Mark...

One of the newer and nice error trapping features in T-SQL is the implementation of the "Try-Catch" block, which I used quite a bit when working on developing corporate databases.

Another was the use of "select" to to test out functionality.  for example, with T-SQL in a database manager I can write the following...

          select 'trailing ''la'''  -> which returns trailing 'la'  when run.

You can do the same with the following Firebird-SQL statement...

          select 'trailing ''la''' from RDB$DATABASE

However, it took me a while to find this capability and I only did so by accident.  This is where a book on "Learning Firebird SQL" would of great assistance in moving the Firebird up the popularity rankings.  In fact, such books are far more helpful than the language references since these latter books assume that you have a working knowledge of SQL and do little to promote the Firebird database beyond the core community.

As it regards the above example, if I remember correctly, I believe there is a similar statement construct in Oracle.

However, the T-SQL implementation is the most elegant and the easiest to use and learn.

Cursors in T-SQL have also gotten very powerful and even more efficient with the later releases of SQL Server.  As a result, their complexity is far easier to learn and use.

As Oracle relies on cursors in a way that SQL Server does not, it is with this database engine that the power of the cursor is its most extensive.

These are just some examples off the top of my head so to speak...