The Database Is The Software Base

When I saw this phrase in TheyWriteTheRightStuff it rang a bell, so I jumped in and Wikified it. I just want to get some thoughts (prejudices) down here and hope someone (I?) will organize/refactor/whatever. (Sorry for the Java bias, and my poor Wiki text formatting skills.)

The idea is to use the database (RDBMS - or OODBMS or OR...) to

• track the project - not to replace an app like MS Project, but to augment it.

One [database contains] the history of the code itself -- with every line annotated, showing every time it was changed, why it was changed, when it was changed, what the purpose of the change was, what specifications documents detail the change. Everything that happens to the program is recorded in its master history. The genealogy of every line of code -- the reason it is the way it is -- is instantly available to everyone.

-- http://www.fastcompany.com/online/06/writestuff.html

• hold metadata, configuration, etc.
  • Just eliminating a zillion "property files" and scripts is a reason in itself.

  • A relational or OO model is better than a flat file for representing metadata.

  • A relational or OO model is better than a hierarchy (XML) for ditto.

XML is overused for this kind of thing (cf. WagTheDogAntiPattern).

• Cuts down on LanguageTribbling.

• Facilitates replacing manual editing of text files (properties, scripts) with controlled maintenance.

Any property file mechanism can be refactored to operate this way.

On the JavaLobbyCommunityPortal? project, we are in the middle of this

process: first we're moving all direct access to the Properties object

into one class. Once it's encapsulated, the Properties file can be

replaced by DB tables, and the "editing" will be done through the admin console.

-- TomRossen

---

So it would appear that parts of the world are finally realizing that the "file" is an artifact of the way computing was done forty years ago.

My question, now, is whether or not the ObjectOriented community can figure out how to clue in the relational DB world about using persistent objects instead of rows in tables. In every example listed here, a "good" distributed persistent object implementation is compellingly better than either its flat-file or relational counterpart.

Of course, the fact that OODB vendors all died ten years ago doesn't make our job any easier. Perhaps if they had spent more time making indexing and metastructure work -- and less time lying about their vaporware -- we might all be better off. But I digress.

Hey - you caught me in the middle of trying to clean up my crummy formatting! Anyway, I think there are some pages in this Wiki that are a little less pessimistic about OODB than you are - not that I remember where they are at the moment. I wonder if you would think Oracle 8i/9i allowing Java stored procedures is a step in the direction you advocate. - TomRossen