Re: Informix vs. Sybase vs. Oracle vs. (gasp) MS SQL Server

From: Anthony Mandic <no_sp.am_at_agd.nsw.gov.au>
Date: 1997/12/03
Message-ID: <3484D255.410E_at_agd.nsw.gov.au>#1/1


Tim Schaefer wrote:

> I know one rather goofy organization that is using Informix for their
> hotel reservations, and they definitely depend on row-level locking.
> I've watched the user load increase at a rate that would make me cringe
> if it was any other data base product, and the equipment was top of the
> line. Overhead? Sure. I like their approach, put the best damn db
> engine on the fastest hardware you like, and go from there. It would
> be ludicrous to think this operation would get by on PLL.

	Hmmmm... looks like a poor physical/logical design coupled
	with a poorly written front end. Maybe we can send Jeff Wong
	in to straighten them out ;-)

> I would hate to have to work around PLL, rather I'd recommend a different
> product, that not only has RLL, but many other features that Sybase does
> not. PLL is an achilles heel in OLTP, otherwise Sybase would be the mecca
> for a lot of companies. This is not what is actually happening. The only
> thing saving Sybase is Powerbuilder. Sybase is not alone in their problems,
> witness some of the other elitest companies, who, talking tech, had what
> they thought was the best. They overpriced the products, and subsequently
> lost a lot of business. I think Unify had to learn this lesson too. Now
> perhaps it's Informix's turn. :-)

	I'd used Unify (years and years ago). It was fast but cludgy.
	Market opinions aren't shaped by logic, people make choices
	by making subjective value judgments. In other words, they
	buy what they like, irrespective of the fact that they may
	not know much.

> Somebody here in c.d.informix mentioned a while back the "20/80" rule: use
> 20% of your customers to get 80% of your revenue. I believe this approach
> shows time and time again that it will result in failure. Also witness using
> the same approach towards feature sets. Informix does not sell RLL alone to
> bolster their products, or just the dubious TPCs, rather the product mix they've
> had definitely outguns Sybase hands down. Only recently has Sybase figured
> out that they needed a tool company to bail them out. The engine alone isn't
> worth it. Think about it, if Powerbuilder only worked on Sybase, would anybody
> buy it?

	Well, Sybase users who wanted a front end might. I have an
	aversion to PowerBuilder (but keep this under your hat).

> As for Oracle, every industry needs a company that sells based on marketing
> rather than actually having good products. They've got market momentum,
> but not necessarily anything beyond that. Unfortunately they've proven that
> the Microsoft method of sales can work for database companies as well.

	I'll leave this for Oracle users to followup on (if they
	aren't already bored).

-am Received on Wed Dec 03 1997 - 00:00:00 CET

Original text of this message