When I learned SQL, I was taught that the way to do a join was with a WHERE clause. For example, if you have orders:

1
CREATE TABLE orders (
2
   order_id bigserial PRIMARY KEY,
3
   customer_id bigint REFERENCES customers(customer_id),
4
   ...
5
)

and order lines:

1
CREATE TABLE order_lines 

The video for the September 8, 2009 SFPUG meeting is now available:

For downloading, the direct link is here.

This work is licensed under a Creative Commons Attribution-Share Alike 3.0 United States License.

The video for the August 11, 2009 SFPUG meeting is now available:

For downloading, the direct link is here.

This work is licensed under a Creative Commons Attribution-Share Alike 3.0 United States License.

Jeff Davis has written a superb article about the problems with NULL in SQL. He has it exactly right when he says:

I think the best way to think about NULL is as a Frankenstein monster of several philosophies and systems stitched together by a series of special cases.

The closest single operational definition of NULL I

For the last few months, I've been honored to videotape PostgreSQL events. The last one was PGDay San Jose 2009, which was the day before the start of the 2009 OSCON. I didn't attend OSCON this year, in part because one of the great pleasures of OSCON for me was getting to go to Portland; visiting San Jose just isn't

From a demo of PostGIS:

Presenter: “Of course, this is all fake data.”
Audience member: “I would have thought that the use of Godzilla would have indicated that it was all fake data.”

From the PostgreSQL day today at OSCON:

Q: “What did XML get you (in this design)?”
A: “XML got us pain.”