˲ Commentaries. These generally
shorter documents discussed specific
facets of blockchain technology in greater depth than seen in other documents.
Analysis. Four members of our
group participated in the analysis of
collected documents. We continued
gathering and reviewing documents
until each felt that the last three to five
documents read revealed no new information; this is a commonly accepted stopping criterion in grounded theory that ensures all core (not one-off)
ideas have been identified. A technical
companion to this article contains the
complete mythological details: the
type of coding used at each stage and
Results. The analysis revealed a set
of 75 interconnected concepts that
define blockchain technology. These
concepts are grouped into five broad
˲ Technical properties—the components that make up blockchain technology. Examples include decentralized governance, a consensus protocol,
and an append-only transaction ledger.
˲ Capabilities—the high-level features provided by the technical properties. Examples include automatic
executions of code (such as, smart contracts), internal auditability, and access control.
˲ Technical primitives—the building
blocks used to construct the technical
properties and capabilities of blockchain technology. Examples include
timestamps, hash chains, and peer-to-peer communication.
˲ Use cases—classes of systems
that the literature identified as applications of blockchain technology.
Examples include cryptocurrencies,
supply-chain mana gement, and identity management.
˲ Normative properties—
representative of what people hope to achieve
using blockchain technology. Importantly, these properties are not
provided by the use of blockchain
technology, as the technical properties and capabilities are. In general,
normative properties relate strongly
to the hype surrounding blockchain
technology. Examples include public
participation, trustlessness, and censorship resistance.
While the concepts defining block-
chain technology are divided into these
stakeholders and their competing in-
terests, and the regulatory landscape.
While there is valuable information
to be learned from industry, analyzing
these sources also brings challenges,
including imprecise terminology and
errors in knowledge; inclusion of hype;
and researcher bias.
The well-established research
method known as grounded theory3, 15
was used to rigorously analyze the data
in a way that directly addresses each
of these three limitations. Grounded
theory helps researchers identify high-level themes and processes within
qualitative data sources generated by
humans and filled with imprecise terminology and descriptions. Additionally, grounded theory limits the impact
of researcher bias, ensuring the themes
and processes are derived from the data
and not from the researchers’ preconceived notions of what the data says.
Materials. The following methods
were used to gather materials:
˲Following RSS feeds that track
news and publications related to blockchain technology.
˲ Downloading materials published
by blockchain consortia (for example,
Hyperledger, the Decentralized Identity Foundation).
˲Reviewing documents from major accounting firms, banks, and tech
˲ Browsing news articles and blog
posts related to blockchain technology.
˲Reviewing submissions to the
ONC (Office of the National Coordinator of Health Information Technology) for the Blockchain in Health Care
In reviewing these materials, we
also followed references and included
those documents if relevant. In total,
132 documents were collected and
split into three categories:
˲ High-level overviews. Often prepared by investment firms, these overviews of blockchain technology provided an enumeration of efforts at using
blockchain technology in practice.
˲ System designs. These papers proposed ways blockchain technology could
be used in a specific system (or, less frequently, reported on a pilot study).
Figure 1. Technical properties for blockchain technology.
Shared Governance and Operation
Resilience to Data Loss