Re: Unknown SQL

On Wed, 30 May 2001 19:50:37 +0100,
"Sam" == Sam Staton <sam.staton_at_bigfoot.com> wrote:

>> this is a very simple and good example where a flat, textual representation
>> of data seems to beat objects. I wonder why none of the other gurus here has
>> come up with the question yet.

Sam> It is worth mentioning that the relational model does not facilitate
Sam> for functions of aggregates (as Mikito hinted). The addition of the
Sam> GROUP BY / HAVING clause to SQL caused a large amount of hastle with
Sam> getting the specification right, and made the spec considerably more
Sam> messy, since it is not natural to the relation model (according to my
Sam> lecturer, Ken Moody, Univ of Cambridge UK).

I think agree with the 'not natural' bit, but on the other hand, since RDBMSs allow you to deal with (large) sets of tuples, it is very convenient, nay essential, to have ways of 'summarizing' such data using the aggregate functions. So while it may not be entirely natural (however defined), I would not call a SQL database without COUNT, DISTINCT, GROUP BY, HAVING, SUM, AVG, MIN, MAX etc. a proper RDBMS.

Sam> Thus if aggregation is difficult in an object-oriented context this is Sam> not necessarily a disadvantage compared with the relational model.

Mmmm ... I guess this is (another) instance of the 'prior art' raising the bar.

                                                                      Philip

-- 
If you have a procedure with 10 parameters, you probably missed some. (Kraulis)
-----------------------------------------------------------------------------
Philip Lijnzaad, lijnzaad_at_ebi.ac.uk \ European Bioinformatics Institute,rm A2-08
+44 (0)1223 49 4639                 / Wellcome Trust Genome Campus, Hinxton
+44 (0)1223 49 4468 (fax)           \ Cambridgeshire CB10 1SD,  GREAT BRITAIN
PGP fingerprint: E1 03 BF 80 94 61 B6 FC  50 3D 1F 64 40 75 FB 53