Re: Is SQL silly as an RDBMS<->app interface?
От | nolan@celery.tssi.com |
---|---|
Тема | Re: Is SQL silly as an RDBMS<->app interface? |
Дата | |
Msg-id | 20030713151714.11944.qmail@celery.tssi.com обсуждение исходный текст |
Ответ на | Is SQL silly as an RDBMS<->app interface? (Antonios Christofides <A.Christofides@itia.ntua.gr>) |
Ответы |
Re: Is SQL silly as an RDBMS<->app interface?
|
Список | pgsql-general |
> I understand that SQL is the interface between apps and RDBMS's because > of history, not because it is correct design. Could you point me to a > link or book or paper that deals with this paradox? Thanks! I'm not sure what you mean by 'correct design'. I think you should go back and read the works of Codd and Date on the development of relational databases. One point that was made early on is that RDBMS theory doesn't guarantee efficiency, but it does guarantee consistency and accuracy in the results. (Efficiency is an implementation issue, though without high-speed computers relational databases would probably not be practical anyway.) The same can be said of the SQL standard and any 'natural language' query. There is always a way to structure a query properly, even though it may be ugly-looking. It isn't about how the query looks, it is about being able to ask the question--ANY question. Back in the late 60's, when I was a budding CS/EE major, the big hardware 'concept' was associative memory, which was supposed to be more human-like in terms of how it functioned, though grossly inefficient in terms of its use of hardware cycles. -- Mike Nolan
В списке pgsql-general по дате отправления: