[bumblebee]

If it will help, I can change the song matching behavior to be as follows:

1) Look for an identical match. If one is found, load it.
2) If no match is found, find all song titles that contain the given title. If only one match is found load it.
3) If multiple matches are found, determine which one is the closest to the given title (in terms of the number of extra characters).  If there is a clear winner, load it.
4) If multiple titles have the same number of extra characters, load the first one in the list.

Mike

Complicated but it just might work
I have many songs in notation with different keys or different real books (different chords). I wonder if this would work with them also. I presume putting its source in the filename might work.
(Clarification, this refers to many songs of the same title but different keys, chords or whatever).

I was thinking about being able to use a number as another option. Songs could have individual ID numbers (perhaps as metadata in a field).
Or within a filename or title in a specific format e.g. #35# so it can be filtered out from display. The same song (but different keys or lyrics etc.) all can be found by each slave by searching for the number sent by the master.
Would something like this work also?