Re: The wisdom of the object mentors (Was: Searching OO Associations with RDBMS Persistence Models)

Neo wrote:
> :) While RM does have a broad scope where it excels, is it
> God's ultimate gift to mankind? Could someone convince me
> it is, by showing how RM is better at modelling/querying
> the simple example at:
>
> http://www.dbfordummies.com/example/ex005.asp

I LMAO when upon reading this in that site's FAQ:

  Q : Is DBD based on the Network, Hierarchal or Relational

      Data Models? Or some combination thereof?   A : DBD isn't based on any common data model. DBD isn't

      the combination of various data models.

  Q : What data model is DBD based?
  A : DBD's data model is named TM. It's details are not

      being disclosed currently.

Why would someone want to use a data management product with an undisclosed data model?

Anyhow, because I'm a n00b and want to practice I will play with that example. (Note I'm not trying to /convince you/ of anything. Your tenor suggests that is not possible.) Here is my relational database:

colors :
name part color
---- ---- -----
john skin black
john hair red
john hair blue
mary skin white
mary hair red

textures :
name part texture
---- ---- -------
mary skin smooth
mary hair silky
mary hair smooth

Now for me this is easier to reason about than the hierarchy presented on that page (recreated at the end of this post). Please understand, I am not a n00b when it comes to hierarchy / network models. I've used them for years. Even so, I still find them complicated and more time consuming to analyze than relations which immediately seem more simple to me.

For example, formulating this example in a relational way it was immediately obvious that we have texture information solely for mary. It also exposed logical questions about the model itself. Must hair and skin always have a color or texture? Should skin and hair color values belong to the same domain?

As for the example queries, the lispy syntax is different and I would love to code up the equivalent example queries for this relational model in either Tutorial D or a lispy version of it. However, my query language skills are still meager and thus anything I code would probably suck. Maybe someone more skilled would be interested in doing this.

Finally, I would like to point out there may be flaws with the (undisclosed) DBD data model similar to the ones with other MV systems exposed by Bob Badour in his "red blue car" analysis from some time ago. He and others thoroughly demolished the MV advocates in that long thread. Perhaps someday a more knowledgeable regular here will have the time to post a "red blue car" /redux/ thread with cleaned up versions of the MV and relational example databases along with the proposed MV "solutions" and the problems with them. I for one would appreciate that because eventually I got lost in that thread and didn't understand the /final/ MV "solution". I mean to say I could not understand how it actually solved the problem.

(original "red blue car" example begins in this thread I /think/ http://groups.google.com/group/comp.databases.pick/browse_frm/thread/8f0304d1a1ff9939 )

• Keith --

[original hierarchy from link]

person
  john
    skin

      black
        color
          black
    hair
      red blue
        color
          red
          blue

      white smooth
        color
          white
        texture
          smooth
    hair
      red silky smooth
        color
          red
        texture
          silky
          smooth