Interoperability

167

168

transacting partners decide to adopt a different platform. Hill (1995, p.120) observes that “[…]the more specialized a resource becomes, the lower its value in alternative uses […] The contingent value of a specialized resource exposes its owner to a greater risk of "hold-up" than the owner of a generalized resource”. To address these concerns, the second level of interoperability is discussed. If TradeLens will be made interoperable with other blockchain-based shipping platforms, the risk of forfeiting relationship-specific investments is reduced for potential adopters, which could in turn increase the likelihood of them adopting the platform.

9.1. Interoperability between TradeLens and legacy systems

A crucial step TradeLens took to support integration with participants’ legacy systems, and ensure simpler interoperability within the industry, was to make its Application Programming Interfaces (APIs) freely available.

As noted by respondents from Mærsk and GTD solution, there are two different options for connecting to TradeLens. Ecosystem members can interact with the platform by either linking it to their proprietary IT systems via APIs or through a user interface. Linking the proprietary systems with TradeLens is particularly interesting for big organizations that are processing large amounts of data. As the Head of Strategy and Operations at GTD Solution/TradeLens observed:

“[…] if you think of a shipping line or a terminal, the amount of data that they are using is considerable, and those data need to be in our platform at scale. And that's not going to happen if you have to have 40 people in a room typing data into a system that can then go to TradeLens”.

Organizations may also opt for the option of using a user interface, instead of integrating their existing systems with the platform. This is particularly relevant for smaller firms, who either do not have a core system with which to integrate, or consider such an integration too expensive.

GTD Solution and IBM also offer on-boarding support for companies that decide to join. The On-boarding team provides guidance and assistance with data and process mapping as well as integration testing, to ensure firms’ current operational systems are able to both feed, as well as consume data from TradeLens. Additionally, TradeLens started to collaborate with third-party integration providers, who may also offer on-boarding support to prospective clients. This may be a particularly interesting option for companies who are hesitant to let on-boarding teams from GTD Solution or IBM too close to their proprietary data.

The support offered by GTD Solution and IBM, as well as open APIs reflect the ambition of the two companies to make connecting to TradeLens as easy as possible, and further motivate industry

169

actors to adopt the platform. Some interviewees, however, suggest that integration with legacy systems is not a trivial issue. They argue that it is difficult to assess the maturity of the proprietary IT systems, and that the integration will require organizations to change their internal processes and “tech set-up” in order to accommodate working with the APIs from TradeLens. This can be particularly difficult for companies using a lot of customized software, accumulated over the years. The Digital product manager at Mærsk argued that: “Over time you build a lot of technical depth, and just by introducing new modern APIs, that's never going to solve the legacy problems, or the technical depth that you may have built up over 20 years. In supply chain, a lot of the software is bespoke software. And a lot of it, with large players as well, is something that they've built in-house so we're not talking about standard software”. Some respondents also noted that simply introducing new technology will not resolve persistent data quality issues. As observed by the CTO of Youredi, TradeLens’ integration solutions provider: “When updating an existing integration from old batch-based to a modern API […] many players will give you the impression that using new technology will somehow magically fix problems with the old technology, and this is very often not the case. Implementing a new API might give a company the opportunity to fix issues with old technology, but new technology is no guarantee that issues will be fixed. Fixing data quality errors usually requires improvements to several downstream systems in the process”.

9.2. Interoperability between TradeLens and other blockchain platforms

Interoperability between TradeLens and other blockchain-based shipping platforms is a critical consideration for potential joiners, as it decreases switching costs and the risks related to asset specificity. A report by World Economic Forum (2020) suggests that companies do not want to get locked in to a specific blockchain platform, as this could inhibit their possibilities for external collaboration in the future.

Considering this level of interoperability is important for TradeLens, since respondents seem to be in agreement that more than one blockchain-based platform is likely to exist in the future. The CDIO of MSC for example suggests that, “The world needs more than one platform, that's for sure. There has to be interoperability because you will always have some of the parties using platform A and some of the parties using platform D or E or F, because that's going to develop”.

This sentiment was recently also confirmed by Michael White, CEO of GTD Solution, who acknowledged that TradeLens will need to be interoperable with other platforms, such as GSBN (Johnson, 2020).

170

An important development in terms of interoperability between TradeLens and other shipping platforms was announced in March 2020, when developers from Oracle, IBM and SAP disclosed they had completed cross-network testing, and were able to connect consortia of firms, clustered on different platforms (Allison, 2020). Since TradeLens is run on an IBM blockchain (based on Hyperledger Fabric), and GSBN is run on Oracle blockchain, this could mean that the risk for partners to join either of the platforms will be diminished considerably, as transaction-specific costs and switching costs are significantly reduced.