more questions from the reasonably na ïve regarding cursors, record locking, and committin g

I originally posted this questions on the SQL light mailing list but in 
second thought, some of them may be more appropriate to pysqlite.

I think some of these questions fall in the category of "running in the
dark with scissors".  If it's a simple matter of I need to read
something, just point me at the appropriate documentation that I've 
obviously messed.

I'm trying to get a handle on this cursor thing.  it looks like it 
preserves state of some sort but I'm missing what state it is 
preserving.  For example, do I need a different cursor for each table in 
the database?  the reason I ask is that in my application, multiple 
tables will be in play at the same time in different threads and processes.

Record locking.  I did read http://www.sqlite.org/lockingv3.html  I do a
fair number of read/modify/write cycles on individual rows in all of the
tables from multiple processes.  if I interpret the documentation
correctly, I need to create a transaction and when I am done changing
things, "commit" which is really just turning on auto commit?

Committing.  how often should one commit?  Is it even necessary if you
close?  What happens in case of a crash?  Does journaling save one's
bacon?  If the change has not been "committed" will be visible to other
processes using the same table?

Many thanks for your answers and patience.

-- 
Speech-recognition in use.  It makes mistakes, I correct some.