Keywords

1 The Origin of Distributed Ledger Technology

Fueled by the highly volatile development of cryptocurrencies, like bitcoin, the distributed ledger technology (DLT) has gained attention from technology experts across industries. Currently, DLT is following the Gartner [1] innovation hype cycle and is in the “phase of disillusionment”. By following this cycle, it is expected to reach a stable level of maturity in the future [2].

The technology of distributed ledgers and blockchain (BC) is built on the concept of ledgers. These initially come from the accounting sector and are a collection of a group of similar accounts in double-entry bookkeeping. Accounting ledgers summarize financial information as debits and credits and show the current balances of single accounts. Therefore, bookkeeping processes are building the foundation for most of the transactions of current economies [3].

The introduction of the internet and the connection of digital ledger to network systems, like enterprise planning systems, led to another increase in efficiency and enabled automated transactions across companies. Today, nearly every company, but especially the E-Business sector, uses some kind of digital ledgers in their IT landscape. While the value generation happens through cooperation across companies, each company keeps its own record of transaction between partners. This happens in each company in specific central databases. The rise of distributed ledger technologies allows the use of shared digital ledgers for value generation. Through these new technologies, a more efficient use of each companies’ capabilities can be enabled [4]. As like many emerging technologies, there is some confusion what the technology exactly covers and what its boundaries are. Based on the expected importance for economic transaction processes, a clear understanding of what the BC technology and the related DLT entail and how they could be classified is necessary and will support efficient implementation of market use cases. This generally valid clarification has not been settled by researchers, regulators or general accepted experts yet.

The first examples of these technologies were BCs like Bitcoin [4] and further developments enhanced it. As the applicable use cases evolved the technology evolved with it to a point where the earlier definitions do not fit the current state of the technology. For this reason, there are currently varying opinions in the scientific literature what BCs are, and where the borders to DLTs are. This also influences the practical applicability, as many practitioners manly think of cryptocurrencies when speaking about DLT and BC technology.

Therefore, this paper aims to determine a common understanding by deriving characteristics of both terms. By this, the research addresses scientists and practitioners at the intersection of DLT use case applications and IT architects, in addition to that the research delivers a benefit for the whole DLT community. To derive the common understanding, a scientific literature review was conducted, which focused on the used definitions in current scientific publications in the area. These definitions were then split into their single components resulting in a quantitative analysis of the mentioned characteristics. Using these, definitions for both terms were derived. To additionally understand how the terms relate to each other, a market research was conducted to either identify BCs that not fulfill the DLT definition or DLTs that not fulfill the BC definition and therefore to understand if both technologies are on the same level, or one is the umbrella term of the other. Afterwards the importance of DLTs for e-business is explained and an outlook with further research topics will be given.

2 Analysis of the Terms DLT and BC

2.1 Scientific Literature Review for DLT and BC

A systematic literature review was conducted following the vom Brocke et al. [5] approach. After the definition of the review scope and the conceptualization of the topic, the literature search, an analysis and synthesis led to a final research agenda.

The search was limited to results starting from 2008, as this is the year Nakamoto [6] released the bitcoin white paper that started the surge of BC solutions. The search was also limited to peer-reviewed, full-access papers and focused on the search terms “Distributed Ledger”, “DLT types”, “DLT ecosystem”, “distributed ledger types”, “distributed ledger ecosystem” and “DLT” in the databases IEEE xplore, Springer Link, ACM and EBSCO host. For each search term the databases were analyzed, and a set of papers selected. After this procedure, a total of 74 relevant papers remained, which then were examined.

The analysis of the literature is split into two parts, each of which focuses on either DLT or BC. The goal is to create a quantitative analysis for the core characteristics of each term. The expectation behind this quantitative analysis is that core characteristics are mentioned in most of the definitions. On the other hand, characteristics that are only mentioned a few times are not considered to be as important and are not central aspects of the term [7].

2.2 Shared Identity and Characteristics of DLT and BC

The research revealed different views regarding the relationship between DLT and BC. Some authors suggest that the terms are synonyms and represent the same concept [8]. Some view DLT as the umbrella term for concepts like BC and similar technologies [9, 10] and, additionally, there is a third group that views DLTs and BCs as related but independent technologies [11]. To ascertain which view may be considered the most reliable, it is necessary to first determine the characteristics that define DLTs and BCs.

In each of the selected papers, either the term DLT (22 definition attempts), BC (44 definition attempts) or both were defined (8 definition attempts) or shortly described. The number of analyzed papers has been raised until the ranking of optional and core features has stabilized. The said definitions were then extracted and split into their components for each term. These components were then grouped into related features and each of the mentioned characteristics counted. Depending on the number of mentions, the characteristics were split into two groups: core characteristics that must all be fulfilled in order to be classified as a DLT or BC; and optional features that may be fulfilled but are not mandatory. Characteristics mentioned by less than 20% of the sources were categorized as optional [7]. The complete list of mentions for each component can be found in the appendix (Fig. 1).

Fig. 1.
figure 1

Number of mentions for each characteristic of DLT and BC

Full size image

The core characteristics for DLT are described in detail in Table 1 and visualized in Fig. 2.

Table 1. Description of the DLT characteristics
Full size table
Fig. 2.
figure 2

Characteristics of DLT based on the quantitative analysis

Full size image

From these core characteristics, a definition for the term DLT can be created:

“Distributed Ledger Technologies are one type of distributed database shared over a peer-to-peer network, where transaction data is synchronized between nodes of the network and the data is immutably stored and secured through cryptographic techniques. Decisions in the network are managed through consensus algorithms.”

Additional to the characteristics shared between DLTs and BCs, there are some core characteristics of BCs that are only optional features of DLTs (Table 2).

Table 2. Additional BC characteristics
Full size table

Based on these additional characteristics mentioned by the authors, a new definition for the BC term can be created:

“A blockchain is a transaction based, chronologic, immutable and synchronized distributed ledger shared over a peer-to-peer network. In a BC, transactions are stored in interlinked transaction sets, referred to as blocks. They execute and record single transactions using consensus algorithms and bundle them into transaction sets using cryptographic techniques.”

When comparing the core characteristics of both terms it becomes clear that all of the DLT core characteristics are also core characteristics of BCs (cf. Fig. 3).

Fig. 3.
figure 3

Characteristics of DLT based on the quantitative analysis

Full size image

On the other hand, some of the BC core characteristics (chained blocks) are only optional features of DLT. This leads to the conclusion that BCs are a specialized form of DLT. On this basis, the statement can be made that, “Blockchains are a subtype of DLT”. To prove this, there must be DLT types that fulfill the DLT core characteristics but not those of BCs and the opposite situation should not exist.

2.3 DLTs Besides BCs

The previous literature search identified papers that focused on initial BC implementations like Ethereum, on certain DLT aspects (e.g. consensus algorithms, smart contracts) or their applicability to certain use cases, but not on the overall structure of DLT concepts. To gain this information, an online market research for current “DLT” solutions in the BC community is necessary. This approach will use a framework created by Cambridge University [33] during a DLT market research. To obtain data of smaller DLT-framework projects, a search for postings on non-scientific media like blockchain news websites, e.g. coindesk and coinhero, or in topic specific DLT and BC community boards and blogs similar to where the idea for the first BC bitcoin started, was executed. The goal of the market research is to check if there are projects that fulfill the DLT criteria and not the BC. For this reason, only a small sample of different projects were analyzed and clustered into groups. Further research is needed to gain a complete overview of the current DLT market.

Seven projects (IOTA, peaq, Hashgraph, Corda, ThunderChain, Radix DLT, Hyperledger) were selected for further evaluation and clustered into four groups (directed acyclic graphs (DAG), sharded ledgers (SL), hybrid ledgers (HL) and BC). One example of each group can be found in Table 3. These groups represent a snapshot of the current DLT market without any claim to completeness.

Table 3. Analysis of DLTs besides BCs
Full size table

While this is only a small sample of DLTs besides BC, it allows the creation of clusters of the different approaches. The goal of these new approaches is to mitigate the issues that exist with current BC solutions. One is the difficult scalability of classic BCs. Another issue is the limited bandwidth for transactions which is the result of the synchronicity of all nodes. New blocks can only be created in the same speed as a block needs to be propagated through the network. To counter this, directed acyclic graphs (DAGs) use an asynchronous and non-deterministic way of structuring data. Instead of interlocking blocks into a chain, DAGs use a treelike data structure where a transaction can have multiple predecessors. This allows a higher bandwidth and increased scalability. Another cluster can be built for DLTs that do not share all the information with all nodes. These sharded ledgers only share between nodes that have an interest in the information. This is useful in cases where data privacy has to be protected and only certain nodes have the right to see the information. Another approach is the combination of a BC with other technologies. ThunderChain combines a fast protocol with a central service (accelerator) with a decentral BC as a backup. These hybrid solutions use the advantages of BCs and expand them with other technologies for different use cases. The following Fig. 4 shows a categorization of these subtypes and their characteristics.

Fig. 4.
figure 4

Categorization of DLTs

Full size image

The categorization is based on the overview of Hileman and Rauchs [33] and part of the IEEE reference framework for DLT/BC [34]. It is extended by the defined subtypes of DLT as well as examples for each category. While the four DLT subtypes cover a wide range of the existing DLT concepts, they may not cover all of them. Newer concepts will extend the range of DLTs and offer new possibilities for future use cases.

3 Relevance of DLTs for E-Business

Within the line of argumentation in Sect. 2, it was shown that all of the categorized DLT-solutions have an impact on all kinds of networked E-Businesses. The goal of the E-Business economy is the connection of different demanders and suppliers for uninterrupted trade. The E-Business sector has evolved from direct business-to-business interactions to a worldwide-decentralized connected transaction network. DLTs are especially designed to connect different economical units over a network without the need for a trusted third party and, as such, DLTs will influence market environments [4, 20]. So far, market platforms in an e-commerce sense are mostly managed by single platform holders that have full control over the said platform (e.g. Amazon marketplace). In a DLT environment, market rules are determined at the beginning and afterwards can only be changed in a decentralized way through the consent of the majority of market participants. There is no longer a central entity to control the market. The single point of failure can be deleted as DLTs are distributed across a node network. As shown, DLTs can also help to reduce the necessity for intermediaries in general, e.g. as part of business transactions. Therefore, DLT solutions will reduce transaction costs in an E-Business environment [35]. Current information technologies can lower the transaction costs because they ease access to information and past transactions and also lower the barrier of interaction between the companies [36]. DLTs have the ability to further lower these costs by eliminating intermediaries and thereby lowering the cost of manual labor. They also provide the opportunity to model relationships through the use of smart contracts (even though not all DLTs support smart contracts).

4 Conclusion

The paper has shown that a maturing level of development of BCs and DLTs can impact future economies and reduce the inherent transaction costs. Therefore, the importance of a clear understanding of the technology has been stated and this issue has been addressed through a scientific literature review that explored 74 definition attempts.

During this process, key characteristics of the BC technology and the DLT itself were identified and the overarching position of the DLT ascertained. It was indicated that the BC and DLT community has grown into a serious industry, which further develops specialized solutions and tools to improve established business transaction processes and tries to reorganize centralized ledger systems by improving the way transactions are handled and connected ledgers communicate, synchronize and update. As shown, some DLTs that are labeled as BC solutions do not fulfill the identified core characteristics and are therefore not BCs. This issue was resolved in a new hierarchical categorization of the DLT solutions. Further updates of this categorization will be necessary in the future, when standardization boards, e.g. ISO, DIN, ASX, will release their own DLT definition. Additionally, new technological solutions will enhance the spectrum of DLTs. As the categorization (Fig. 3) does not claim to be complete, further research is necessary to obtain a fully comprehensive view.

This future research should update the categorization of Fig. 4 in regular (e.g. annual) research loops and should include a backward comparison. This is necessary if new DLT data models have occurred on the market and therefore could impact the DLT or BC definitions or expand the DLT categorization on a horizontal view. This research expands the volume of underlying information and will therefore raise the quality of the definitions and the categorization even more. It should include additional criteria’s, which address especially business scientists.

Additional research should focus on extending the DLT categorization for practical usage. The extension needs multiple perspectives. First, by going deeper into each category and analyzing their architecture and additional information for each DLT type. Second, by extending the number of existing DLT types with new developed types. The mentioned future research could be handled by additional systematic literature and market analysis and should be proved by qualitative expert interviews.