Re: open source PostgreSQL not supportable?

From: Frank Hamersley <terabitemightbe_at_bigpond.com>
Date: Thu, 12 Jan 2006 14:51:19 GMT
Message-ID: <H1uxf.214543$V7.21954@news-server.bigpond.net.au>

Christopher Browne wrote:

>>Marc Schoechlin wrote:
>>
>>>Hi !
>>>DA Morgan <damorgan_at_psoug.org> schrieb:
>>>
>>>>Some day open-source databases may be competitive with those built by
>>>>the big three just as Linux is now competitive in the area of operating
>>>>systems. But that day, for databases, is not today nor anytime within
>>>>the next two to three years.
>>>
>>>This is FUD :-)
>>>I used postgresql in software-projects for the first time in 2001
>>>for really business-critical systems without any real problems or
>>>limitations.
>>
>>And precisely when did your implementation pass an ISO9000 audit?
>>
>>And which accounting firm agreed to sign off on the financial
>>statements it produced?

>
> You're just making this up as you go along.
>

>>Did the FTC agree that it meets the terms of FACTA?

>
> Why on earth would the FTC have to evaluate a database implementation
> against a legal act that requires that three companies release credit
> reports to consumers??? That, of course, is the main purpose of the
> Fair and Accurate Credit Transactions Act.
>

>>Who audited it for Basel II compliance?

>
> Again, not relevant.

Correct - although confidence must be established with the regulatory bodies - hence XL doesn't cut it.

> Basel II is a regulatory mechanism to evaluate
> how financial institutions define their capitalization models. In a
> sense, it doesn't even exist yet.

Incorrect - the Accord is out there now - adoption is imminent. Most organisations would be assembling regression datasets to convince their regulators they are ready to throw the switch next year.

> Some countries plan to implement
> the accord in 2008;

Wrong - its 2007 for most countries, 2008 for the US and other wimps.

> that involves regulating _banks_, and their
> financial models, not the database engine implementors.
>

>>Or HIPAA?

>
> Peter Wayner's work, _Translucent Databases_, nicely documents how
> this kind of securing of information is properly handled within the
> application design, such that you do not *wrongly* entrust database
> engine implementors with control over whether or not the data is
> secured.

You should still ensure as far as possible that it requires a conspiracy to subvert the system. Then add the Wayner factor!

>>Basically what you have written reads as:
>>"I threw together something and it runs."
>>
>>Not to be too flip here ... but so can my mother.

>
> You're being nothing *but* "flip."
>
> But I suppose academics can afford that, as long as they already have
> tenure.

Cheers, Frank. Received on Thu Jan 12 2006 - 08:51:19 CST