I've tried and never quite succeeded in producing a comprehensive list of all the glitches that you might encounter with drag and drop. I think the only acceptable number of glitches should be zero, but RM is not at the zero defect level on drag and drop. Of course, some of the "glitches" are sometimes viewed as advantages. In any case, we have to remember that behind the scenes, drag and drop is a GEDCOM export and import which automatically includes RM extensions to GEDCOM. That being said, here is a list from memory.
- Drag and drop goes by the GEDCOM options on which facts are exportable or not. There is no "drag and drop all facts" option. To avoid losing facts, you have to go into Lists->Facts and turn on GEDCOM export for all your facts and then go back and turn off GEDCOM export for any facts that you want to leave turned off. There is no warning about lost facts.
- Trailing carriage returns in notes are lost.
- Description fields longer than some limit - maybe 72 characters or something like that, I don't remember the exact number - are truncated.
- Some DNA data is lost.
- Unused places, sources, user defined facts, and user defined source templates are lost. This is often used cited as an advantage rather than as a glitch.
- User defined source templates become duplicated rather than being put back together properly. The net result is that Master Sources which use user defined source templates do not merge correctly.
- Color coding is retained, but Named Groups are lost. Similarly, the Set Relationships indicators are lost.
- "Not a problem" indicators are lost.
I don't use shared facts, but i believe that when I played with shared facts and drag and drop that they exported correctly. I could certainly be wrong about that. If shared facts don't drag and drop properly, that would be just one more in a very long list of reasons not to use them.
I don't use RM's to do list or correspondence logs. I do use RM's Research log. I haven't tried a drag and drop to be sure the Research Log is handled correctly.