Re: Early and late binding.

"mAsterdam" <mAsterdam_at_vrijdag.org> wrote in message news:43d226c9$0$11073$e4fe514c_at_news.xs4all.nl...
> In prolog there is no difference. That is the other extreme.

In Lisp also, there is no difference. A list can be constructed as data, and then evaluated as process.
This is particularly useful in automatic programming and/or automatic debugging.

> In SQL itself you can can mix DDL and DML, if you so decide.
> Some applications do. The times I saw that mix used I did not like it,
> and could trace it down to poor design. Maybe this is necessarily so,
> maybe not. Thoughts?

It depends.

Some DDL, like CREATE INDEX or DROP INDEX, has relatively little consequence (except for performance). It wouldn't be necessarily a "bad thing" (tm) to put these operations into the stream of end-of month processing, intermixed with ordinary DML.

Also, if the underlying information requirements, or the known part thereof, is a moving target, then DDL is called for. This moves into the domain of mountain man's evolving schema. If however, the information requirements are static, and the application is constantly doing DDL to reflect ongoing routine transactions, something is probably wrong somewhere. Received on Sat Jan 21 2006 - 18:50:37 CET