The revolutionary potential of blockchain has been much touted in many fields including research and higher education. In this post, Martin Hamilton discusses some of the potential applications of blockchain to academia and raises key questions about how these systems could be implemented and safeguarded from malicious exploitation.
Blockchains are all the rage right now. They’ve joined cloud computing, big data and artificial intelligence in the pantheon of go-to tech psychobabble, regularly featuring in the word salads produced by thought leaders and influencers, as they prep for what they hope will become their TED talk. If you believe the hype, blockchain can cure cancer, solve Brexit and perhaps one day it will also turn out to be a floor wax and a shoe polish.
There are plenty of introductions to blockchain out there, so I won’t waste time and space here repeating them, but there are some important fundamentals to understand before we can think about the academic context of this technology.
Essentially, blockchains are very slow databases that data can be added to but not (typically) updated or deleted – once you put something ‘on’ the blockchain it’s there for good. And like many conventional databases they can be smeared (distributed) across lots of computers, so they aren’t reliant on one computer being online and working properly.
Blockchain enthusiasts are still struggling to figure out how to split a big blockchain database into lots of little bits, so you might still need some high-performance computers if you plan to store lots of data. This is due to the fact that blockchain ‘transactions’ (like ‘updating the Land Registry to record that Martin bought the house’) typically need to be confirmed by all the computers that the blockchain is smeared across, which can take quite a long time.
Most blockchains use quite small data blocks – if we want to store big files like images and PDFs they end up having to be chopped up into lots of little bits, or stored somewhere else, with a pointer to that other place (like a URL) put onto the blockchain.
What does all this mean for potential blockchain applications in research and education?
In my horizon scanning report on the potential of blockchain I considered a few use cases and how blockchain might help. For example:
- A universal user ID or self-sovereign identity that you own and control – choosing who you release what information to, like ORCID on steroids
- An academic portfolio of grants, publications etc that can be updated directly by your funders, editors and collaborators
- A unified mechanism for tracking citations and data-reuse, facilitating open science and reproducibility
- A universal academic transcript for all learners, at all levels of education – perhaps with transferrable credit building up on an earn-as-you-learn basis
Garbage in – garbage out
At face value there is much to get excited about here, but in a piece for UKSG Insights looking at potential blockchain applications for libraries, I wanted to highlight one particular problem that blockchain itself cannot solve – we could call it ‘garbage in, garbage out’ but more accurately it is the issue of false or potentially malicious information stored on the blockchain.
Consider the following example: We recently heard that images of child abuse have been stored on one popular blockchain database, which arguably makes the database itself illegal in many countries. In the UK, for example, mere possession of these images can lead to sentences of five years imprisonment, with sentences for distribution of up to ten years in prison.
Another example of malicious information is data that has been crafted to interfere with software and systems, such as the smart contracts that would underpin use cases, like my hypothetical land registry application. In one recent example, US$79m of Ethereum cryptocurrency was stolen by attackers exploiting a bug in a smart contract used by the Parity software, a popular ‘wallet’ for storing cryptocurrency.
Data could also simply be false, though. There has been a lot of excitement about using blockchain technologies to create a digital provenance scheme building on the legacy of initiatives like Fair Trade. However, if we are to be confident about the provenance of anything from a chocolate bar to research outputs (which can sometimes be falsified), then we need to be confident about who is putting the data onto the blockchain, and the data itself. And even with the very best digital infrastructure, people will put lies on the blockchain.
To make informed use of new technologies like blockchain, it follows that we need to do something which is quite alien to many technologists – we have to ‘threat model’. What might a bad person attempt to do? What would the impact be? How could we guard against it?
It transpires that many of the potential mitigations negate the supposed advantages of blockchain technology – for example, if we simply want to be confident that our academic profile, or learner record, is not going to be abused by pranksters, then perhaps it shouldn’t be an immutable database that anyone at all can add entries to.
For blockchain to work, especially in education, there must be some sort of gatekeeper, if only to separate the wheat from the chaff. For instance, sometimes honours are revoked, or marks adjusted on appeal, so there must be a mechanism to record this – in a way that isn’t prejudicial for the plaintiff. If anyone can peruse the blockchain (through one of the many explorer applications, perhaps) then these adjustments will be there for all to see, because the blockchain itself is immutable.
Never say never though, because it is always possible to have sensitive information, particularly personal information, stored elsewhere, and keep a reference to it (like a URL) on the blockchain. But then what is the blockchain really for? What are the semantics of putting that pointer onto the blockchain? In the case of an academic record should we opt for ‘LSE graduates’, ‘Former students of LSE’, or just ‘People who in some way have had an association with LSE at some point’?
Beginning of the end, or end of the beginning?
It may seem that I’m saying that blockchain is more hype than substance, and indeed there is a considerable amount of dot-com type irrational exuberance. However, the reality is that blockchain technologies are at a similar point to the early World-Wide Web of the 1990s, or to go even further back: potentially as transformational as the telegraph was in the Victorian era.
Blockchains are not going to go away any time soon, but we have yet to see their final form – just as today’s web bears very little resemblance to the medium for scientific discourse originally envisaged by Sir Tim Berners-Lee, all those years ago.
This blog post is based on the author’s article, Of modems and pixie dust – blockchain demystified published in UKSG Insights.
Note: This article gives the views of the authors, and not the position of the LSE Impact Blog, nor of the London School of Economics. Please review our comments policy if you have any concerns on posting a comment below.
About the author
Martin Hamilton leads the future and emerging technologies team for Jisc, operators of the Janet network. His role is all about generating and channelling new ideas and building partnerships to bring them to fruition. Martin is particularly interested in the societal implications of ubiquitous robotics and artificial intelligence, and humanity’s emergence as a true interplanetary species.