I can send mail using database mail perfectly well but always have to specify an "email address" in the recepient. In the past the alphanumeric pagers I have work with...

And please, post DDL too so that we are not blindly guessing ...

>>((oh btw - the views are generated dynamically using sysobjects to determine table requirements, so as new tables are added, so too is the view modified)

<<

Just remember to run sp_refreshview...

or if you can use openrowset

select a.*

from openrowset('SQLOLEDB',...., 'exec proc1') a

inner join

openrowset('SQLOLEDB',...., 'exec proc2') b

where ...

>>This uses Double Precision, which I've never heard of in T-SQL until recently, but it still has trouble with certain values. <<

From BOL definition of float:

The synonym for *double precision*...

Or

SELECT STUFF(@Var,@Position,1,'?')

Be careful not to generalize client side trnsaction handling, because it will deteriorate the concurrency of the system due to longer transaction times (which is the advantage of doing it...

The things that have not been mentioned, are where 2005 is a REGRESSION from 2000 ...

To the shame of M$ there are *many*.

Here are some:

1.Replication Management: Replication monitor is way...

Craig, Try to manipulate a TEXT variable on the server ... you'll understand what "difficult to manipulate" means

Not sure about what system (or settings) you have but:

select *

from

(select 'Fred Johnson' mycol

union all select 'Fred') data

where mycol LIKE '% %'

Returns *only* 'Fred Johnson' to me

-First make sure that

fd.synonymID

dm.synonymID

d.synonymID

dm2.synonymID

are all of type *int*

-Secondly if your sum is bigger than an INT the cast should be made to BIGINT.

Cheers,

tough luck. When you need trailing columns that are blank you must at least provide the last comma (csv) or use "" for your char fileds. The only solution is...

Although I got it wrong the words "any constraint" means more than just PK, FK and UQ. There are also "DEFAULT" constraints, "CHECK" constraints which are definitely not captured in...

I have had same experiences with M$ more than once... and in an Corporate environment

My personal opinion is that although you are getting better "cohesion" with the sepparate methods, this is probably one of those examples where it may not be what you want....