Blockchain since its very inception has been widely sought by numerous technologists, researchers or entrepreneurs for finding killer-apps that can bring blockchain to the mainstream. One response is to utilize blockchain to solve the traceability problem and many thus start to advocate that with blockchain one can build the best traceability solution. It seems to many that as long as one system is built with blockchain, it automatically becomes invincible and omnipotent in tackling the underlying problems. Truth or lie? This is an important question to answer!
It is true that blockchain can bring some intrinsic values like openness, transparency, security, data integrity, indisputability, transaction atomicity and traceability. Applying blockchain to solve traceability in industries like agriculture could be a good option as one enabling technology. However, one has to understand that blockchain alone cannot provide a solid and secure solution for solving traceability issues.
When it comes to traceability, it means establishing a chain of trustworthy informational data points from the origin to the final destination. When the process moves from one point to another along the chain, one must make sure there’s no falsification, alteration or degradation of the information to be injected into the chain. It has to be verified and approved by (preferably digitally signed by) a predesignated individual entity who is responsible for committing such information or truth into the chain. Since there could be many such middle points existing with the chain. Any failure or attack to one or several such middle points would immediately result in a failure of ensuring the traceability.
Let’s say, there are 100 such data points within a chain, and each data point has 0.01% chance of of being attacked or malfunctioning resulting in data entry failure. So the total chance of failure in ensuring traceability is:
1 — (1 — 0.01%)¹⁰⁰ ~= 1%
That means after 100 times of repeating the end-to-end traceability process, the chance of failure is 100%! That makes the whole system rather vulnerable and impractical!
One could argue that the failure rate on each middle point can be greatly reduced to near zero. However, this is rather a lie in my view. There could be unintentional or intentional failures or errors along the chain process. What’s most importantly, there are heterogeneous types of systems that provide the data points to the traceability chain and some systems may need to interact one another before the information is generated and stored within the chain node.
Let’s look at what could possibly make up the complete traceability end-to-end system.
One Example of an End-to-end Traceability System
There are following possible types of uni- or by-directional information transmitting or propagation along the full traceability chain:
- Human to human
- Physical to digital
- Human to machine (a centralized information system)
- Machine to machine (between two centralized information systems)
- A centralized system to a decentralized system (blockchain)
- One blockchain to another blockchain (Cross-chain)
For a centralized system to input data onto a blockchain, one needs to deploy so-called oracle software that connects to the blockchain network and is able to transmit system or human generated transactions onto the blockchain. The transactions passed by the oracle system carries the traceable information.
As you can see, one typical traceability solution could be composed of many heterogeneous components that may not share the same standard or implement a common spec. which makes the communication even more difficult and prone to failures, esp. when human beings are involved in the whole process.
Summary
It sounds all pessimistic, right? Yeah, to build a trustworthy and reliable traceability solution can be really hard and costly! Blockchain is not and will never be the silver bullet to this problem. But like many practitioners have said, blockchain improves the whole system in terms of data sharing efficiency and transparence and integrity protection on-chain. It is up to the solution designer of any traceability system to take care of every node within the traceability “chain” to ensure even the weakest link within the chain has been taken good care of and won’t pose any potential risk or the effort of compromising the system would cost more than brining the failure to the whole system.
So one might be wondering, what would be the ideal future for solving the traceability problem? My simple answer is that only until every component within the system has been nicely digitized and decentralized with cryptography-level security would one be able to see such a truly reliable and trustworthy traceability system! That is however technically possible! :-)
