I talk from Sao Paulo, Brazil.

I develop system for industries and all of many then, I use SQL Server 2000 ( Small Business Package ).

My projetcts are all in VB6.

And some time ago I read an article in a Brazilian .Net Magazine about StrataFrame.

I was very interested in use strataframe, and I saw a demonstration.

It´s really fantastic.

But, some questions made think about use strataframe.

one of them is, how I can manage table strutucts manually in merge replication databases?

specially in tables that are in replication. because new tables, we can create in query analiser. but in tables that are in replication, the fields are create using enterprise manager replication monitor, adding columns in articles!

How DDT create all structures alone, this fields will not be create.

Buy.

No, not really.  We use SMO (SQL Server Management Object) to deploy most of the structures and only use T-SQL when using pre and post deploy scripts, etc.  If you allow the DDT to automatically create your CRUD sprocs, then you can see the sprocs that will be created.

We also do a lot of other stuff behind the scenes so that you don't have to actually upgrade structures in order (i.e. going from version 1 to 3, for example). 

You do get the source code hen you purchase SF so you can see what we do, but it is not all T-SQL script that gets used.  Hopethat makes sense.

A little trick that you can do though is use a packet sniffer to see all of the T-SQL code that gets executed.  Ultimately all code is broken down into T-SQL script...even through SMO.