Either there are more rows than you think, or there are triggers on the table doing things behind the scenes.

Given your sample data, this appears to do what you need:

insert into GWB ([Subject])

select name

from #StandardSubject SS

left outer join [#gwb-ok] GWB

on GWB.[Subject] like left(SS.Name, 3) + '%'

where [Subject] is null;

From...

Lynn Pettis (1/29/2010)

---

I seem to remember reading something somewhere (twitter...

I'd have a table of prerequisites, most likely. Lists "If you are" and "then you can become".

Going to give another vote for a standard archiving package for this.

Rather than archiving into a database, I might just export to csv or something simple like that. Would...

I thought the same thing.

Sure...

They have. Use top, and invert the order.

lmu92 (2/1/2010)

---

I'm struggling with a little ethical issue:

Let's assume you find the contact information of a company a friend of yours is working for in some sample data posted on...

amin 70510 (2/1/2010)

---

1. I want to express my appreciation for quick and detailed responses.

2. If I was "know it all" and had ego problem, I would not be posting a...

To do that, you'd have to first partition the data into separate files for each month, then you could back those files up.

Is the data partitioned?

I was less restrained than...

Sounds to me like you know more about this than everyone else does, so we'll wait with baited breath for your next revelation on it.

Since I already mentioned the limits...

From what I understand, most tempdb operations already take place purely in RAM, till they get too big to fit there. Temp tables and table variables both work that...

Computers can't think. They don't use logic. They just do what they're told. ...

These two articles and their discussions might help:

http://www.sqlservercentral.com/articles/Auditing/63247/

http://www.sqlservercentral.com/articles/Auditing/63248/