How about just doing triggers. Also you may want to look at some transaction log reader options if you want to see 'before' data.

Had one years ago from a third party vendor <cough>'CommerceOne'</cough> that was almost that long but it had a bunch of subqueries (about 40 or so if I recall)....

It might also be alphabetical location between the two. Suffice to say SQL Server does not translate.

No, unicode does not convert to something simpler, so the pictograph for Godzilla does not move down to become the letter A.

You can do a bit of that sort of...

If I am understanding question #1 correctly, you are asking if what is entered for a string has an impact on the collation used. No, it does not the...

A kill can have block implications outside of the realm they are aware of.

Depending on the nature of how connections are made there is a risk you may kill something...

Yes I have and my answer at the time was no...

You could look at offering it if they are using a read only connection or something, but yeah the rollback...

Is the first column a primary key? If so you could just do an update...

You could add the indexes in after you are done initializing by putting them directly into the replicant database after data is loaded. We do that to speed up...

Setting up replication for large objects can be fun...

Are you replicating indexes? You may want to try replicating without creating the indexes (it will force the primary key I...

So you would want 08/19/2010 13:15 to show up as 08/19/2010 13:00 ?

I think the batch size setting only applies to replicated transactions not the initialization....

I don't think you can really do anything about the batch size with merge replication

Have you considered setting the recovery mode on the target to simple?

When you say 'database' is this Linux process even trying to connect to SQL Server or is it just putting files in a folder?