities. In such cases, a competitive dialogue can be considered as part of the procurement process: after several suitable vendors have been identified, the EMB can conduct a structured dialogue with all of them to identify feasible alternative solutions for the EMB’s requirements.
Vendor Lock
Where technology is proprietary to a vendor, where data formats are not open or when an EMB relies heavily on a vendor for its electoral operations, it risks being locked into a particular vendor. In other cases, electoral stakeholders may have strong preferences for a well-established and trusted vendor and may not want the EMB to engage any alternatives. Any such tie to one particular vendor should be avoided to make sure the EMB remains in control of the systems it uses and the costs incurred.
Open Source vs. Proprietary Systems
When the transparency of an electoral ICT application is important to stakeholders, they frequently request open source software be used, which has source code that is accessible to all and comes with a software licence that allows free usage and distribution. Access to source code is one requirement for experts to understand exactly how an ICT system works. In addition to the added level of transparency, open source software is also considered to be cheap and secure, and it limits vendor lock. At the same time, vendors tend to build their business models on proprietary software with closed source codes, mainly to protect their intellectual property and sometimes to limit access to security issues to a small internal audience.
Proponents of open source electoral management software argue that elections technology is ‘mission critical’ for democracy and should therefore be completely transparent and owned by the public. Initiatives such as the Open Source Digital Voting Foundation (OSDV) work on the development of fully open source electoral systems. However, open source initiatives for electoral management systems currently find it difficult to gain momentum and are increasingly confronted with a market that is already served by experienced vendors and their proprietary systems.
Vendors are increasingly recognizing the high demand for transparency by many electoral stakeholders. Norway made open source software a requirement for its Internet voting system and the selected vendor’s source code was published online.
While vendors may not always be ready to switch to open source and free use of their software, they are usually willing to disclose their source code for public scrutiny. Such disclosure can either be limited to certain time frames, controlled environments or selected experts, or it can be complete and public. Increased transparency and access to proprietary codes often come with significant additional costs; depending on the application in question, an acceptable trade- off may need to be made.
Commercial Off-the-shelf Systems vs. Customized System Development
A fundamental decision for electoral administrators is whether to obtain ready-made ICT solutions or pursue the development of a custom-built system. A ready-made system can usually only be adapted to a certain extent, and may therefore require changes in the electoral process to match the system. Such changes are in some cases beneficial and reflect best practices for using electoral technology; in other cases they are required for the sole purpose of making the selected technology work in the given context. By contrast, a custom-built system may more closely fit the existing electoral process. However, customized development is a lengthy and expensive process, and poses extra challenges in managing the development process.
