Model Doc Processor

Automation for consistency of implementation

At EC Wise, our method for creating physical databases for our clients follows a programming metaphor:

• The "source code" consists of business model entities defined in one or more XML documents.
• The "compiler" is represented by a custom .NET application that parses the XML model document.
• The "executable" build product includes HTML documentation, statistical reports, and SQL scripts targeting several RDBMS platforms.

This gives EC Wise the following advantages:

Encourage language-agnostic thinking

Although the ModelDoc Processor can accept a Word document as input, in practice the convention of specifying the data model in an XML dialect allows the offshore team to focus on the logical requirements of busines relationships, reducing the overhead of reading and writing English.

Early validation

Having multiple authors increases the probability of errors in the data model specification. The model XML can be validated against a specific XSD, even before the changes are saved in the editor - making it difficult for errors to be introduced in the first place.

Accurate documentation

Given a known revision of the data model, the generated documentation is guaranteed to accurately represent the corresponding physical database schema. There is no chance of errors arising through human error (typing mistakes, or comprehension errors) during the documentation process.