Align, Refine and Define Data with Steve Hoberman
It was great to spend time with Steve Hoberman again. It has been five years since his last webinar, and it was lovely to listen to how the Data Modelling process of distilling business requirements has improved, changed and morphed due to the significant change that NoSQL has introduced into the field of data management.
His new requirements validation process talks about more than just how Data Modelling. It also talks about how we need to adjust to new ways of building applications that collect the data we need. And dare I say how we need to adapt to changes in life.
First, we need to align our understanding with what the business requires. We need to get a common understanding of the Business Terminology to do the alignment. Steve has adjusted the terminology he uses to align with NoSQL. I still don't see a precise definition of the business terminology.
I am increasingly convinced that using the Conceptual Model and a formal Business Definition Writing process (@Terry Smith and Mark Atkins , please read their book - Data Governance needs Risk Management, AKA - fix your business terminology) gives us the best result, as it includes all the business rules that will ensure the business expectations of the data.
The NoSQL model focuses on the Terminology and Process Queries at this stage. It was interesting to hear an adage that we use with Design Patterns: Form follows Function. In this case, read it as Form follows Queries.
You may be wondering how different this is from procedural programming. The difference begins at the next step (Refine), introducing Hierarchies.
The Refine step creates the required levels of document hierarchy according to the query. Steve refers to this as a Query-focused Data Model.
The Hierarchy is an essential change for NoSQL data modelling. Document and graph databases handle the hierarchy structure very well. A collection of nodes in a graph database can be considered a document.
The Define Step enhances the hierarchies with what Steve calls the "secret sauce", and now we see a model that closely represents a namespace definition of the document.
Recommended by LinkedIn
I want to thank Steve for sharing and teaching us how to realign our thinking.
From a Data Governance perspective, the Business Terms Model and the Logical Model are critical for understanding the essential business terminology, rules and sets that should be adhered to across all data systems: relational (master data), dimensional and NoSQL.
Webinar Details
Title: The Align > Refine > Design Approach to Data Modeling
Webinar Date: 5th Feb 2024
Presenter: Steve Hoberman, DMC
Meetup Group: INs and OUTs of Data Modeling
Modelware Systems Blog Post:
Tags: #datamodelling, #nosql, #alignrefinedesign, #datasherpa
Data Solutions Technical Lead at 4Sight
11moPlease share the recording
Senior Consultant Team Lead at Mibcon DIMA | Customizing and extending SAP PowerDesigner | modeling | DWH | Sybase
11moCan you please share the recording with me?
Hi Howard, this looks like it was an excellent presentation - can i get a link to see it?
Freelance Data Management Consultant, Metadata Specialist, Data Modelleur, Data Analist
11moYes please
Data Science Manager | Consultation | Senior Data Scientist | Machine Learning | Artificial Intelligence | GCP, Looker, Tableau, Snowflake, PowerBI.
11mointerested