Re: Relational query with path expressions

On Apr 4, 11:05 am, paul c <toledobythe..._at_oohay.ac> wrote:
> David BL wrote:
>
> ...
>
> at least one nuance of TCLOSE here without mention of recursive
> structures to ease the practical problems. more mixing up of physical
> and logical, not to mention the 'choices' presented as alternatives,
> when one of them Codd called the 'connection trap' about forty years
> ago.

Where was there a nuance of TCLOSE?

Mixing of physical and logical? You think the path expressions imply some mapping to a physical implementation? Actually they are nothing more than a convenient way to express queries at a logical level for a given a relational schema.

The proof is in the pudding as they say. Whether you dislike the intuition behind the path expressions or not, you have to admit they are remarkably concise for every example from Date's Intro book. Care to comment on that?

With regards to the 'connection trap', I already commented that n-ary relations are not generally decomposable into binary relations in the sense of a join dependency. You appear to have missed the point (where I rejected the idea of "navigating" from attribute to attribute along an "access path").

> just a suggestion, start a phony family tree at ancestry.com, try to
> put in a source for a marriage between two people and then check if
> ancestry's pathetic interpretation of the hierarchical gedcom
> so-called standard assigns the source to both spouses.
>
> also consider the rather logical definition of cousin on wikipedia and
> ask whether the offered syntax can ever possibly deal with zeroth
> cousins (siblings) or second cousins, not to mention cousins several
> times removes.
>
> you have every right to express various motives here, but they look to
> me, too scatterbrained for most non-mystics to comprehend.

Evidently you don't realise I was regarding the unary functions derived from binary relations as merely a curiosity, and not to be taken too seriously.

For someone who thinks of himself as a non-mystic, you seem preoccupied  with intuition rather than the mathematical content of my original post, which defined a query language. If you think it's problematic then please explain why without your hand waving. Received on Sat Apr 04 2009 - 06:32:54 CEST