This is probably a crazy question because I don't really know what I'm talking about. But maybe someone who does will indulge me. :-)
We essentially want to repeat a SQL query to a production database every few minutes, and when it returns a certain result, then run a job (but we don't want to run an actual job to do the query that often, because this particular workflow has about 300 instances active at any given time).
So were thinking to put it in a SQL variable, and use that in a pre-condition.
However, DBAs say we should use bind variables for this query. Is it possible in the format of a SQL variable object? Or is that like making seven red lines but some of them are green? (http://www.cnet.com/news/this-is-how-an-engineer-feels-when-hes-surrounded-by-idiots/)
(P.S. this function is currently done by 300 perl jobs running in parallel, doing the check every few minutes, which seems like a perfectly reasonable approach. But there are long-story problems with it.)