Re: Locally Managed Tablespaces ... again!!!

"David Sharples" <david.sharples3_at_ntlworld.com> wrote in message news:bGIT9.274$CI5.2092_at_newsfep4-gui.server.ntli.net...
>
> If Oracle's own instructors (not all probably) are still teaching that it
is
> better for performance, then the myth wont go away anytime quickly
>

To be fair to the instructors, it's hard to teach anything otherwise than as written in the course documentation: and in this case, the course documentation itself is still peddling the 'separate indexes for performance' myth.

Bear in mind that instructing is an art and a skill in itself. You can't reasonably demand that every (or even 'nearly all') instructor be a technical wizard when it comes to Oracle's workings. If they can teach well, that is all you can really ask for. But because they aren't technical gurus, many of them will teach the material they are given by the Corporation without much pause for deep thought. Just occasionally, you'll encounter an instructor who not only likes instructing but actually wants to know some detail about *what* he's instructing, and therefore goes to great lengths to pick apart the myth from the truth, the fact from the error. They are, however, few and far between (and not just at Oracle).

It's also the case that with Oracle's course scheduling techniques (pile 'em in, bums on seats), and the number of instructors available, the time for that research has to come out of the instructor's own weekends/evenings. Provision for genuine research in what is laughably called 'Oracle *University*' is, er... shall we say 'somewhat lacking'. You might think, in plainer terms, that it's actually pathetic. But I couldn't possibly comment.

Regards
HJR Received on Fri Jan 10 2003 - 21:06:22 CST