Jim,

>Your suggestion is not relational at all. It is simply concealing a record oriented ISAM approach to data with a wrapper that makes it look relational.

Somewhat true, yes, this is the purpose.
>
>To what authorities do you attribute the statement, "SQL is a bad implementation of the relational model"?

I think you will have read DATE: "An intruduction to modern database systems" (or something). Date here describes perfectly why SQL is a bad implementation. Just recently I've read somewhere that even oracle is only a 60% implementation of the relational model. In addition I read some hardcode DB oriented magazines in which the writers do state the same. Too me this is NOT an opinion but a FACT.

>xBase is NOT relational.

Certainly NOT, Neiter is SQL-Server nor ORACLE their implementation of the relational model is fairly weak.

>

It fails most of Codd's original 13 rules.

True,

>

It is record oriented, the relational model is rooted in mathematical set theory.

Though it's record oriented, we could skip this part and work set-oriented: this is also possible.

>

It deals with record pointers, the relational model states that the only way to identify an instance of an entity is through the primary key.

We don't have to use record pointers, nor the commands which affect the current record, We could also use SQL commands.

>

It requires that the programmer know the physical storage structure to deal with the data, relational model says the developer need know only the logic structure and need know nothing about the physical structure.

So much for your generated primary keys :), yes it's true, but if you look at my concept, it isn't hard to automate the process and let the computer determine the structured hierarchy by specifying a limited SQL query (no group by and having clauses), though it capabilities might reach further. Maybe this wish could also being described as a FILTERED SQL resultset which is editable, and the indexes are available (both implicitly and explicitly). This FILTERED SQL resultset also must be available when joins are used.

>

xBase has many commands in it that allow access to the data by bypassing the relational logical structure, relational theory says the only way to access the data is through its relational structure.

But we can use the relational approach, can we ?

>That's just the beginning of the list that could be built. Arguing that xBase is more relational than SQL is pure hogwash.

You'll never see me saying this, I'm only saying that this solution is more toward the relational model than the SQL-select implementation of SQL. You might want to check this by studing (in the book of DATE) the mathamatical operations you can do with relations (Sadly i don't own the book anymore, maybe you've got an ISBN number). You may find much of simularities between the operations in the relational model and what i'm trying to do here. That i must use indexes is just an implementation detail, but could be hidden if the wish was applied

>Relational databases were invented by E. F. Codd and SQL was invented by E. F. Codd for the purpose of working with and manipulating relational database in a purely relational fashion.

To my knowledge SQL was invented by IBM (though i don't know if E.F. codd did help with this implementation). If he would have much to say about the implementation, SQL would not be as worse as it is today... Or Codd and DATE have a huge difference in opinion up here, which i think is highly unlikely.

Walter,