Talk:Chapter 10
From Structured Analysis Wiki
How should data dictionaries be organized, alphabetically or by logical data store. It seems that both organizations could be useful? Does any one have any thoughts on this. Also perhaps we could add a small example data dictionary here to help explain the concepts.
Comment on how data dictionaries should be organized
In almost all cases today, a systems analyst will have an integrated development environment (IDE) or Computer-Aided Software Engineering (CASE) tool that will support an automated form of data dictionary. There's some discussion of this in Appendix A.
As a practical matter, this means that the systems analyst should be able to use the automated tool to search for data dictionary entries by name, keyword, author, logical data store, or other attributes; and if it's necessary to print a hard-copy report, it should be a simple matter to do so alphabetically or in any other desired order.
Remember: we now live in the age of Google; people expect to be able to search for almost anything, and find it easily!
Logical Data Dictionary and CASE Tools
The way I have organized our data elements is both in a data dictionary and also in logical data stores. The data dictionary would have the details as described in this chapter and the logical data store just collect the data elements logically but does not contain the associated informaiton about each element.
Also I read Appendix A as we currently do not use an CASE or IDE with a data dictionary tool in it, and I was curios about how to use the databvase (Sybase, etc) to create a data dicionary. Wouldn't the database data dictionary be the physical one. I am also assuming that there would be both a logical and data dictionary creating during system development. The logical DD would be created during requirements analaysis and the physical DD is created during technical design.
Too bad that the comments do not state the person who entered and the time when they were entered.
Just add 4 tilda (~) at the end & it "signs" & timestamps your contribution. DEddy 16:01, 23 February 2007 (PST)
Data Dictionary vs Metadata Repository
Ed - how do you want to treat the evolution/mutation of terminology over time? Specifically: what was called a "data dictionary" in the late 1970s when you originally wrote the book, has now been inflated to "metadata repository" or even "enterprise metadata repository" (ignoring the possible common spelling of metadata, meta-data and meta data) Personally I think the original term "data dictionary" should be kept for consistency, but a note early in this chapter noting/footnoting? that "data dictionary" has fallen out of fashion and is commonly (plus much more broadly) applied to repository 30 years later. DEddy 20:26, 22 February 2007 (PST)