Re: In an RDBMS, what does "Data" mean?

In message <9X%vc.5730$tB.3511_at_newssvr32.news.prodigy.com>, Eric Kaun <ekaun_at_yahoo.com> writes
>> So, if you don't have experiments to show that real-world data ALSO
>> comes in tuples (or a close approximation thereof), then you can't
>> conclude that a relational database is a good place to store real-world
>> data.
>
>Sure you can; evidence <> proof. The nice work logicians and mathematicians
>have done with predicate calculus over the years, while perhaps not
>corresponding to "the real world" (tm, MTV Networks), gives us nice
>machinery with which to manipulate... well, data. What, precisely, would
>allow you to conclude that a <datamodel> database is a "good place" to store
>real-world data?

Yup. Evidence does not equal proof. But that was not what I was getting at. Note my careful use of the phrase "or a close approximation thereof" :-)

If "real world" data is not a close approximation of "relational data", then it is reasonable to conclude that a relational database is not a good place to put it ... :-) And if the two are a close approximation, then a relational database may not be the *best* place, but it has to be a *good* place.

Don't forget - I'm a scientist :-) If the stats are 95% confident, that's not "proof", but it's "good enough".
>
>> Sorry for ignoring the rest of your post, but this is ABSOLUTELY
>> FUNDAMENTAL!!!
>
>Perhaps, but I still don't think "data comes in tuples" is anything like an
>axiom. I could certainly be wrong.

Read C&D's first rule! "Data comes in rows" - which is as far as I can make out, a synonym for "data comes in tuples". I'm sure a relational guru will disagree, but I can't see the difference ...

Cheers,
Wol

-- 
Anthony W. Youngman - wol at thewolery dot demon dot co dot uk
HEX wondered how much he should tell the Wizards. He felt it would not be a
good idea to burden them with too much input. Hex always thought of his reports
as Lies-to-People.
The Science of Discworld : (c) Terry Pratchett 1999