Design principles behind developing a standard

Open Data Product Specification is a vendor-neutral, open-source machine-readable data product metadata model that defines objects and attributes related to data products. The first production level version 1.0 was published on 5th Feb 2022. The latest version is 2.1 which was released on 22nd Aug 2023.?

The idea of defining data product standard was born in 2019 as part of developing a Data Product Toolkit for data productizement purposes. Version 1.0? of the Data Product Toolkit was published in Miroverse for all to use. After that Data Product Toolkit evolved into a complete package to do data productizement with 6 canvases and the latest release (2.1) was done in June 2021. After that, I have dedicated more time to the Open Data Product Specification.?

Initially a side project, ODPS evolved into a robust data product metadata model, integrating business metadata, pricing plans, SLAs, DataOps, Licensing, and security. The ODPS offers a metadata model that is shared among tooling and automation. I would say ODPS has the potential to become the “Swagger” of data economy. ODPS is now listed as a data product model in the EU Data Spaces tool stack.?

That alone is a tribute to ODPS developed originally by a handful of people from Finland since the alternative standards in the list are coming from global giants like W3C (DCAT) and TM Forum.????

Adoption of ODPS is increasing due to practicality?

During the year 2023, more and more organizations have approached me with interest in building Proof of Concepts in which ODPS is used. Even before this ODPS has been on the radar of multiple organizations from Fortune 500 companies to governmental institutions. Here is a very fresh just days old testimonial providing some evidence that our design principles are working. The quote is taken from PoC discussions with the customer organization:?

But what are the mentioned design principles helping us in ODPS development and in driving faster adoption and challenging the giants in standards?

Why do we need design principles?

Early in the development 9 design principles were defined. Design principles are crucial in developing a standard because they provide a framework for creating a standard that is clear, consistent, user-friendly, interoperable, secure, and adaptable. They help ensure that the standard serves its intended purpose effectively and stands the test of time.?

Nine design principles of Open Data Product Specification

Here is a screenshot from the early days regarding the design principles (still valid and in use)

Some of the principles are no-brainers but need to be stated. Some principles are easy to follow while for example “make it hard to misuse” is a balancing act. Making something hard to misuse can require strict rules and details, but at the same time ODPS is not aiming for perfection, it is driven by “must be practical”.?

Reusing existing solutions has included also looking at what works in DCAT and other standards. One of the key principles is the Developer eXperience. Onboarding the standard, “getting it” easily without a week-long study, and a short path to apply is very important for the ODPS.?

“Make it long-lasting” manifests in practice by using clear versioning (SemVer) and having a Governance model. We have spent hundreds of hours studying the landscape of data economy, data-driven value delivery, and existing solutions. Internally we have had hundreds of hours of debates and an uncountable number of tests to find the best possible approach to solve problems at specification level. This is how “Do the hard work to make it simple” manifests in our lives.?

The value of having design principles has been priceless. It has at the same time made our lives easier, but also occasionally harder. What seems to be the result based on early evidence is that the design principles work and guide us towards easy-to-understand and value-creating standard. And that is all that counts. The Design principles are also part of the ODPS Governance model.?

The bottom-up approach to serving the needs of the practitioners?

Take a look at the Open Data Product Specification model and provide feedback (Github issues, links provided in specification). Our principle is that development ideas must come from the practitioners, their needs drive the development of ODPS, not bureaucrats or people in ivory towers. The maintainers of ODPS are here to serve the needs of the business and practitioners.??