Re: The MySQL/PHP pair

"Kenneth Downs" <firstinit.lastname_at_lastnameplusfam.net> wrote in message news:2df962-hee.ln1_at_pluto.downsfam.net...
> PMFJI. A friend of mine refers to DBAs and programmers as "natural
enemies
> in the wild." Wish I could disagree with him.
>
> Is there some sound basis upon which we can relieve the natural division
> between programmer and dba? My own answer divides the client into
"layout"
> and "process". For layout you need someone with a keen eye to mark up
some
> HTML (oops, I mean XHTML (oops, I mean XML)), but the drum I beat around
> here says the procedural code in the client is all dependent upon the
> definition of the database, so may as well generate it.
>
> This leaves the programmer prototyping new cases that the code generator
> cannot seem to handle, but not doing much else.
>

That's all programmers were really doing a half a century ago. Of course, the concept of a "code generator" back in 1954 would have been "a FORTRAN compiler".

> Are there are theories on this? Are we stuck with it forever? If not,
what
> is there to be done?

I suggest getting and reading "Object Oriented Analysis" by Peter Coad and Edward Yourden. It's a little out of date, and it covers a lot of things you probably already know. But it really helped me to see the unity behind the fractured state of system analysis.

Maybe there are some more current works that people can suggest in place of this one. But it has to be one that asserts that their soultion is NOT "the silver bullet". Received on Wed Nov 10 2004 - 15:26:49 CET