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

"Anthony W. Youngman" <wol_at_thewolery.demon.co.uk> wrote in message news:v5os1HDGTSrAFwl8_at_thewolery.demon.co.uk...

> PERFECT! That's a quote I would have loved to have had available to me
> earlier :-)

The thing is, I'm not satisfied with the arguments of either the pro relational camp or their challengers in this forum.

There's clearly a lot of intelligence and erudition in here, but it seems to be savagely misused, on both sides of the argument.

I've used the power of relational joins, ever since I was first exposed to the concept. And my first use involved nothing more sophisticated than Datatrieve and indexed files on a VAX. And the theorists in this forum who dismiss that as "not relational" have a fundamental synapse missing with regard to the connection between theory and pragma.

I've never used PICK, but from what I've read in here, if one were to study the reason why certain PICK applications were (and possibly still are) successful, and do the same for Datatrieve, one would find a surprising overlap. And I think data models would play a minor role in both studies. I'd love to see some rational discussion of that. But we'd have to get away from some of the cultural norms of this forum.

For me, the migration from Datatrieve to VAX Rdb/VMS, and later from that to Oracle were pretty natural. While I find much to criticize about SQL, it's far, far better than the access languages that grew up around CODASYL databases! If a better language can be designed, implemented and adopted, I'm all for that! But don't expect me to wait! Received on Fri May 21 2004 - 15:30:25 CEST