Two years ago, Andrew White introduced the idea of blockchain data synchronization for master data.  He had identified technological complexity of distributed IT landscapes as the main obstacle for semantic data synchronization. It is evident that technological gap is not the only one and not the most important impediment here.  Business, non-profits and government organizations are reluctant to empower any intermediary party owning the central hub to establish the synchronization policy.  They simply don’t want to grant an authority and superpower to an external party. We should mention GS1 and IceCat for retail products and ECCMA for industry and military products as examples of semi-successful intermediaries struggling to expand their services and user base. On September 2017 collaboration with IBM and Microsoft was announced to leverage GS1 Standards in their enterprise blockchain applications for supply chain clients. Actualog provides a platform that allows create and use shared product classifier, product templates and product cards.

Craving for product data unification and standardization but willing to keep full meta data and product data control companies could not create and utilize vendor-independent supply chains.

Blockchain is a set of technologies allowing data synchronization without intermediaries, so it could naturally solve the dilemma mentioned above. A blockchain is a ledger of facts, replicated across several computers assembled in a peer-to-peer network. Facts can be anything from monetary transactions to content signature. Blockchains is not only the technology, it is an organizational concept that allows independent parties to work together without any coordination center.

Blockchain master-data synchronization sounds brilliant, shining even more brightly while technologies are maturing. The devil is in the details as usual, particularly in consensus algorithms in our case. Prove of work (the traditional mining) that proved itself so well for the bitcoin experiment is wasteful, slow and expensive for master-data synchronization. Prove of stake could be an option, but what is at stake? How we can estimate and evaluate assets owned by master-data distributed register?

Companies and organizations need full control over their master-data assets, its structure and content. Synchronization should be an option, not an obligation. An owner of the master-data register needs freedom to define and adjust synchronization rules and procedures. Blockchain synchronization for master-data is about enablement, is should start from the extensions in master-data registers allowing data provenance identification in the distributed environment. Anonymity works fine for bitcoin, in the corporate business environment we prefer to know our partners.

It is required to define common dictionary for the meta-data, for example, what is a «product», «variant», «class», «category», «attribute», «property» or «catalog» in synchronized datahubs. Synchronized master-data registers should used unified data structure, they could serve as buffers between corporate master-data pools, sitting behind firewalls.

How a blockchain ledger for product master data might look like, what kind of facts we should register there? Product information management starts from the classifier, we would need product categories and attributes to define products.  We will need a ledger for classifier information and a ledger for products.

Open Technical Dictionary (OTD) and Data Quality ISO standards (ISO 22 745 and ISO 8000) provide an efficient conceptual foundation for distributed hubs, even though OTD standard excludes harmonization and standardization of terminology. OTD terms: classes, measures, units of measurement, values, etc. could be basic block for classifier ledger. Creation and editing of concepts are basic facts in the classifier ledger, in addition we need information whether those concepts were accepted and approved by the particular datahub.  Products ledger keeps information about product creation, modifications, it also keeps reference to classifier ledger concepts used to define a product.

An attentive reader might ask what’s the point in blockchain to synchronize master data if there is no need for consensus, no anonymity? Blockchain-related hype might help to accumulate resources needed to synchronize master-data, like a banner of modernity, success and victory. In Actualog we are going to add some blockchain flavor ASAP 😊


Leave a Reply

Your email address will not be published. Required fields are marked *