Flipping through Books On Line also serves well as introduction...

/Kenneth

So the deal is to start numbering the deptmt with the highest # users at 200, then on from there as the #users descends? And this is also a oneshot...

This is your basic import-scenario where you may have a staging table loaded with new data that should be updated if exists, or inserted if new.

I usually like to do...

Pardon me for perhaps straying a bit OT, but I feel that a few things may need to be sorted out.

We talk here about keys and indexes in a very...

Heh.. asking for a concensus on modeling?

Would probably be easier to get a concensus on religion or politics

As with most things, there's a...

Well, I've never used this particular style to alias myself, I just know that it's there and need to be 'looked out for'. Who knows? Perhaps it will go away...

Most of the times, wierd stuff like this is due to the data in some way...

Like things that makes a difference between writing a TDS stream to a result window compared...

This is how it's worked since the dinosaurs...

It's an old 'trap' when unintended, the solution is to keep tounge in cheek and place your commas appropriate.

/Kenneth

Yes, I know that. I wrote that you need a numbers table.

My aim isn't to always at all cost post 100% conclusive close-the-case posts.

I'm more interested in getting people to...

Well, 'paying attention' only gets you that far, and going ANSI lessens that burden somewhat.

Apart from that, the legacy join syntax is way overdue by now, and really shouldn't be...

The problem is that soundex / difference don't cut it.

It doesn't work as well as one would like.

As for how to use it, it's pretty well documented in BOL.

If...

What? And rob you of the opportunity to show how to do it?

Full service we may provide, though the best path to learning...

A method to make it easier to avoid such accidental crossjoins etc, is to abandon the legacy join syntax used here, and rewrite it using ANSI join syntax.

/Kenneth

Here's a way to do it.

select

x.custid,

       n.n

from

Another thing is - the queries is said to be the same, but what about the answers..?

Are both queries on both servers returning the exact same result?

/Kenneth