Maximizing Code Quality and User Skills with Git and Blockchain: Part 1

This article was written with the help of Jasper.ai

Currently, Git uses a merkle tree data structure to perform settlements and reconciliations for code changes in a commit, similar to the way that blockchain transactions are processed. This data structure allows Git to track the settlement and reconciliation of the code change itself, as well as the user information included in the commit hash. However, by incorporating blockchain technology, it is possible to track the user, commit, and change over time in order to create a rating for code quality.

To do this, whenever a commit that passes the validation function in an associated validator feature is made, the value of the unique user token is updated along with a multi-dimensional objective value score. If the code is later changed to fix a problem with the function, a similar value score is updated on the record, adjusting the user’s rating in discrete transactional steps. The result is that the chain always maintains the user’s current aggregate value score across multiple dimensions of value. This feature enables automatic leveling and promotion within organizations using the system, effectively creating a real-time skill tracking and rating function.

However, the cost of performing this sort of tracking for an indeterminate number of users would be exponentially higher for a completely off-chain system, as dedicated resources would have to be maintained for the data and business logic layers. This feature is just one of a set of features that together represent a granular skill level assessment and real-time performance record for an entire organization.

The last mile problem in this scenario exists in multiple components. One is the actual Git change, which is typically handled through a process called a pull request, which requires peer review and corrections to the code before a commit will be merged to a branch representing versioned code. However, there is potential for abuse and gaming of the system, so this problem will require more thought and careful observation in order to completely address major incentive issues. There is potential to encapsulate many parts of the standard agile software development process in blockchain or in smart contracts, but for now, the majority of processes are off-chain, with high-value novel processes such as the rating system added first.

Another dimension of the last mile problem is in the consumption of the user rating system by third parties. This service amounts to a domain-specific social credit system, and most of the components of the display and incorporation of user rating values will be handled by off-chain software systems. It is likely that an event-driven service for each dimension (code quality, complexity, breadth, logic) and a pub/sub model of streaming events will be used. Potential customers for this service could include recruiters, other companies, the users themselves, and academic institutions. Team ratings can also be used as validation for management systems and development methodologies.

There is a wide range of potential for such a system. The rating system is both value-chain enhancing and value-chain replacing. This service will be developed as open-source software, so there is no intention of competing with any other business solution. The product is the team enabled by the system and the products they create. There is potential for both permissioned and permissionless versions of the system, but the permissioned version is likely to hold the most utility for private enterprise, while the permissionless version may be more useful for the community.