[XML-DEV Mailing List Archive Home]
[By Thread]
[By Date]
RE: [xml-dev] To Normalize or Not to Normalize
- To: 'Robert Koberg' <rob@...>
- Subject: RE: [xml-dev] To Normalize or Not to Normalize
- From: "Bullard, Claude L (Len)" <len.bullard@...>
- Date: Tue, 13 Dec 2005 15:53:08 -0600
- Cc: 'XML Developers List' <xml-dev@...>
Insight.
It is a common experience. With the current chasm
crossings to web-based tech, possibly very common.
As Frank says, research first, of course. Now what
questions are important?
My first clue is that denormalization is either intentional
(performance, for example) or laziness (multiple developers,
not enough discipline or communication), historical (obtained
the code from another company and built on it), or accidental
(didn't know two fields were the same thing).
Names seem easier but can be harder.
Of course, finding out where the customer intends to spend
money or resources now, next and after next is important.
len
From: Robert Koberg [mailto:rob@...]
Can there be any other answer than "It Depends" ?
What's the budget? What's the time frame? How important is the app?
Given the all too common and absurd answers to the above questions, my
strategy would be to bail out, start my own company and do what I want :)
seriously, what are you expecting here?
|
