Re: Nulls, integrity, the closed world assumption and events

From: Neo <neo55592_at_hotmail.com>
Date: 16 Jan 2007 13:37:54 -0800
Message-ID: <1168983471.869532.214700@38g2000cwa.googlegroups.com>

> Person(id, cars) Car(id, manufacturers)

Ok, so initially given john has no cars, tom has two cars, and bob has one, the MV solutions looked similar to:

Person Car
john

bob        nissan, porsche
tom        aston_martin

and with the new data requirements of 0-to-many mfgs, it became:

Person Car
john

bob         ->nissan, ->porsche
tom         ->aston_martin

Car                 Mfg

nissan
porsche
aston_martin gm, honda

What is used to link the cars in the two files? Their name or some type of ID? Does splitting the data from one file to two affect existing queries, ie find all persons who have nissan and porshe? I assume a human splits the data from the single file to two files, or is it handled automatically by db itself?

Because dbd is highly systematic, it doesn't require a schema change by user or splitting of data to different tables/files by user and pre-existing queries are largely unaffected. For example the following query to find persons with nissan and porsche stays the same before and after the new data requirements:

(& (get * has nissan)

    (get * has porsche)) Received on Tue Jan 16 2007 - 15:37:54 CST