Hi: Is has been a source of frustration for years. I asked techies inside WorkFront how they learn the data model with no meaningful data dictionary and was told rather consistently that it is tribal knowledge – word of mouth. I struggle with the idea that they have no internal documentation that describes the business semantics. That is developmental chaos. Sure, their data model is unique and likely held as company confidential. I have worked with many software companies that felt giving out the business semantics of their data model somehow reduces their competitive advantage or opens them up to complaints of "your data model is not third normal form. Can you adjust it to be Third Normal Form, please?" It's not worth the complaining from those who want to out-tech your data team. It's an ingredient of your secret sauce. Having said that, the value in keeping information in a database is ONLY realized when it is taken out, in the form of integrations, reports, and screens. Getting the data out creates the value. Knowing what you're taking out definitely increases the probability that each integration, screen, and report is properly understood. Not having a data dictionary with business semantics increases the probability of misinterpretation, or worse, not getting all the value from your data that you need. So I've thought of paying for a data dump in MS Access form from AtAppStore, then running it through software to produce an ERD, then create a Wikipedia article(s) or something we can all edit with the business semantics I can figure out. If the company won't do it, maybe we should. Time. I just need the time. ��
