Re: SQL server Vs Oracle

David <desertfox_at_thegrid.net> wrote in message news:_bi_2.780$i4.68153_at_alfalfa.thegrid.net...
> will win. Do you for a moment think that the 'limited' desktop market is
> insignificant?

In terms of database servers yes, the desktop market is completely insignificant. Or do you classify as database server a Dbase system?

> actually said. I do not work for MS and could care less about their
> marketing. Is your hatred of MS so deep that you have to use it as a Straw
> Man and not discuss issues that are actually a part of this discussion?

Sorry, I don't hate them. I use their products in my PC's too. Because I have to. The reasons why are irrelevant here. And some of them are extremelly useful to me. What I hate is people that accept their marketing drivel as some form of gospel without looking at the past history of this industry. Which is NOT the same as saying that I hate your ideas or point of view, OK? We are simply exchanging views and arguments, or can't that be done anymore?

>
> As a side point, you are aware that marketing is more important from a
> business standpoint than technological advantages?
>

Sure. Completely irrelevant though in a technical discussion. Any tecnical argumentation that uses marketing "facts" is therefore flawed in this context and deserves to be shot at. Metaphorically speaking, of course! :-)

> The Devil is always in the details.

Precisely. And the details say: the "add cheap hardware" argument is completely flawed in view of the long past, mid-term and recent history of this entire industry.

> How many 'rules of thumbs' have evaporated into thin air as technology has
> marched forward thus eliminating these things we take for granted.
>

Name ONE in the field of databases and data management.

NO, faster/larger disks/CPUs are not the reason why proper application and database design can be relaxed: the result of that is a spiral of even higher demands, as current state of affairs is proof plenty. And the reason proper design rules were thought out in the first place!

The problem is not new. Trying to solve it by means of marketing IS new, though. And it is proving beyond any doubt that it is the wrong solution.

> How do you think MS is able to run their entire site using IIS? They have a
> massive server farm to make up for NT's scalability weakness. So in their
> case, using more hardware WAS the solution!
>

Wrong. ADDING more hardware, the initial argument, is NOT the same as USING more hardware. In the context of this discussion, adding more hardware means adding more power to an existing single system. More memory. More disk capacity and I/O paths. Faster CPU. Like MS abundantly recommends as the "cure all" of any NT problems. Solves NOTHING except in very limited cases.

USING more hardware to spread the load accross parallel processing systems is nothing new and we certainly need not NT for this or to prove it. Been
done since the year dot of this age. What is so special about MS and NT that makes this suddenly such a GREAT technological solution?

>
> Interesting how your logic tells you that your 450 MHz PII runs Word slower
> because its design is 'utter crap'. Once again, try substantiating your
> arguments with some proof rather than making these sweeping generalizations.
> I would imagine Word97 has a hell of a lot more features than Word2.
>

From a design point of view, a word processor that has grown by two orders of magnitude in its storage requirements, CPU needs and memory demands and continues to deliver a simple fact: word processing, is IMHO a piece of design crap. Regardless of whatever "nice" features it may have. Let me explain with an
analogy: suppose someone added great features to your prefered model of car, such as anti-lock brakes, continuously variable transmission, auto-diagnostics and such. But to do this they had to make it the size of an interstate transport
truck with two trailers. Great show, lots of features. But all you wanted was something to take you to the supermarket and back... Of course, because the price of this monster is the same as last year's model, you think you are getting real value for money?

Like all analogies, this may be flawed. But it explains my point of view a little bit better, I hope.

> >Look at the history of this industry for the last 20 years! Have you EVER
> >seen an hardware improvment that wasn't immediately matched by more
> >complex software? This then grinds it to a halt and makes a yet faster
> version
> >of the hardware needed.
> And what is so diabolical about this? Faster hardware and more software to
> match it... that actually sounds like a good deal! Progress.. you gotta love
> it.

Absolutely nothing is diabolical about this. Why bring religion here? What is incorrect is to use the "add cheap hardware" argument as a solution when everybody knows it isn't, because it has never solved anything. As the history of this industry abundantly shows. I had this same discussion on Compuserve in 1993. Then the argument was the same. I argued then that in 5 years, we'd all
be discussing yet again how "adding hardware" would be the solution to the demonstrable problems of a flawed MS solution. Here we are again EXACTLY as I predicted, "adding the hardware". The problem still the same: bad design, worse consequences.

Have we saved ANY money? NOPE, absolutely NOT, we went through the upgrades to VL Bus/PCI/AGP, the upgrades to 486/Pentium/PII/PIII, the upgrades to networking capacity, the upgrades to memory bus systems and their string of incompatibilities, all the expense involved in re-training/re-toollin/re-equiping.
Cost of all that? I'll let you work it out...

The result? "Add more hardware", of course! Yeah, right!

When is this industry gonna open their collective eyes and stop re-inventing the wheel every six months? There is no such thing as a "silver bullet" solution
to what are in essence problems identified long time ago. The more we spend trying to masquerade (market) products into these magical bullets, the more we'll get
involved in this spiral.

>
>
> Old crappy systems are always going to be around in one form or another. If
> you want to live in a world of fantasy where MS will never dominate the
> market it chooses to compete in, then go right ahead.

So far they have competed in the PC/LAN arenas. After nearly 15 years, they won. Great. All the power to them. I'll be long gone before they win anything else.
And no, flogging a product for free to create a market is marketing, not a technical achievement.

> For the rest of us, we
> will continue to watch MSFT stock double and split year in and year out and
> simultanouesly watch their market share grow and grow.

Keep watching. I prefer the real world where people actually have to DELIVER something. Surprising how effective that is in weeding out "addon hardware"... And market share price was never a measure of any technical advantage.

--
Cheers
Nuno Souto
nsouto_at_nsw.bigpond.net.au
http://www.acay.com.au/~nsouto/welcome.htm Received on Thu May 13 1999 - 06:49:18 CDT