Re: the relational model of data objects *and* program objects
Date: Fri, 15 Apr 2005 16:33:00 +0200
Message-ID: <425fd0a5$1_at_news.fhg.de>
> No matter what you call it, the customer is paying for a system that keeps
> accurate records. That means the database is the first and final concern.
> In a database app, there is no theory of code that trumps the needs of the
> database. Rather, as "form follows function", the code theory needs to
> come after the database theory.
The customers pay for two things:
1. keeping records, and
2. manipulating records
So these two sides cannot be separated. We need not only to store our
files but also move them between folders and update the records. So code
and data exist together and we hardly can concluse that database is the
final concern. I would say that they have equal rights.
But finally customers pay for a system that works but it is not an easy
task nowdays. One of the problems is that we are not able to solve tasks
which were primitive yesterday. We cannot say where our data is stored
and where is the database located. One and the same data is everywhere:
it is in cache (many levels), it is on web pages, it on disk and on
tape, it is in an application sever, it is in memory, it is in
processor, it is in the network, it is on paper etc. The code takes also
very different forms: web page scripts, intermediate languages (Java
byte code, .Net), native code, database languages (PL/SQL etc.), EJBs
etc. Note that it is one and the same system with one and the same data
and one and the same code. To make such a system work is not a simple
task and it is where the existing technologies do not help. So they
exists certain necessaty for something new.
alex