Posted by: nostra
« on: August 24, 2010, 11:18:38 pm »The feature is pretty buggy and I plan to fix it soon, but in a meantime it is really better to turn it off.
The purpose of that behaviour is to avoid the creation of duplicate records. I consider it a design flaw—which I hope will be fixed in 1.0. The check for duplicate records should happen after the new item has be selected and downloaded. Any existing (invisible) records can then be positively identified based on URL and deleted.