Re: Other tree representations in SQL
Date: Sun, 8 Dec 2002 23:02:46 +0000
Message-ID: <DA3Bro2W+889Ewtm_at_diamond9.demon.co.uk>
In message <IvwH9.2325$La5.561977498_at_newssvr10.news.prodigy.com>, Greg
Boland <gregb_at_snet.net> writes
>I posted Top only because you don't need to read the rest to get an idea of
>what I'm thinking. The tree model, no matter how you do it, is contrived
>under the sql model.
Yup! There's a structure clash.
I have a sneaking suspicion that there's a model somewhere that fits both. To my mind the traditional relational database looks like a system for locating a datum using Cartesian coordinates. The tree structure looks like a system for locating a datum using polar coordinates.
Perhaps if we reframe relational algebra as relational geometry there's a Riemannian solution that unites the systems. (More likely it will add another mutually incompatible model.)
-- Bernard Peek bap_at_shrdlu.com www.diversebooks.com: SF & Computing book reviews and more..... In search of cognoscentiReceived on Mon Dec 09 2002 - 00:02:46 CET