The Forum for Discussion about The Third Manifesto and Related Matters

Please or Register to create posts and topics.

SQL 1992 `UNION JOIN`-- huh?

I've just bumped into

UNION JOIN

This join type have been introduced in SQL-92 language standard, but disappeared in later versions of SQL standard.

What was anybody thinking? (I use "thinking" very loosely.)

I thought the SQL Committee ratification process ground so slowly it takes forever to get even sensible ideas like NATURAL JOIN agreed. (1992 also was when NATURAL JOIN got blessed -- although many vendors to this day don't support it.) Do I call this other thing UN-NATURAL JOIN?

Quote from AntC on January 10, 2022, 10:10 am

I've just bumped into

UNION JOIN

This join type have been introduced in SQL-92 language standard, but disappeared in later versions of SQL standard.

What was anybody thinking? (I use "thinking" very loosely.)

I thought the SQL Committee ratification process ground so slowly it takes forever to get even sensible ideas like NATURAL JOIN agreed. (1992 also was when NATURAL JOIN got blessed -- although many vendors to this day don't support it.) Do I call this other thing UN-NATURAL JOIN?

UNION JOIN did not make it into the 1992 standard.  Nor did JOIN ON FOREIGN KEY or JOIN ON PRIMARY KEY.  I drafted the proposal to remove these items from the working draft, shortly after I had joined the UK group known as UKDBL.   I also suggested to my UK colleagues the possibility of adding NATURAL JOIN, and that resulted in the proposal that was accepted by the international committee.

Hugh

Coauthor of The Third Manifesto and related books.
Quote from Hugh on January 10, 2022, 12:08 pm
Quote from AntC on January 10, 2022, 10:10 am

UNION JOIN

This join type have been introduced in SQL-92 language standard, but disappeared in later versions of SQL standard.

What was anybody thinking? (I use "thinking" very loosely.)

I thought the SQL Committee ratification process ground so slowly it takes forever to get even sensible ideas like NATURAL JOIN agreed. (1992 also was when NATURAL JOIN got blessed -- although many vendors to this day don't support it.) Do I call this other thing UN-NATURAL JOIN?

UNION JOIN did not make it into the 1992 standard.  Nor did JOIN ON FOREIGN KEY or JOIN ON PRIMARY KEY.  I drafted the proposal to remove these items from the working draft, shortly after I had joined the UK group known as UKDBL.

Thank you Hugh for the historical perspective. I'm left wondering why anybody thought UNION JOIN was worth including in the "working draft". Had some vendor already implemented it, and was trying to get it 'blessed'?

Or did it come from some academic/theoretical perspective, as post-1992 with Snodgrass re Temporal Data?

 I also suggested to my UK colleagues the possibility of adding NATURAL JOIN, and that resulted in the proposal that was accepted by the international committee.

Hooray! But it didn't get much uptake amongst vendors. And there's a few vicious propagandists on StackOverflow alleging NATURAL JOIN is a disaster waiting to happen.