Re: ORACLE RAC CRASHES ORBITZ

From: Darin McBride <dmcbride_at_naboo.to.org.no.spam.for.me>
Date: Tue, 29 Jul 2003 22:05:45 GMT
Message-ID: <ZcCVa.557958$3C2.14299559@news3.calgary.shaw.ca>

Daniel Morgan wrote:

> Darin McBride wrote:
>

>> Daniel Morgan wrote:
>>
>> > Blair Adamache wrote:
>> >
>> >> It's possible that DB2, Informix, Sybase and 'C' compilers have all
>> >> caused crashes. Never heard of a C compiler crash that interrupted
>> >> production and got the same level of coverage as Orbitz did in eWeek
>> >> and Computerworld. I guess when you buy the back page of the Economist
>> >> and Businessweek and use the word "Unbreakable", you attract a little
>> >> more attention - the same way a blivet attracts flies.
>> >>
>> >> <snipped>
>> >
>> > So now Larry is responsible for the amount of news coverage? What has
>> > the amount of coverage got to do with anything?
>>
>> I think the point is much more simple than you want to admit:
>>
>> * Oracle advertises "unbreakable".
>> * (It is claimed that) Oracle broke.
>>
>> All I can say is that I hope that Oracle execs (who must have approved of
>> the "unbreakable" campaign) have lots of seasoning to add taste to their
>> feet.
>>
>> > I thought we were involved in technology not marketing hyperbole. Leave
>> > the P/R nonsense to the weasels. Yep Oracle put out the phrase
>> > "Unbreakable". P&G put out the phrase "Whiter and brighter"? That's
>> > what they pay people in marketing to do in lieu
>> > of having real jobs. Or don't you  remember when someone got paid a lot
>> > of money to add UDB to DB2? So why do you go after one and not the
>> > other?
>>
>> 1. I've not heard of anyone saying, "Here's some data that can't be
>> stored in the Universal database."
>> 2. I've not heard anyone claim that DB2 is the only "universal" database.

>
> At one level I agree with you. Nothing is unbreakable. It is impossible to
> conceive of a product that put on the wrong hardware, hosted on the wrong
> operating system, run through the wrong web server, used to execute bad
> code, and managed by a bunch of monkeys wouldn't break. Anyone that
> believes otherwise is dummer than a bag of hammers.
>
> Assuming you, and others, bear no relationship to the above referenced bag
> ... why the literal interpretation of what is clearly marketing hyperbole?

Is it marketing hyberbole when Sun advertises 99.999% uptime? Generally, I take anything that is subjective to be hyperbole ("Coffee Crisp is a nice, light snack" - whether it is nice, light, or a snack [vs breakfast] is all pretty subjective). Anything that is objectively verifiable, I, and I suspect many others, would take much more literally.

If Sun only manages 99.99% uptime, don't you think that their competitors would call them on it? Someone may make a decision, after reading spec sheets, to go with a certain vendor based on those spec sheets, and a competitor proving the spec sheets were wrong would likely also be factored in somewhere. Oracle claimed 100% uptime. Yes, you and I know that's basically impossible. But that's the claim. And it's not a subjective one.

In short: don't make marketing claims that you can't back up. That's not hyperbole. That's lying. Even if it's wishful thinking, it's still lying because it is said with such an air of confidence that the confidence is the lie (it's unknown and said as if known).

I wouldn't expect any differently should my own employer be caught lying like that. (Personally, I'd be affronted if they tried, and may even think about employment elsewhere, and this I'm saying even though my second-line manager is participating on this thread.) I don't expect untruths to be masqueraded as "hyperbole" in one place any different than any other place, whether that's Oracle, MS, IBM, or any other vendor. Received on Tue Jul 29 2003 - 17:05:45 CDT