Re: Reminder, blatant ad

Jan Hidders wrote:
> dawn wrote:
> > Jan Hidders wrote:
> >
> >>dawn wrote:
> >>
> >>>I'd be happy for you to spell it out, Jan. You are pro-RM and
> >>>pro-XQuery, is that right?
> >>
> >>And pro-Pick, if that happens to be the best tool for the job at hand. ;-)
> >
> > And a politician too. A question, then, is when you think a non-RM
> > model would be advisable.
>
> That question is way to broad and over-simplifying so I cannot give a
> clear and meaningful anser to that. It's like asking 'when would you not
> use a hammer?'. Are we talking about about the RM as a data-model per
> se? Or as a data model at the external level of a DBMS? Or at the
> logical level of a DBMS? Are we talking about existing DBMSs or
> hypothetical ones? These are all different questions with different
> complex answers.

OK, I'll take a quick stab at a clearer question. If you were writing a "data processing 101" type of application, such as a "Pet Store" app (if you are familiar with the Java and other Pet Store examples) or the Joke Response System described on my web site http://www.tincat-group.com/swdev/jokesreq.html and you had complete freedom in tool and standards selections, would you be more likely to choose a Relational Model to work with your DBMS or an "XML data model" or other model that includes attributes with list values?

Additionally, if you chose a SQL-DBMS tool for practical purposes such as wide-spread use and availability, is there something lacking from current tools that use alternative data models (e.g. XML database, Cache', PICK databases such as IBM U2, Berkeley-DB,...) that would cause you to reconsider if such features magically appeared?

Thanks! --dawn Received on Thu Feb 09 2006 - 01:54:06 CET