Know, don't think

I recently had the opportunity to sit down with a great SQL Guru ( he'd call himself a tinkerer ).  It was over dinner prior to SQL Server Saturday 60, and I gained a lot of nuggets of goodness.  In no particular order:

1. Remember, despite what the relational zealots say, a database is a file system: Input - Process - Output; deal with it, and use it to your advantage.

2. Know, don't think you know.  In other words, use the scientific method to be sure that he assumptions that you are making are correct.  Create a test, and experiment to verify your results.

3. Watch out for parallax.  In short: If you are looking at your speedometer, if you look from the right side, you are going 50; if you look from the left, you are going 75, but only if you look right on do you see that you are going 65.  So don't taint your results by looking at it with a bias.

4. Test with millions of records, not 50.  Just because something works with 50 records means that you just don't have any syntax errors.  If it doesn't work ( perform ) with millions of records, you are not doing it right.

5. Whatever you do, stay far, far away from torpedo fuel.

The last two points are important, but more "you had to be there":
1. Take your wife out on dates.
2. Cats are sentient.

You really had to be there.

Welcome to my Node

Welcome to my blog.

This has been a long time overdue, but the Node has been kicking around in my head since the good ole days when you could explore all the sites in a Yahoo branch; back when the Internet had an end; an edge.

I'll post what I find useful; what caused sweat and tears on this side and will help the next wave; because lets face it, none of us would be as far along if it wasn't for those that blazed the trail for us ( or at least left a trail marker now and again! )

So please feel free to reuse anything you find here; if you find it useful or find a better way, feel free to shoot me a message; otherwise, take care, and thanks for stopping by.

Austin