Applying

blockchain
technology
in global data
infrastructure
ODI-TR-2016-001

Open Data Institute

1 | Open Data Institute 2016 | Applying blockchain technology in global data infrastructure

Table of contents
Executive summary

Introduction 3
The foundations of blockchain technology

The technology landscape

Scaling and linking blockchains

13

Privacy and security

16

Practical experimentation

20

Conclusions 22
About the ODI

23

Bibliography 24

Authors: James Smith, Jeni Tennison,

Peter Wells, Jamie Fawcett, Stuart Harrison
Editor: Anna Scott
Design: Christie Brewster

Executive summary
Blockchains, or distributed ledgers, are part of a new area of technology that is generating a
lot of interest. They have the potential to become an important component of our global data
infrastructure.
In this report, we present an overview of Blockchain technology and issues that come with it,
for a non-technical audience who seek to understand the potential of distributed ledgers and
blockchains in a commercial or policy context. We focus on non-financial use cases, both to
avoid duplication of other work and to explore the wider impact of the technology.
Some of the key areas covered in this report are:
the basics of blockchain and distributed ledger technology
the landscape of use cases and applications
how blockchains will need to link into our global data infrastructure
how blockchains will need to grow and scale over time
the potential privacy implications for personal data in distributed ledgers, and
the risk of adding personal information to blockchain systems without careful
design
a practical exploration of building a blockchain system for data storage
We conclude that distributed ledgers are a potentially important area of technology, but that
we must avoid being swept up by blockchain hype, and remember to focus on solid user
needs first, before we choose the technology that we will use.
We recommend further that this is best done by organisations convening across sectors to
identify common data infrastructure needs.

2 | Open Data Institute 2016 | Applying blockchain technology in global data infrastructure

Introduction
Data is essential for the modern age; it is infrastructure for the whole economy. It underpins
public service transformation, business innovation and democratic engagement. It connects
sectors.
Data infrastructure includes technology, processes and organisations. The Open Data Institute
(ODI) is in the process of developing its design principles for strengthening the data infrastructure
we rely on to build tools, products and services that benefit everyone. Sometimes we talk
about our data infrastructure like we do our road infrastructure: roads help you get from A to
B data helps you get to a decision. Sometimes we need a super highway or a motorway;
at other times a country lane. Data infrastructure needs to be as reliable as necessary and as
open as possible.
A new class of data infrastructure technologies has recently emerged, known as distributed
ledgers. Blockchains are one specific type of technology in that domain, and though far from
the only type, they are receiving a lot of attention. Blockchain technology emerged from the
digital currency Bitcoin, and has been hailed as a revolutionary step forward for data storage
and the decentralisation of computer systems.
With the support of Deutsche Bank, the ODI has been exploring emerging data infrastructure
technologies including distributed ledgers and blockchains their potential and the issues
they might bring. We have explored the technology landscape, raised questions based on
our recent work on data infrastructure and personal data, and begun to investigate technical
implementation. Each of these activities has led to new insights into the current state, and
potential, of these technologies.
While there are many technologies relevant to data infrastructure, we have focused recently on
blockchains because of the hype surrounding them, and a concern that it may lead to negative
outcomes if they are used in the wrong ways.

Peak of Inflated Expectations

Plateau of Productivity

VISIBILITY

Slope of Enlightenment

Trough of Disillusionment
Technology Trigger

Gartner Hype Cycle

TIME
3 | Open Data Institute 2016 | Applying blockchain technology in global data infrastructure

Many technologies go through a similar hype cycle; the peak of inflated expectations can
lead to the trough of disillusionment, before eventually we reach the plateau of productivity
(Gartner, 1995). Blockchains and distributed ledgers are definitely on the upwards slope
towards the peak, and in our work we hope to help accelerate through the cycle, and rather
than increasing the height of the peak, to instead raise the floor of the trough. Good ideas die
in the trough, and we want to help those good ideas survive through the cycle to success.
It is only together that we will be able to answer the question of whether blockchains are as
fundamental for forward progress in society as the Magna Carta or the Rosetta Stone (Swan,
2015, p. viii), whether they are irrevocably bound to the failed Bitcoin experiment (Hearn,
2016), or whether the truth lies somewhere in between.
In this report, we hope to help the reader form their own opinion on that emerging question,
and provide a roadmap for further exploration.

The foundations of blockchain technology

Blockchains provide a way to store information so that many people can see it, keep a copy of
it, and add to it. Once added, it is very difficult to remove information. This can reinforce trust
in a blockchains content.
At its heart, a blockchain is simply a new kind of database, with a unique set of properties.
Rather than being stored in a single location, this database is shared across the Internet, with
many people holding a copy of it. The synchronisation between all these different copies
so that there is still a single recognised truth is one of the unique aspects of blockchain
technology.
In his article Avoiding the pointless blockchain project, Greenspan (2015) outlines these
properties as follows:
Shared read: blockchains are a structured data store that many people can read
Shared write: as well as read, many people can write data into the database
Absence of trust: the different writers do not have to trust each other not to
manipulate the shared database state (i.e. you cannot change information I have
added)
Disintermediation: there is no need for a trusted intermediary to enforce access
control (or none is available)
Transaction interaction: records in the database depend on and link to each
other
Validation rules: the rules around database transactions are well-defined,
such that anyone with a copy of the database can validate that it has been
maintained correctly

4 | Open Data Institute 2016 | Applying blockchain technology in global data infrastructure

If an application does not need to meet all the above criteria, then it does not require a
blockchain solution; there are much more mature technologies available that meet many of
the above criteria (including traditional databases) and can also support distributed service
and organisational models.

How data is stored in a blockchain

The fact that writers do not need to trust one another not to change information in a blockchain
is down to how data is stored in the system. A blockchain stores a series of transactions
which can be data of any sort in blocks, which get added one after the other in a continuous
chain. Importantly, blockchains are considered append-only, or immutable, data stores. Data
can be added to the chain, but because each item depends on the one before it, data cannot
be removed or changed without recalculating every subsequent transaction.
Because of this, once data is embedded in a blockchain it cannot be altered without that
change being detected and potentially rejected by the other nodes in the network. This is
useful for data that users need to trust, because it provides a guarantee that the data has not
been changed since it was put in the blockchain, but also has other implications, which we
discuss below.

Blockchain distribution and data consensus

The disintermediation property arises because blockchains are distributed; there is no central
storage location, no primary copy. Blockchains are maintained by a peer-network of nodes;
every node has a copy of the blockchain and has equal authority to add to it. Every node
publishes that data for other nodes to pick up and use.
The size of the network is important, and ensures the immutability of the database. A single
node could change historical transactions and recreate a valid blockchain, but because other
nodes in the network check the data before they accept it, it becomes impossible for a single
node to make a historical change. More than 50% of the network needs to agree on the truth
for it to be accepted. Conversely, there is a danger that if more than 50% of the nodes (or
miners, in Bitcoin parlance) decide to change history, they can.
As new blocks may be added on the end of the chain by many nodes at once, blockchains
need a resolution mechanism to decide which block is accepted by the network. Bitcoin uses a
proof of work algorithm, where nodes must prove they have solved a complex cryptographic
puzzle, purely as a way of deciding which node gets to add its data first. Other systems use
different methods such as the proof of stake algorithm, which requires that nodes prove
they own a certain amount of an asset, such as the cryptocurrency of the system. Choosing a
resolution mechanism for a blockchain can greatly affect the scalability of the system and how
much energy it uses, particularly as proof of work algorithms require a lot of computing power.
5 | Open Data Institute 2016 | Applying blockchain technology in global data infrastructure

Distributed ledgers, blockchains, or the blockchain?

Confusing terminology is often used around blockchains, which are themselves a type of a
more general distributed ledger technology (DLT).
Some people have a tendency to talk about the blockchain, which normally implies the
blockchain that underpins Bitcoin, as opposed to a blockchain.
The ODI uses blockchain terminology as follows:
Distributed ledger technology: the concept of a distributed shared database of
transactions
A blockchain: a particular implementation of a distributed ledger system that
uses the same basic technical architecture as Bitcoin, but could be independent
of that system
The Bitcoin blockchain: the blockchain that runs the Bitcoin digital currency. In
the wider world, this is often confusingly referred to simply as the blockchain,
but we avoid that usage

Permissions
While classical blockchains (such as Bitcoin) are public and allow anyone to write to them,
some distributed ledgers work differently. Distributed ledgers can be classified as:
Public: anyone may have a copy of the database and anyone may write to
it. This is the classic Bitcoin-style public ledger. Sometimes referred to as
unpermissioned or permissionless
Permissioned: anyone may have a copy of the database, but only certain
parties may write to it an audit log for government information, for example
Private: only certain authorised users have access to the database, whether for
reading or writing a blockchain used internally within an organisations firewall,
for example
The concepts of permissioned and private blockchains are controversial in the blockchain
development community, as they may reintroduce trusted intermediaries into the system and
therefore undermine one of the unique aspects of the technology.
Permission types in blockchains are aligned with the closed / shared / open components of
the ODIs Data Spectrum, which classifies data by who has access to it.1

See http://theodi.org/data-spectrum.

6 | Open Data Institute 2016 | Applying blockchain technology in global data infrastructure

Closed data is only available to a single individual or organisation (similar to a private

blockchain); shared data is shared under various permission methods with a limited group
(similar to a permissioned blockchain); and open data is freely available to all to access, use
and share (similar to a public blockchain).

The three generations of blockchain technology

Blockchain technologies and their applications generally fall into one of three generations,
defined by their use and capabilities:
Digital currencies (Blockchain 1.0): storing and executing transactions carried
out in purely digital currencies (i.e. Bitcoin)
Coloured coins (Blockchain 2.0): using small Bitcoin transactions to represent
assets that are transferrable but not digitally stored on a blockchain, such as
real estate or diamonds
Smart contracts (Blockchain 3.0): transactions store executable code which
runs when certain criteria are fulfilled, allowing complex behaviour and fully
distributed applications (Dapps)

7 | Open Data Institute 2016 | Applying blockchain technology in global data infrastructure

Section summary
Blockchains (or distributed ledgers) are a new type of database
with unique properties
Blockchains suit applications where multiple readers and writers
need to use the database without a trusted third-party
Data stored in a blockchain cannot be changed afterwards
Public, permissioned and private blockchains are differentiated
by who can read or write data within them
Smart contract systems enable blockchains to run executable
code, allowing much more complex behaviour

The technology landscape

The recent explosion of interest in blockchains and distributed ledgers has created much
speculation on the application of these technologies. While there are many ideas being brought
to the fore, there seem to be few actual implementations being referenced.
As part of our research, we explored this rapidly evolving space to pull out tangible examples
of tools, technologies and applications currently in operation.
Our efforts focused primarily beyond financial applications. Because of the association with
Bitcoin, there has already been a great deal of research on blockchains for cryptocurrencies
which we did not want to replicate. We also focused on finding applications that illustrated the
use of the technology in solving real problems and examined a variety of tools that aimed to
make the underlying technologies more accessible to developers.
Given the rapid growth in this field, this section only captures a snapshot of the activity that
was being undertaken in early 2016. It should however serve as an illustration of the maturity
of the field and a guide to potential applications.

Trends and observations

From our landscape research, we were able to distinguish a number of interesting trends and
observations.

8 | Open Data Institute 2016 | Applying blockchain technology in global data infrastructure

The baseline technology is unclear

Our first observation is that the baseline technology underpinning different tools and solutions
for blockchains is often difficult to understand. For instance, in many cases it is hard to
distinguish between the use of a blockchain versus the Bitcoin blockchain. This indicates
that the technology stack for distributed technologies is still very fluid and in the process of
being defined.
The fluidity of the technology stack presents further issues in classifying any particular
solutions as either tools, technologies or applications. For instance, Bitcoin is an application
of blockchain technology, but it is also being used as a platform on which other applications
are built.

Emerging platforms
At the current stage of development, platform technologies have not yet fully emerged. Many
blockchain systems are bespoke, and combine both platform and application functions. As
the field matures, commoditised platforms that allow applications to converge on a common,
underlying technology stack are likely to appear.
Some of the most interesting applications, at least from a technical perspective, are those that
use multiple distributed technologies and blockchain implementations in a single service. One
example of this is Alexandria,2 a system designed to allow users to distribute digital content.
It uses Bitcoin to handle payment, Florincoin3 (an alternative cryptocurrency) to create a
searchable index, and the InterPlanetary File System4 or IPFS (a distributed hash table or
DHT) for content storage and distribution.
Such emerging hybrid stacks show that complex services can be entirely implemented in a
shared, distributed manner. We find one promising stack to be the combination of Ethereum5
for the application or logic layer, IPFS for bulk storage, and BigchainDB6 for database storage.
This shows the beginning of a true platform ecosystem mirroring the common LAMP7
webstack.

2
3
4
5
6
7

See http://blocktech.com.
See http://florincoin.org.
See https://ipfs.io.
See https://ethereum.org.
See https://www.bigchaindb.com.
Linux, Apache, MySQL, and PHP.

9 | Open Data Institute 2016 | Applying blockchain technology in global data infrastructure

The blockchain field is focused more on technology than

real-world problems
Startups working with blockchains tend to be highly focused around technology. Many
distributed technology startups begin by laying out their technical vision, rather than focusing
first on an end-user need.
Many blockchain solutions are also related to other disruptive and emerging trends in the tech
sector, such as the Internet of Things and the sharing economy. The focus on technology over
real-world problems is a common feature in those areas as well, and can be compounded by
the combination.

Existing applications of blockchain technology

As part of our research into distributed ledgers, we compiled a list of the tools, technologies
and applications that we looked into.8
From marriage on the blockchain (Woods, 2015) to AI gladiators (Borah, 2015), blockchain
technologies are being experimented with in many different ways. In a crowded field where
many aim to capitalise on associated hype it can be hard to identify useful applications of
blockchain and distributed technologies that might have a lasting effect.
We have identified several categories of notable applications. These are generally focused
around a real-world problem that requires the unique properties of the technology. The best
examples are from organisations that were committed to solving a problem before they adopted
blockchains as a possible solution, as opposed to many who attempt to fit the problem to
thetechnology.
The following sections identify the areas we think are worth watching, along with examples of
organisations in the space.

Document and Intellectual Property verification

The following services are designed to prove the existence of digital documents by recording
a unique identifier and timestamp in public blockchains.
SealX and GeniusX: contract notary services and IP certification systems
(bitproof.io)
Blocksign: a document verification service (blocksign.com)
Factom: a service for distributed record storage (factom.org)
8

This list is published under an open licence for anyone to access, use and share at dlt-research.labs.theodi.org.

10 | Open Data Institute 2016 | Applying blockchain technology in global data infrastructure

Monitoring supply chains

The following services aim to improve supply-chain transparency and efficiency by tracking
materials and products through the production process. How physical products are reliably
linked to digital assets is an open question.
EverLedger: stores records of diamonds to prevent insurance fraud and
diamond smuggling (everledger.io)
Provenance: tracking supply chain provenance (provenance.org)
Skuchain: tracking supply chain provenance, especially for food (skuchain.com)

Building a peer-to-peer economy

The following services aim to automate various sharing economy transactions using smart
contracts, without the need for centralised platforms.
Slock.it: a service for renting, selling and sharing things that can be locked;
combining IoT with blockchain (slock.it)
Airlock.me: a decentralised electronic security service offering keyless access
protocols for smart property (airlock.me)
Lazooz: a decentralised transportation platform for ride-sharing (lazooz.org)
WeiFund: a crowdfunding and equity platform based on Ethereum (weifund.io)

Governance
The following services aim to use smart contracts to inform decision-making processes,
and in some cases create Decentralised Autonomous Organisations (DAOs) or even
decentralised nations.
Colony: a tool to help organise distributed, democratic company-like entities or
organisations (colony.io)
Freecoin: an alternative currency for democratic social organisations
(freecoin.ch)
BoardRoom: a service that organises decision-making processes
(boardroom.to)
BitNation: a collaborative platform for governance for organisations and even
virtual nations (bitnation.co)

11 | Open Data Institute 2016 | Applying blockchain technology in global data infrastructure

Digital content distribution

The following services aim to provide a means to distribute digital content equitably whether
for IP protection, or fair remuneration in a way that does not rely on centralised platforms.
ALEXANDRIA: a standard allowing users to distribute digital content
(blocktech.com)
UjoMusic: a service for distributing creative digital products, such as music,
focused on generating greater transparency, fairness and profitability
(ujomusic.com)
Blockai: a digital asset distribution service (blockai.com)

False promises
While there are promising applications as identified above, a great many of the ideas out
there are vapourware, with no viable implementation or model. For instance, development on
Honduras land registry, which is being turned into a distributed ledger by Factom, has stalled
with no working system (Rizzo, 2015). This is particularly important, as this example is used
repeatedly to show that blockchains can be useful in traditional government applications, but
has not yet shown any results.
There are also many instances of old ideas being brought back to life with an application of new
technology sheen. For instance, tracking benefit payments and how they are spent is a policy
idea that has been proposed and rejected in the past, but is now reappearing with blockchains.
Many such projects failed for good reasons in the past, and the addition of blockchains will not
change those reasons, which is more often social or cultural than technological.

Section summary
The distributed ledger technology stack is still emerging, with
unclear boundaries between platforms and applications
Many blockchain applications are technology-centric rather than
focused on user needs
While there are good examples of blockchains being used to
build useful services, they are currently in the minority
Interested parties must be alert to blockchain hype and myths
when surveying the landscape, and should focus on those
applications solving real problems

12 | Open Data Institute 2016 | Applying blockchain technology in global data infrastructure

Scaling and linking blockchains

Blockchains are emerging from their origins in cryptocurrencies and being explored as a
mechanism for storing data of other kinds. We are very early in our understanding of when
and how best to use blockchain technology. We need to anticipate and plan for what happens
when blockchains scale from low levels of use to potential ubiquity for other applications, like
recording marriages (Alexander, 2014), registering land ownership (van Wirdum, 2015) and
maintaining supply-chain provenance (Provenance, 2015).
Blockchains are maintained by a distributed network of nodes: computers that store the
blockchains and may add data to them. There are drivers for having a few blockchains that are
each maintained by a large number of nodes and for having many blockchains that are each
maintained by a small number of nodes. It is likely it will end up somewhere in the middle. The
suitable number of blockchains needed to support the applications that use them will change
over time. We have to ensure that it is possible to adjust: to split blockchains or to merge
blockchains as required, and to migrate data between them.

What are the drivers for having fewer blockchains?

Blockchains are attractive as a data store because their distributed nature makes them robust
and tamper-proof. Robustness ensures that the data is always available. A blockchain being
tamper-proof guarantees data integrity; even if some nodes are compromised, the other nodes
will not accept changes they make to the blockchain.
A blockchain that is maintained by a single node could be struck by a hardware failure, or
could rewind and rewrite the blockchain it is maintaining without detection.
Blockchains that are only maintained by small numbers of nodes can get into situations where
the majority of the network is owned by a single organisation or cartel. This happened with
the GHash Bitcoin mining pool in 2014 (Hern, 2014), and was the reason behind Onenames
recent migration from Namecoin to the Bitcoin blockchain (Onename, 2015). When more than
half a blockchain is owned by a single organisation, it is possible that they can collaborate to
alter the content of the blockchain or to accept invalid transactions.
The fact that small networks of nodes undermine the utility of a blockchain is a driver towards
having a few, large-scale blockchains maintained by many nodes.

13 | Open Data Institute 2016 | Applying blockchain technology in global data infrastructure

What are the drivers for having more blockchains?

The size of a blockchain grows over time because it is an append-only data store: you can add
data to a blockchain, but you can never remove it.
At the time of publication, the Bitcoin blockchain is around 70GB in size. It has been growing
steadily by about 2.5GB/month (though that rate is increasing). With a limit of 1MB/block
and one block every 10 minutes, the maximum rate of increase in size will be just over 4GB/
month. While there is work underway to change this, the choice of size is controversial and the
outcome unknown in the long-term.
The Bitcoin blockchain is relatively small, as transactions take up very little data. Other
applications for blockchains may require more storage, or a speedier rate of growth (larger
blocks and/or more frequent addition of blocks to the chain).
Every node in a blockchain network needs to be able to store and process the entirety of the
blockchain. Blockchains that are used by lots of applications will be large in size. The vast
majority of data within a blockchain supporting multiple applications will be irrelevant to the
application any particular node is interested in. Some nodes might only be interested in land
registry data, some only in statements of copyright ownership.
Large blockchains require nodes that are interested in a given application to take on all the
data from other applications using the blockchain. They might hesitate not only because of
size but due to ethical concerns about the data those blockchains contain. These are drivers
towards having more, smaller and more specialised blockchains.

Linking blockchains
To find the balance between a few large and many small blockchains, blockchains will need
to be able to split and merge over time, and for this to work, they will need to be able to refer
to each other.
It is also important that new data infrastructure, built on blockchains, is compatible with the
Web we have already. We will need to be able to link to items in a blockchain from outside,
on the wider Internet. After all, data is at its most useful when it can be referred to and linked.
Therefore, transactions on blockchains should have stable URLs and an equivalent of HTTP
redirection to point to updated locations. We will need a standardisation effort to create
standard URL schemes for blockchain transactions, if these systems are to link into our global
data infrastructure.

14 | Open Data Institute 2016 | Applying blockchain technology in global data infrastructure

Data standards and archiving

How do we standardise storage in systems so that we get a single network of data, as
opposed to having to use a different storage system every time we want a new type of
information? What are the data protocols for distributed storage? How do we talk about, and
perhaps enforce, ownership and licensing?
Like other Internet phenomena, blockchains should be automatically archived for posterity.
Whatever the archiving institution whether the Internet Archive or something else they will
need to be able to discover the data that need archiving, automatically, and standards will be
an important part of this.

Section summary
Fewer, large blockchains will be more secure
More small blockchains will be more scalable
Blockchains will need to be able to split or merge over time to
balance their size and security
We need to create URL and related standards to link blockchains
into our global data infrastructure

15 | Open Data Institute 2016 | Applying blockchain technology in global data infrastructure

Privacy and security

The irreversibility and transparency of public blockchains mean they are probably unsuitable
for personal data. We need to be careful when designing blockchain systems not to infringe
on peoples privacy, and to account for a world in which we have doxing, identity theft and the
right to be forgotten.
The examples in this section are not necessarily mentioned because anyone has suggested
providing these kinds of data in blockchains. But blockchains are being investigated for a
range of purposes, across the Data Spectrum, and we need to explore their limits.

What irreversibility means for privacy

There are types of data for which the difficulty of change could lead to problems particularly
personal data.
For example, in the UK personal insolvency notices are published in the London Gazette when
people are declared bankrupt. They are published in paper copies of the Gazette (which are
relatively hard to get hold of or search) and published on the Web. Requests are sometimes
made for notices to be removed from the website to avoid them being as easy to find, because:
people do not want their personal insolvency to be known about (they may have
a right for that event to be forgotten under the law)
people have made a transition to a new gender; they may have a legal right for
previous data to be altered to ensure it is interpreted according to their new
gender
the notice includes the current address of a person who is attempting to avoid
an abuser, and thus the notice increases the risk of discovery and harm to that
individual
It is easy to imagine other scenarios where blockchains could be used to hold data about
people that might seem innocuous at the time but where the situation changes such that data
should no longer be held in the same way. For example, a recent change in UK law means that
company directors birth dates are no longer published by Companies House.9 Data about
people applying for planning applications, holders of licences and those in public office is
routinely published and could change in similar ways.
Beyond government, blockchains offer a great opportunity for people to collaboratively create
datasets in a peer network without a central authority. We can see the evidence of the demand
for data rating teachers, about landlords, about where sexual offenders live all of which
would have a significant privacy impact if published (Neyfahk, 2015).
9 UK Parliament (2015) Small Business, Enterprise and Employment Act, 96(3). Available at: http://www.legislation.gov.uk/
ukpga/2015/26/section/96/enacted. [Accessed 2016-05-30].
16 | Open Data Institute 2016 | Applying blockchain technology in global data infrastructure

Arguably, we have a situation even now where once data is published on the Web, it can never
truly be removed. Certainly removing a page from Googles search results under a right to be
forgotten does not actually remove it from the Web, rather it makes it harder to find. What is
different about blockchains is that if a court were to attempt to legally compel the removal of
data from them, it would be both hard to do and have very disruptive side-effects, which we
explain in the next section.

Removing data from a blockchain

Imagine a blockchain were found to contain the names and addresses of children at risk
of abuse (we deliberately pick something most people would say should not be publicly
available). To clear that data out, over half the nodes that maintain the blockchain would have
to work together to rebuild the blockchain since that data was added. This process is similar to
rebuilding from a backup: while being rebuilt, the blockchain would be rewound to a previous
state, days or weeks or even more out of date. During this time (and rebuilding blockchains
deliberately takes time), the data would not be up-to-date. This might also be a time when
unwanted changes to data that was trustworthy could creep in.
Alternatively, a court could try to compel the entire set of nodes to be shut down. Putting
aside that nodes may reside in different legal jurisdictions, that would have huge practical
implications. It would mean removing all the rest of the data held in the blockchain as well as
the target of the order. Given the use of blockchains that people envisage often involves the
same blockchain holding many types of data and supporting many types of applications, there
is a real risk that, pragmatically, bad data simply has to continue to exist in order to prevent
massive disruption to the provision of good data for other applications.

What transparency means for privacy

When new data is added to a blockchain, peers in the network check it to ensure it is valid, to
avoid fraud by rogue nodes. The data that the peers need to check must be stored transparently
in the blockchain.
Sometimes as with Bitcoin personal data is required in order to verify that a transaction in
the blockchain is valid. For a node to check a Bitcoin transaction, it must have access to all
previous transactions and be able to check that the person giving the Bitcoins actually has
them to give. It must therefore be possible for any node to reconstruct the full financial history
of every person exchanging Bitcoins: how many they have, where they got them from, and
whom they spend them with.
This is intensely personal information, highly revealing of the details of someones life; after
all, would you publish all your bank statements on the Web? The only shield is the pseudoanonymity of the Bitcoin address, which is easily breached if the address is associated with
17 | Open Data Institute 2016 | Applying blockchain technology in global data infrastructure

a donate button on a blog. Those who trade Bitcoins are therefore advised to hold several
addresses and not to transfer Bitcoins between those accounts to avoid others linking them
together. It is not clear how many Bitcoin holders are aware of this risk, and security-throughobscurity is well-known to be insufficient.
Some of the proposed uses for blockchain such as to record auditable benefits payments
threaten to expose this kind of information about a much wider range of people, the benefits
they receive and with whom they spend them.
Blockchains do not have to expose personal data directly to reveal private information about
people. A blockchain recording visits to health practitioners (including midwives, mental health
teams and AIDS clinics) does not need to include the entirety of someones health records to
reveal information about them. Much like phone records (Mayer & Mutchler, 2014) or browsing
histories, this metadata may be sufficient to reveal personal details.

Designing privacy-protecting blockchains

There are ways to design the content of blockchains and the network that supports them, that
limit the level of disclosure that they entail.
First, whereas anyone can join the Bitcoin blockchain, it is possible to use a permissioned
distributed ledger as a method of resolving conflicts within a peer-to-peer network of trusted
nodes. When nodes can be trusted, they can control what becomes public, and therefore
hide data in the blockchain that should not be shared. The security of all the nodes in such a
trusted network needs to be guaranteed as every node will have a copy of all the relevant data,
and the network needs to be protected against spoofing, but, in general, if you have a trusted
network many privacy issues are no more problematic than they are in centralised systems.
Second, blockchains could be used purely to provide a timestamp for information held
elsewhere. Content that could require redactions in the future can be made available as usual
on the Web, with transactions in the blockchain containing simply a pointer to the content
and its hash. If the content needs to be taken down, the fact that the content existed at a
given point in time could remain in the blockchain; the stored hash alone would not enable
the reconstruction of the removed content. If the content needs to be changed, the existing
hash would no longer match the content, so applications are able to detect that something
has changed. If the changes are legitimate, the reasons for them could be stored in a later,
overriding, transaction.
This pattern of using blockchains purely as a timestamping mechanism and not as a data store
has the additional benefit of being more likely to scale in the face of large amounts of data needing
to be recorded. On the other hand, it shifts the burden of robust, distributed data storage, which
is one aspect of the interest in blockchains, to other protocols, whether the Web, BitTorrent, or
IPFS. Projects that are focused on robust distributed storage might not need a blockchain at all.
18 | Open Data Institute 2016 | Applying blockchain technology in global data infrastructure

Finally, it is possible to encrypt data stored within a blockchain. The main problem with this
approach is that if the decryption key for encrypted data is ever made public, the encrypted
content is readable by anyone with that key; there is no way of encrypting the data with a
different key once it is embedded within a blockchain. Conversely, if the key is ever lost, the
data cannot be read. And there is the problem of sharing the key for the data amongst all those
who legitimately need to be able to read it.
On top of that, a well-used blockchain will be a potentially eternal datastore, and over a
sufficiently long period any encryption will be broken whether by discovery of loopholes,
backdoors, or the advent of new techniques such as quantum computing.
Regardless of the approach taken to designing blockchains, every blockchain contains
transaction data. That data needs to be designed so that it is not disclosive in and of itself,
which may be a tricky balance as that data might also be necessary to assess whether the
transaction is valid and therefore prevent fraud or errors. Transactions should also be designed
so that they cannot be used to add comments that might include personal data.
Blockchains are not necessarily bad for privacy; it all depends on how they are designed. As
stated by Vitalik Buterin (2015): blockchains do not solve privacy issues, and are an authenticity
solution only. Anyone experimenting in the area should be thinking through the implications.
As the ICOs guidance around privacy by design suggests, designers should be carrying out a
privacy impact assessment or similar process up-front, to ensure that the transparency of the
information stored in the blockchain does not infringe on peoples privacy. Unlike with other
technologies, getting it wrong is really hard to reverse.

Section summary
Immutable data storage in blockchains may be incompatible with
legislation which requires changes to the official truth
Once added, removing data from blockchains can be impractical
and highly disruptive
Even if personal data is not stored on a blockchain, metadata can
be sufficient to reveal personal information
Blockchains by themselves are not a solution for personal or
private data
Any encryption used is likely to be broken in the future
Bad blockchain design decisions are very hard to reverse

19 | Open Data Institute 2016 | Applying blockchain technology in global data infrastructure

Practical experimentation
During early 2016, the ODI Labs team carried out some practical exploration of blockchain
technologies and how they could be applied to data infrastructure. This was done in order
to understand the technology more deeply, but also to drive out further issues through
experimentation.

Creating a blockchain
The first step was to create a blockchain to experiment with. The team opted to build on
Multichain,10 a blockchain software toolkit that supports Bitcoin and other blockchains, but
adds support for metadata and custom assets.
As software developers, it was reasonably simple to get a system up and running, although
this would be confusing and difficult for anyone without technical experience. There is a clear
need for user-friendly prototyping systems so that non-technical users (for instance policymakers) can explore their own use cases.

Storing data on a blockchain

The team built a small open source software library11 around the blockchain that would allow
arbitrary information to be stored with any transaction. In the standard Bitcoin-style system
we were using, only a small amount of information can be stored with each transaction.12 The
total amount of information that could actually be stored was only 640 bits per transaction. By
comparison, a single tweet can hold up to around 2650 bits (Munroe, 2014).
One of the use cases in mind for this data storage was to be able to store a cryptographic hash
of the contents of a URL. The hash is a fingerprint for the data pointed to by the URL; the
data cannot be recovered from it, but a third party could use the hash to detect if the data has
changed since the hash was created.

Auditing open data

The Government Digital Services in the UK has written about guaranteeing the integrity of a
register (Potter, 2015), explaining how a distributed database could be used to check that an
official register has not been tampered with.
For our exploration, we decided to follow the example proposed, and create a blockchain
10 See http://www.multichain.com.
11 Available at https://github.com/theodi/multichain-client.
12 Using OP_RETURN metadata.
20 | Open Data Institute 2016 | Applying blockchain technology in global data infrastructure

version of the Food Standards Agency premises ratings dataset. If the FSA website was
compromised, and ratings altered, the information on our blockchain which could not be
deleted or altered could be used to detect the tampering. We successfully imported the full
rating data (rather than just an audit hash) into the blockchain, storing the premises ID, along
with the inspection date and rating.

Searching blockchains
While the team successfully stored open data in a blockchain, various issues emerged when
trying to use the data.
Firstly, finding a particular record is non-trivial. To find a particular food hygiene rating for a
premise, we needed to carry out a brute-force search of the blockchain. Many search systems
are available, and indeed there are many sites for searching and viewing the Bitcoin blockchain,
which index the blockchain into a searchable database. The problem then becomes that you
are now relying on the integrity of your search index to ensure you find the right information.
If the idea of the blockchain is to prevent centralisation (due to lack of trust or other reasons),
we cannot trust search indexes maintained by others. In order to find anything, not only would
each blockchain client have to have the entire blockchain stored, but also store (and update)
a search index built from it, which could be considerably larger. There is a need for search
indexing to be built into blockchain software if it is to be used for data.

Incentivising maintenance
The test blockchain we created consisted of only two nodes, mining and sending fake currency
to each other. The key to the integrity of blockchain technology is its decentralised nature.
Once one person or organisation owns the majority of nodes in a blockchain network, this
integrity breaks down, as discussed earlier.
If we were to open up our audit blockchain to the wider public, how would we incentivise
mining? With Bitcoin, once a miner adds new transactions to the blockchain, there is a fixed
reward (currently 25 BTC, almost 7,500 in todays exchange rate). As the size of the blockchain
network increases, so does the computational power required to add new blocks (if using a
proof of work algorithm); ODwyer and Malone (2014) have compared the electricity usage
of the Bitcoin network to that of Ireland. As a result of this, and the rewards at stake, there are
entire server farms devoted to Bitcoin mining.
Currently, the test blockchain we created is tiny, and a few well-intentioned people might be
happy to mine it for free. However, as the blockchain grows, mining will get harder, and the
worthless currency the blockchain rewards miners with wont help them cover the electricity
bills the computation will end up costing them.
21 | Open Data Institute 2016 | Applying blockchain technology in global data infrastructure

Section summary
Simple prototyping tools are required so that non-technical users
can explore blockchain use cases
Only a very small amount of information can be encoded with
each transaction in current blockchain software (including
Bitcoin)
Distributed search indexes will be required for trusted search of
blockchain data, increasing the space requirements

Conclusions
Data is part of the infrastructure of the modern world. It is essential to the operation of society,
and it is vital that we learn how to build, maintain and strengthen our data infrastructure.
Distributed ledgers are a potentially important technology for enabling a shared data
infrastructure, and are worthy of investigation.
However, new technologies always go through a hype cycle. The challenge at the beginning of
that cycle is to identify the uses and applications that will stand the test of time. Blockchains
are unusual in that mistakes made in early deployment could last a long time, and could cause
significant damage, especially if deployed carelessly with personal data.
Blockchains could be used to build confidence in government services through public
auditability. They also hold great potential for collaborative maintenance of data assets,
enabling widely distributed data collection and publishing for applications such as supply-chain
information. Smart contracts also have promise for the future across many applicationareas.
However, in our research we have seen many cases where people attempt to bolt old, failed
or impossible policy and business ideas onto the new technology, or to unnecessarily reinvent
things that work perfectly well. Many other cases show familiar organisational models being
rebuilt as permissioned ledgers based on blockchain technologies, but this ignores the core
innovation of the technology and its promised transformation.
We have seen many ideas that would put new personal data into blockchains but learnt that,
if misused, this will create significant new privacy issues. The core problems that blockchain
technologies help to address of distributed maintenance by collaborating organisations is
of growing importance and an area that shows some promise, but few are considering it.
Success in data infrastructure design will come from convening sectors (such as finance,
22 | Open Data Institute 2016 | Applying blockchain technology in global data infrastructure

agriculture, or healthcare), identifying common challenges and then determining which

technology approaches whether blockchains or not are the most appropriate in helping to
address them.
Blockchain technology is a new and powerful tool in our toolbox. We must use it when it is the
right tool for the job at hand.

Recommendations
Organisations must remember to start with user needs, rather
than preselecting technologies that may or not be appropriate
Organisations considering blockchain technology should be
aware that there are many other distributed technologies
available, and it is important to assess needs properly in order to
arrive at the right technology choice
Organisations should convene across sectors to identify
common data infrastructure needs, and then decide how those
can best be met and which technologies should be used
When experimenting with blockchain technologies, researchers
and developers must be aware of the privacy implications of
storing information in a public immutable database; what is done
cannot be easily undone

About the ODI

The Open Data Institute (ODI) connects, equips and inspires people around the world
to innovate with data. It is independent, nonprofit and nonpartisan, founded in 2012 by
Sir Tim Berners-Lee and Sir Nigel Shadbolt. From its headquarters in London and via its global
network of startups, members and nodes, the ODI offers training, research and strategic advice
for organisations looking to explore the possibilities of data.

23 | Open Data Institute 2016 | Applying blockchain technology in global data infrastructure

Bibliography
Alexander, R. (2014). The First Blockchain Wedding. [Online] Available at: https://bitcoinmagazine.com/articles/firstblockchain-wedding-2-1412544247 [Accessed 2016-05-20].
Borah, P. (2015). DAO Wars. [Online] Available at: https://github.com/consensys/dao-wars [Accessed 2016-05-23].
Buterin, V. (2016). Privacy on the Blockchain. [Online] Available at: https://blog.ethereum.org/2016/01/15/privacy-on-theblockchain/ [Accessed 2016-05-20].
Hern, A. (2014). Bitcoin currency could have been destroyed by 51% attack. The Guardian [Online] Available at: https://
www.theguardian.com/technology/2014/jun/16/bitcoin-currency-destroyed-51-attack-ghash-io [Accessed 201605-20].
Hearn, M. (2016). The resolution of the Bitcoin experiment. [Online] Available at: https://medium.com/@octskyward/theresolution-of-the-bitcoin-experiment-dabb30201f7 [Accessed 2016-05-20].
Gartner. (1995). Hype Cycle Research Methodology. [Online] Available at: http://www.gartner.com/technology/research/
methodologies/hype-cycle.jsp [Accessed 2016-05-20].
Greenspan, G. (2015). Avoiding the pointless blockchain project. [Online] Available at: http://www.multichain.com/
blog/2015/11/avoiding-pointless-blockchain-project/ [Accessed 2016-05-23].
Mayer, J. & Mutchler, P. (2014). MetaPhone: The Sensitivity of Telephone Metadata. [Online] Available at: http://webpolicy.
org/2014/03/12/metaphone-the-sensitivity-of-telephone-metadata/ [Accessed 2016-05-20].
Munroe, R. (2014). Twitter. In: What If? London: John Murray, p. 217221.
Neyfahk, L. (2015). Californias Sane New Approach to Sex Offenders. [Online] Available at: http://www.slate.com/articles/
news_and_politics/crime/2015/04/california_s_sane_new_approach_to_sex_offenders_and_why_no_one_is_
following.html [Accessed 2016-05-20].
ODwyer, K. and Malone, D. (2014). Bitcoin Mining and its Energy Footprint. [Online] Available at: https://karlodwyer.github.
io/publications/pdf/bitcoin_KJOD_2014.pdf [Accessed 2016-05-21].
Onename, (2015). Why Onename is Migrating to the Bitcoin Blockchain. [Online] Available at: http://blog.onename.com/
namecoin-to-bitcoin [Accessed 2016-05-20].
Provenance, (2015). Blockchain: the solution for transparency in product supply chains. [Online] Available at: https://www.
provenance.org/whitepaper [Accessed 2016-05-20].
Resnikoff, P. (2015). Im Imogen Heap. And This Is Why Im Releasing My Music on Blockchain. [Online] Available at:
http://www.digitalmusicnews.com/2015/10/05/im-imogen-heap-and-this-is-why-im-releasing-my-music-onblockchain/ [Accessed 2016-05-20].
Rizzo, P. (2015). Blockchain Land Title Project Stalls in Honduras. [Online] Available at: http://www.coindesk.com/
debate-factom-land-title-honduras/ [Accessed 2016-05-23].
Swan, M. (2015). Blockchain: Blueprint for a new economy. 1st ed. OReilly Media.
van Wirdum, A. (2015). Honduran Govt to Build Land Registry Initiative on Bitcoin Blockchain. [Online] Available at: http://
cointelegraph.com/news/honduran-govt-to-build-land-registry-initiative-on-bitcoin-blockchain [Accessed 201605-20].
Woods, T. (2015). This couple got married on the blockchain. [Online] Available at: https://technical.ly/brooklyn/2015/11/11/
couple-got-married-blockchain/ [Accessed 2016-05-23].

24 | Open Data Institute 2016 | Applying blockchain technology in global data infrastructure