“Not since the Web itself has a technology promised broader and more fundamental revolution than blockchain technology.” —Hyperledger Project
And perhaps not since the Web itself has any technology been hyped so much as blockchain.
Enthusiasm sometimes makes it difficult to find the “edges,” to even know what role a blockchain is performing.
At its core, a “blockchain” is a type of database. Like any other database, it stores information. And like others, it has strengths and weaknesses. The strength of blockchain is that it secures its data content so it can’t be modified without evidence. That makes it ideal for authentication.
A database—like a Swiss-army knife—is a general-purpose tool that can support a range of applications. But each has its place.
As often happens in such cases—with a gale-force excitement at its back—it appears that blockchain can do anything.
This can be misleading, because when we look at the latest demonstration of blockchain’s power, we may have the impression that we’re seeing something that can’t be done in any other way. But in fact, as some nay-sayers are quick to point out, 99 percent of what we’re seeing could be implemented with any off-the-shelf database and conventional programming language.
In the supply-chain industry, for example—already heavily committed to blockchain—viewers typically are presented a demonstration of seafood traceability, or coffee trading, or mango-tracking, or a smart contract for corn or soy that automatically imposes penalties on the seller, etc.—the list is now quite long.
Detractors will explain how these applications could be implemented entirely without blockchain. But that misses the point.
It’s about strengths and weaknesses.
Blockchain’s strength is trust and security—it lets trading partners share authenticated data knowing that it has not been tampered with.
Among its weaknesses, blockchain is slow. The data—usually transactions—are knitted into a time-sequenced chain with cryptographic seals at every link. This makes it secure, but retrieval from the chain itself is glacially slow.
For this reason, every blockchain has a helper-database to handle lookup and retrieval.
Today, the partner is usually a “key/value” database, such as Leveldb, Couchdb, Mongodb, or some other member of the key/value family; sometimes it’s a conventional relational database (SQL). These databases can quickly look up transactions by attributes—parties, dates, price-ranges, etc. The results are pointers into the blockchain itself. Given a set of attributes, the helper database can quickly identify a set of relevant blocks, fish them out, unpack them, and serve up the authenticated data to the client.
From our experience at GeoCertify, this common blockchain data architecture is not the best solution for a full-scale solution for a global supply chain like coffee.
The reason is the “friends-of-friends” problem, or—translated to coffee—the “suppliers-of-suppliers” problem.
Tracking a blended coffee—that is to say, almost any coffee—means traversing a network of relationships that can “fan out” into hundreds or thousands of filaments, whether looking in the backward or the forward direction—and please note that traceability in both directions is required.
Key-value and SQL databases are not the optimal choice for handling this type of retrieval. This is a problem of searching by relationships, not by attributes. As the data gets deeper, the search time begins to explode exponentially.
Therefore, we understand that today’s blockchain architecture cannot yet provide the backbone for a full-scale traceability or distributed supply-chain management system for blended products like coffee.
Having said this, it’s important to emphasize that GeoCertify is absolutely committed to blockchain.
Perhaps we’re biased, but we have tackled some of the core data management issues in the suppliers-of-suppliers problem. We can scale to billions of entities without exploding the lookup times.
Our strategy for blockchain is to use it for its core strength—trust through immutable authentication.
We start from a fully-operational integrated supply chain management system that has already traced 45 million pounds of coffee from 21 origins; it has already addressed end-to-end traceability—backward and forward; it has already handled fan-out to thousands of small-holding farms.
Thanks for this go to another advance in database technology—the graph database. Graph databases have emerged and grown right alongside blockchain, only without the cheerleaders and marching bands.
GeoCertify is adding blockchain authentication to a fully operational graph-based integrated supply chain. Putting these together, we can see a clear path to the next level of scalability and distribution.