- Shopping Bag ( 0 items )
-
All (3) from $57.70
-
New (3) from $57.7
The authors do a good job of hiding much of the complexity of the subject. Though after I wrote that, it occurred to me that part of the credit also and surely obviously goes to the developers who coded the package. The Master Data Services reflects the reality that raw SQL programming is too low level for anyone but dedicated SQL programmers.
What MDS allows is a more top down approach, as covered in chapter 4, for example. Here, you can construct an entire object model hierarchy, where these derive naturally from the real world business context you have to deal with. From doing a top down set of models, the book shows how the user interface lets you rapidly commit this to a functioning database. The given example is worth contemplating for how it intuitively focuses on the business aspects.
Versioning is another big idea in the book. But it differs slightly from the word of the same name as used by programmers. For the latter, versioning is often used to freeze code at different times, and to debug and validate specific versions. To some extent this is also used here. But chapter 5 advises that version management can help you test different business scenarios, by maintaining independent copies of master data. Be careful about this. In more general programming environments, this can lead to awful problems later on, if you want to consolidate [ie. merge] 2 versions, whose data sets and rulesets built atop those data have deliberately diverged. Here's the thing. Most of chapter 5 deals with the details of how to make the versions. Low level instructions that you will certainly need. The problem can arise at the higher level of management of the project. Be careful about how much you want to do with split off versions. How is a lesser issue.