Blockchaintechnology
Blockchaintechnology
net/publication/330350267
CITATION READS
1 571
1 author:
Ronald Rousseau
KU Leuven
526 PUBLICATIONS 13,769 CITATIONS
SEE PROFILE
All content following this page was uploaded by Ronald Rousseau on 13 January 2019.
(freely available)
Ronald Rousseau
and
Abstract
Before going into details, let me first say that I am not a specialist in computer
science or blockchain technology. Consequently, most of the information in the first
five sections is taken from the sources acknowledged at the appropriate places. That
said, I am very interested to find out how this technology will further develop in the
real world and how scientists will use it in their investigations.
Let me begin with a few words about the history of the blockchain technology and the
bitcoin. Satoshi Nakamoto (2008) – probably a pseudonym - invented the blockchain
2
technology as the underlying technology for the bitcoin. The bitcoin itself was
launched in 2009. This technology deals with the distribution of value, such as money
or property rights, without a trusted third party, such as a bank, a governmental office,
a lawyer or a notary.
Before dealing with the blockchain and the bitcoin we first explain some terms used
in this context: ledgers, public-key cryptography, hash functions, nonces and proof-
of-work.
We recall that in common word use a ledger is a book containing accounts to which,
e.g., debits and credits are posted. As such, ledgers have been at the heart of
commerce since ancient times and are used to record many things, most commonly
assets such as money and property. Walport (2016) notes that through history they
were recorded on clay tablets, papyrus, vellum or paper. However, one may say that
in all this time the only notable innovation has been computerization, which initially
was simply a transfer from paper to bytes. Now, however, algorithms enable the
3
This leads us to a short discussion of different types of digital ledgers. On the one
hand we have public, decentralized ledgers which are accessible to every Internet
user. We will see that the bitcoin belongs to this category. On the other extreme we
have the fully private ledger, where write-permissions are monitored by a central
locus of decision-making. Besides write-permissions there are also read-permissions
involved, which may either be public or restricted. A private blockchain amounts to a
permissioned ledger, whereby an organizational process enables the whitelisting (or
blacklisting) of user identities. The difference between public and private blockchains
is the extent to which they are decentralized, or ensure anonymity. Between the two
extremes, there exists a continuum (Brown, 2015; Allison, 2015) of “partially
decentralized” blockchains rather than a strict public/private dichotomy. Although the
bitcoin belongs to the public part, many future applications will probably belong to the
private or partially decentralized part.
Encryption algorithms which use the same key for both encryption and decryption are
known as symmetric key algorithms. A newer class of "public key" cryptographic
algorithms was invented in the 1970s. These asymmetric key algorithms use a pair of
keys, a public key and a private one. Public keys are used for encryption or signature
verification; private ones decrypt and sign. The design is such that finding out the
private key is extremely difficult, even if the corresponding public key is known. The
best known public-key cryptographic algorithm is the RSA algorithm (Rivets et al.,
1978).
The result of an encryption is often called a hash and the action of performing
encryption is often referred to as hashing. Recall that the verb ‘to hash’ means to
chop something up. A hash function is a mathematical algorithm that takes an input
and transforms it into an output. A cryptographic hash function such as the one used
in the RSA encryption scheme, is characterized by its extreme difficulty to revert, in
other words, to recreate the input data from its hash value alone.
In this section we recall the RSA public-key cryptosystem, largely taken from the
original source (Rivest et al., 1978). In this way we provide a simple example of
hashing and of signing. This is useful to understand the blockchain.
The RSA algorithm involves four steps: key generation, key distribution, encryption
and decryption. But first we explain the notion of a public-key cryptosystem. In a
public-key cryptosystem each user places their encryption procedure E in a public file,
hence the name of this cryptosystem. However, the user keeps the details of the
corresponding decryption procedure D secret. The whole procedure has four
properties:
(d). First deciphering and then enciphering a message M returns the original
message M, i.e. E(D(M)) = M.
5
Properties (a) and (d) mean that the operations E and D are each other’s inverse.
This cryptosystem is used for sending messages and for signatures, which is
performed as follows. Alice wants to send a secret message, MA, to Bob. Encryption
and decryption functions are denoted as EA, DA, EB and DB, depending on the owner.
Now Alice encrypts her message using Bob’s public key, leading to EB(MA). Now Bob,
who is the only person knowing DB performs DB(EB(MA)) and reads MA. Suppose now
that Alice wants to sign a document C. Then she performs DA(C). There is only one E,
namely EA, which leads to EA(DA(C)) = C. As EA is public this means that anyone can
check that indeed Alice has signed the document.
Practically, in the RSA system these properties are realized as follows. The public
encryption key is a pair (e,n) of positive integers.
This means that E(M) is the remainder of the division of Me by n. If we denote E(M)
by C (the enciphered message), then deciphering is done as follows:
The encryption key is the pair (e,n) and the decryption key is the pair (d,n). The
public-key cryptosystem works if knowledge of n and e does not help an attacker in
finding d. This leads to the problem of choosing the keys. The integer n must be the
product of two, randomly chosen, large prime numbers: n = p*q. Recall that n is
public, but when n is large enough it becomes practically impossible to find p and q.
To make factoring harder (for an attacker) the primes p and q should be similar in
magnitude but differ in length by a few digits. The integer d is a large integer which is
relatively prime to (p-1)*(q-1). This means that the greatest common divisor of d and
the product (p-1)*(q-1) is 1. Finally e is determined as the inverse of d modulo (p-
1)*(q-1), i.e.
In their paper Rivest, Shamir and Adleman show that this method satisfies the four
requirements for a safe public-key cryptosystems and provide a simple example. Of
course, since the publication of the original paper the basic RSA-algorithm has been
refined to protect against many types of attacks.
The double spending problem is the following problem. If you have a digital asset,
such as digital money, and you want to give it to somebody else, how can one
prevent you from giving it to two different people at (almost) the same time? As this
asset is digital it is, indeed, easy to make copies. We will show how the bitcoin solves
this problem.
A nonce
Proof-of-Work
This idea was further worked out by Black who proposed the so-called Hashcash
algorithm (Black, 2002). Computation is performed using a cost-function. Its outcome,
in this context referred to as a token, should be easily verifiable, but moderately
expensive (in time or in another commodity) to compute. Preferably this function has
a parameter so that, if necessary, the difficulty related to its computation can be
made to increase. Black calls this cost-function MINT because of the analogy
between creating cost tokens and minting physical money. Later, Nakamoto used a
similar cost-function to mint bitcoins.
After the introduction of the bitcoin, scientists realized that the essence of the
blockchain is actually informational and processual, and does not necessarily relate
to the monetary sphere. In this sense, blockchains may exist without an underlying
token or coin.
In the blockchain each digital record is turned into a unique string of letters and
numbers called a hash (which can be seen as a unique fingerprint) and inserted into
a transaction. A transaction is initiated when the future owner of the digital object
sends his/her public key to the original owner. The object is transferred with a digital
signature. Transactions are broadcasted to a network of miners (the nodes in the
network) who check them. Miners turn pending transactions into a block including the
hash of the previous block, a time stamp and a random number (a nonce) (Pilkington,
2016). From this statement we note one of the main properties of the blockchain
technology, namely that it leads to distributed consensus among participating nodes.
In this way the blockchain technology is able to remove the need for a trusted third
party to guarantee a transaction.
Bitcoin is the special case that the digital record represents monetary value. It was
the first decentralized public ledger, and has acquired a global status.
We first point out the steps to run the bitcoin network – a special peer-to-peer
network - taken from (Nakamoto, 2008) and next provide some details. Nakamoto
proposes the following steps:
5) Nodes accept the block only if all transactions in it are valid and not already spent.
6) Nodes express their acceptance of the block by working on creating the next block
in the chain, using the hash of the accepted block as the previous hash.
Bitcoin transactions
To send bitcoins, you need two things: a bitcoin address or wallet (the public key),
and a private key because the blockchain includes a public-key encryption scheme
similar to the RSA one. A bitcoin address is generated randomly, and is simply a
sequence of letters and numbers. The private key is another sequence of letters and
numbers, but unlike the bitcoin address, this is kept secret. A transaction is initiated
when either, the owner looks up the bitcoin address of the future owner, or, the future
owner of the coin sends his/her public key to the original owner, asking him/her for
money. Every coin is associated with an address, and a transaction in the crypto-
economy is simply a trade of coins from one address to another. Note that there are
no physical bitcoins or even digital ones: only records of bitcoin transactions. Another
striking feature of the blockchain is that public keys are never directly tied to a real-
world identity. Transactions, although traceable, are enabled without disclosing one’s
identity. This is a major difference with transactions in real-world currencies that, with
9
If Alice sends some bitcoins to Bob, that transaction will have three pieces of
information:
An input, stating which bitcoin address was used to send the bitcoins to Alice in the
first place. This secures the chain of transactions.
An amount. This is the amount of bitcoins that Alice is sending to Bob. Note that one
may send more than one bitcoin and that a bitcoin can be split into 100,000,000
pieces. Each such piece, i.e. 0.00000001 bitcoin, is called a satoshi.
An output or target address. This is Bob's bitcoin address. Recall that this address is
public.
To actually send bitcoins, you moreover need your own private key. When Alice
wants to send bitcoins to Bob, she uses her private key to sign a message with the
input (the source of the coins), the amount, and the target (Bob’s address).
She then sends them from her bitcoin wallet out to the wider bitcoin network and all
peers trying to solve blocks collect the transaction records and add them to the block
they are working to solve. Miners verify and confirm transactions and get an incentive
for doing this because of attached transaction fees.
A transaction in the bitcoin world is final once it is included in the blockchain, thereby
becoming simultaneously verifiable by many sources. These fully decentralized
blockchains rest on a consensus mechanism of proof-of-work for validation purposes:
in the case of bitcoin, the “longest chain – the chain with the most proof-of-work – is
considered to be the valid ledger (Swanson, 2015, p.4).
We will not discuss how to mine for bitcoins on your own, leaving that to others. But
we will explain the main idea. The bitcoin blockchain is a chain of transactional
records enriched by a subset of so-called miners who solve difficult computational
problems. Miners anonymously compete on the network to solve a mathematical
10
problem, thereby adding the next block to the blockchain. The reward for finding this
next block, namely ‘newly minted’ coins, is sent to the miner’s public address. Miners
may spend these coins at will, using their private key. However, mining cannot go on
forever. When the bitcoin algorithm was created a finite limit on the number of
bitcoins that will ever exist was set at 21 million. Currently (January 2018), there are
about 18 million and 800,000 bitcoins in circulation. That means that slightly more
than two million bitcoins are still to be discovered. New bitcoins must show a proof-of-
work to be accepted. This proof-of-work (PoW) is the so-called Hashcash PoW
(Pilkington, 2015) proposed by Black (2002). For verifying transactions, and
calculating proof-of-work, bitcoin relies on a specific hashing function, called the
double SHA256 hashing algorithm, wherein the target is a 256-bit number (a number
of the order of 10168). To be accepted by the network the SHA256 hash of a block's
header must be lower than or equal to the current target for the block. The lower the
target, the more difficult (and processing time consuming) it is to generate a new
block. For a block to be valid, it must result in a hash value less than the current
target.
To compensate for increasing hardware speed in the real world and varying interest
in running nodes over time, the proof-of-work difficulty is determined by a moving
average targeting an average number of blocks per hour. If they are generated too
fast, the difficulty increases. Similarly, when system-wide mining power increases, so
does the difficulty of the computational problems required to mine a new block
(Böhme et al., 2015, p. 218). This difficulty level is adjusted to keep the pace with
which new blocks are generated constant at roughly one per ten minutes (Dwyer,
2014, p. 5).
11
Nodes always consider the longest chain to be the correct one and will keep working
on extending it. If two nodes broadcast different versions of the next block
simultaneously, some nodes may receive one or the other first. In that case, they
work on the first one they received, but save the other branch in case it becomes
longer. The tie will be broken when the next proof-of-work is found and one branch
becomes longer; the nodes that were working on the other branch will then switch to
the longer one.
We already know that in the blockchain, bitcoins are registered to bitcoin addresses.
Creating a bitcoin address is nothing more than picking a random valid private key
and computing the corresponding bitcoin address. This computation can be done in a
split second. But the reverse (computing the private key of a given bitcoin address) is
mathematically unfeasible and so users can make a bitcoin address public without
compromising its corresponding private key. Moreover, the number of valid private
keys is so vast that it is extremely unlikely someone will compute a key-pair that is
already in use and has funds. The vast number of valid private keys makes it
unfeasible that brute force could be used for that. To be able to spend the bitcoins,
the owner must know the corresponding private key and digitally sign the transaction.
The network verifies the signature using the public key (recall the example of the
RSA-system).
If the private key is lost, the bitcoin network will not recognize any other evidence of
ownership: the coins are then unusable, and effectively lost. With no central bank
backing bitcoins, there is no possible way to recoup losses. Besides the bitcoin other
cryptocurrencies have been invented, the so-called altcoins (this naming may sound
familiar to informetricians). Besides cryptocurrencies, there exist other applications of
the blockchain, among which Ethereum, a blockchain-based platform for smart
contracts, is probably the best known.
Media have given a great deal of attention to the so-called anonymity of the bitcoin.
Yet, reality is less simple. On the one hand, bitcoins are like cash in that once Alice
gives bitcoins to Bob, she no longer has them and Bob does, and there is no other
party that knows their identities. While the public keys for all transactions are
recorded in the blockchain, those public keys are, indeed, not tied to anyone’s
12
identity. On the other hand, unlike cash, the fact that a transaction took place
between two public keys, the time when it happened, the amount that was transferred,
and other information is recorded in the blockchain.
If a person’s identity were linked to a public key, one could look through the recorded
transactions in the blockchain and easily see all transactions associated with that key.
For this reason, bitcoin is not anonymous, but pseudonymous at best.
Tying a real-world identity to a bitcoin address is not as difficult as one may imagine.
For one thing, a person’s identity, such as an IP address, is often recorded when the
person makes a bitcoin transaction. Moreover, it is also possible to guess identities
simply by looking at the blockchain. Brito and Castillo (2013) mention that in an
experiment the identities of 40 percent of bitcoin users were discovered (Androulaki
et al., 2012). Moreover, it has been shown several times that studying the bitcoin
transaction graph with the appropriate tools can lay bare the financial activities and
identities of bitcoin users. We conclude that it is very difficult to stay anonymous in
the bitcoin network and pseudonyms tied to transactions recorded in the public
ledger can still be identified years after an exchange is made.
Finally, a few words about the bitcoin in the real world. Can it be considered as
‘money’? It is not because some newspapers refer to the bitcoin as a currency, that it
actually is. “Real” money has three properties: it can be exchanged for something
else (something of value, such as a loaf of bread or a house); it can be used to store
value (the value of the house you sold can be stored in money) and finally, it has a
sufficiently stable value. Although the bitcoin satisfies the first two requirements (at
least to some extent), it is the third requirement where things go wrong. We know that
even official currencies such as the dollar or the euro are not always completely
stable and hence do not reflect a fixed value, yet having a stable value is not at all
the case for the bitcoin or any other cryptocurrency. In a sense the bitcoin behaves
more like precious metals such as gold, digital gold to be more precise, which is also
to a great extent market-dependent (Krugman, 1984).
5. Possible applications
Nowadays it is more and more realized that algorithms that enable the creation of a
blockchain are powerful, disruptive innovations that could transform the delivery of
13
public and private services and enhance productivity through a wide range of
applications (Walport, 2016).
Chapron (2017) points to four specific areas in which the blockchain technology could
be used: ownership, traceability, incentives and policymaking. We provide some
examples, based on Chapron’s article.
Proven ownership of fishing or hunting rights or the right to protect animals, such as
fish, may prevent selling these rights or denying their existence by corrupt
governments. Traceability starts with humans, leading to undeniable birth certificates
(they cannot be lost anymore), but, of course, includes tracing physical goods
throughout their life cycle. Another example is tracking the origin of green electricity
(Fouquet, 2017). Chapron mentions that by using a portable DNA sequencer illegally
traded animal or plant parts can be spotted. The blockchain could ensure that
conservation and development funding is used as intended (a strong incentive to do
so). If insurance money must be paid, e.g. for crop damages, payments can be made
with minimal delay, although officials are still needed to assess damages. Scientific
advice to cities could be organized along a blockchain framework (Acuto, 2018).
Finally, a public, shared and immutable register of assets and transactions can help
to hold politicians accountable for their actions.
Traceability and ownership are essential for business enterprises. Not surprisingly
large companies such as IBM offer partners a form of private blockchain to track their
goods, see https://www.ibm.com/blockchain/. In such blockchains, identities are
known and no cryptocurrencies are involved.
In a research context the blockchain could help solve the reproducibility crisis, reduce
the power of publishing giants and improve peer review (Van Rossum, 2017). In a
‘blockchained’ science, performing and communicating science would look very
different from what happens nowadays. Indeed: blockchains allow for decentralised,
self-regulating data and create a shared infrastructure where all transactions are
saved and stored. As scientific information is essentially a large, dynamic body of
information related to data that is collaboratively created, altered, used and shared, it
lends itself perfectly to the blockchain technology. Working within a blockchain
14
context would mean that whenever researchers create content or interact with it, this
action is stored in a single decentralized platform. In this way, everyone has access
to the same information. Moreover, in a blockchain for research, critical aspects of
scholarly communication such as trust, credit and universal access can be realised
and safeguarded. “Blockchained science” would make larger parts of the research
cycle open to self-correction, and has therefore the potential to address the
reproducibility and credibility crisis (Van Rossum, 2017).
Now we turn to the peer review process. A blockchain framework could not only
improve reproducibility in general, but would also allow reviewers to do their work
more thoroughly as they have more information available to judge originality.
Encryption would allow reviews to be validated but in this way they remain
anonymous and stored permanently. Moreover, post-publication review in various
forms could be integrated easily (Bartling, 2017; Van Rossum, 2017).
Disseminating content
In recent times the possible role of the blockchain in publishing has been investigated
predominantly in non-academic publishing, where the move to online has led to a
shift in revenue allocation from content creators and publishing companies to hosting
companies, social media giants, and advertising intermediates (Van Rossum, 2017).
The original business model followed logically from the structure of the Web, which
consists of one-way pointers (hyperlinks). Hence, there is no immediate mechanism
for allowing small automatic payments for usage. Given this, the only choice for
publishers is to impose unfriendly paywalls with expensive forms of payments or to
open up content and base their business model on advertising. In a blockchain model
for scientific communication this business model could be a thing of the past. Indeed,
several applications have been developed that allow for content distribution coupled
with micropayments that flow directly to the producers of content (Van Rossum,
2017).
community. Any kind of research output could be endorsed including blog posts, data
sets, software etc.
6. A bibliometric analysis
As we are interested to find out how academic authors have reacted to recent
developments related to the blockchain we performed the following query in the Web
of Science (WoS), on January 4, 2018.
We did not try to fully cover all aspects of blockchains or cryptocurrency but are
convinced that we were able to capture the main ones, at least those included in the
WoS. We also note that the query TS=”block chain” on its own gave many false
positives. The final query resulted in 800 publications (3 had 2018 as publication date
and are not shown on Fig.1) with an average number of 1.6 citations and an h-index
of 17. Yearly publications (all types) are shown in Figure 1. Taking into account that
the year 2017 is not complete, this table suggests an exponential increase. As
Nakamoto wrote his article on the bitcoin in 2008, we start the time axis in the year
2008.
Publications
300
250
200
150
100
50
0
2008 2010 2012 2014 2016
Table 1 shows the ten countries with the most publications. For simplicity we used
whole counts (if a publications is written by authors with addresses in three different
countries, then each country receives a score). Moreover, we follow the WoS in
considering England, Scotland, Wales and Northern Ireland as four different regions.
We moreover compared rankings obtained from our blockchain query with the
ranking based on a query for all publications in the research area of Computer
Science. Results are shown in Table 1.
Some countries such as the USA, England and Switzerland are, relatively speaking,
more interested in blockchain technology than in computer science in general, while
the opposite holds for China, Spain and Japan.
Table 2 shows publication types and the number published for each of them. Not
surprisingly, proceedings papers lead the rankings. Also the relative high number of
editorials and news items catch the eye.
18
We also search for the most cited articles, but found that none of the publications on
blockchain technology included in the WoS is highly cited. This is illustrated in Table
4.
Table 4. Most-cited articles related to blockchain technology in the WoS (PY stands
for Publication Year)
Not surprisingly, the most-cited articles about the blockchain deal with the bitcoin. In
particular they discuss the so-called Silk Road, an online black market, launched in
2011, for selling drugs and other illegal goods. As part of the dark web it was
operated in such a way that users were able to browse it anonymously. Selling and
buying were conducted with bitcoins. Yet it was shut down by the FBI in October
2013. It then re-emerged as Silk Road 2.0, but was again shut down by the FBI and
Europol on 6 November 2014. A new version, Silk Road 3.0, went offline in 2017 due
to loss of funds.
Although these articles are generally poorly cited, Nakamoto’s paper (not in the Web
of Science) is much more cited. We found 403 citations (in the WoS), mostly
recorded for Nakamoto S. bitcoin Peer to Peer (and some variations), but some also
for Nakamoto S., consulted, freely available, technical report or working paper.
1) Advantages
2) Disadvantages
However, Chapron (2017) also mentioned that, when it comes to the bitcoin, this
technology is estimated to consume about 10.4 terawatt hours (TWh) a year, which is
21
almost twice the amount used by Google (5.7 TWh). Of course, most of the so-called
‘trusted’ third parties such as banks and governments, also consume large amounts
of electricity and are expected to oppose this new technology as it would make their
privileged role in society largely or completely superfluous. Notwithstanding
organized crime syndicates, whose Silk Road experiment did not turn out very well,
also those used to act in ‘grey zones’ will probably not immediately embrace a
system that makes ‘everything’ traceable. Although, for instance, laundering money
through bitcoin is possible, this may be seen as more risky than using a more
traditional method.
This leads to the question: is blockchain technology the solution for all problems?
The answer is clearly no. By its nature this technology is not efficient: one registration
takes much more time (Chapron mentions that the bitcoin can only manage seven
transactions per second) than when registration is done by one – trusted – party.
Moreover, nowadays transactions become slower and slower.
We already mentioned the loss of a private key in the context of bitcoins. Similar
losses for contracts or ownership (your house for example) or in the context of
inheritances are catastrophic and as far as we know, no good solutions exist for the
moment.
References
Acuto, M. (2018). Global science for city policy. Science, 359(6372), 165-166.
Androulaki, E., Karame, G.O., Roeschlin, M., Scherer, T. & Capkun, S. (2012).
Evaluating User Privacy in Bitcoin. In: IACR Cryptology ePrint Archive 596 (2012),
http://fc13.ifca.ai/proc/1-3.pdf.
Bartling, S. (2017). Blockchain for Science and Knowledge Creation. Available at:
http://www.blockchainforscience.com/2017/02/23/blockchain-for-open-science-the-
living-document/
Böhme, R., Christin, N., Edelman, B, & Moore, T. (2015). Bitcoin: Economics,
Technology, and Governance. Journal of Economic Perspectives, 29(2): 213-38, DOI:
10.1257/jep.29.2.213
Brito, J. & Castillo, A. (2013). Bitcoin. A primer for Policymakers. Arlington (VA):
Mercatus Center.
Cryptography: https://en.wikipedia.org/wiki/Cryptography
Dwyer, G. (2014). The Economics of Bitcoin and Similar Private Digital Currencies.
July 8. dx.doi.org/10.2139/ssrn.2434628
Krugman, P. R. (1984). The international role of the dollar: theory and prospect. In:
(John F. O. Bilson and Richard C. Marston, Eds.) Exchange rate theory and practice
(pp. 261-278). University of Chicago Press.
Pilkington, M. (2016). Bitcoin through the Lenses of Complexity Theory: Some Non-
Orthodox Implications for Economic Theorizing. Handbook of the Geographies of
Money and Finance. Martin, R.; Pollard.J. (Eds.). Edward Elgar: Cheltenham
Rivest, R.L., Shamir, A. & Adleman, L. (1978). A method for obtaining digital
signatures and public-key cryptosystems. Communications of the ACM, 21(2), 120-
126.
Van Rossum, J. (2017). Blockchain for research. Digital Science Report. DOI:
https://doi.org/10.6084/m9.figshare.5607778