Any opinions on this one?

"LudekSA" <ludek (AT) webmail (DOT) co.za_nospam_> wrote

None that I won't receive criticism for voicing.

"Bob Badour" <bbadour (AT) golden (DOT) net> wrote

Criticise away, Bob.

But if you call Luke a drooling fool, or a moron, or an idiot for even thinking
that Cache might have something going for it, well then . . . .

I've been using Caché from day one (and it's predecessors for many
years before). So the answer would have to be "Yes, a positive one", I
guess. If you have any specific questions I'll try to answer as best
as I can.

Wolf

"LudekSA" <ludek (AT) webmail (DOT) co.za_nospam_> wrote

"LudekSA" <ludek (AT) webmail (DOT) co.za_nospam_> wrote

From time to time some people try to make-up old, obsolete and
discredited approaches and to sell that to the gullibles as a new and
revolutionary stuff.

See:

http://www.dbdebunk.com/page/page/622933.htm


Regards
Alfredo

And yes, I do know the difference between its and it's, in case anyone
wonders. I was in haste.

W.
wolf.koelling (AT) slaughterandmay (DOT) com (Wolf Koelling) wrote in message news:<eef28651.0401230205.74bc1ed1 (AT) posting (DOT) google.com>...

It would be stupid for anyone to criticize a small typo when the post has
much more serious flaws.

If the experience was positive, how does it compare to using a relational
dbms and what is your experience with the relational model?

What, specifically, made the experience a positive experience?

"Wolf Koelling" <wolf.koelling (AT) slaughterandmay (DOT) com> wrote

"Alfredo Novoa" <alfredo (AT) ncs (DOT) es> wrote

Hmmmm! Maybe this is why some people still choose to use
"old, obsolete and discredited approaches". A quote directly
from Chris Date's paper "What do you main, post-relational"

"Second, regarding the argument that "relational technology [is] ...
slow": Well, I might agree for the sake of the argument that it's true
that some, or even all, existing SQL products are slow"

Thank you Chris Date for verifying something that those of use in
the field already knew!

--Corey

"Bob Badour" <bbadour (AT) golden (DOT) net> wrote

The "relational model" is irrelevant if the implementation of that model
fails to meet the performance needs described in the requirements.

--Corey

"Corey Brown" <corey (AT) spectrumsoftware (DOT) net> wrote


Interesting that you picked up on precisely the point I was going to comment
on, but from a different perspective. If every implementation of the
relational model (to date) is slow, how can one defend the point that the
relational model itself is not inherently slow? True, a model is just a
model, but if the model cannot be implemented, of what value is it?

It is this religious tenacity to the moral superiority of the relational
model that makes the trolls so annoying. I spent much too much time on a
thread attempting to point out that one camp consistently debates from the
position of superior theory while the other argues from the position of
superior implementation.

From another perspective, the claim that a relational logical model could be
physically implemented precisely (if not more) efficiently as an object
database ignores half of one of the major tenets of programming:

There is no logical problem that cannot be resolved by adding a layer of
indirection.
There is no performance problem that cannot be resolved by removing a layer
of indirection.

You can't have it both ways.
--
<disclaimer>
Opinions posted are those of the author.
My company doesn't pay me enough to speak for them.
</disclaimer>
--
Jim Melton
Software Architect, Fusion Programs
Lockheed Martin IS&S
(303) 971-3846