Another FTM Issue - Unattached Family Records in My Tree

As I reported here and here I am separating my Family Tree into multiple branches. I have discovered that there are thousands of unattached family records in each of my new branch files.

Reviewing my original Family Tree I also found over a hundred similar unattached family records in it.

I discovered this by reviewing the GEDCOM created by FTM when I export some or all of a tree. These unattached records have no HUSB or WIFE records associated with them and float around the database. For example:

0 @F408@ FAM
1 MARR
2 DATE 12 DEC 1700
2 PLAC Suffield, Hartford County, Connecticut

UPDATE

I just discovered unattached family records with nothing but the level 0 GEDCOM record. Here is a section of the GEDCOM. These are also NOT removed by compacting.

0 @F18656@ FAM
1 MARR
2 DATE 27 FEB 1707/08
2 PLAC Newton, Middlesex, Massachusetts, USA
0 @F18659@ FAM
0 @F18660@ FAM
0 @F18662@ FAM
1 MARR
2 DATE 26 NOV 1919
2 PLAC Proctor, St Louis, Minnesota, USA
0 @F18667@ FAM
0 @F18670@ FAM
0 @F18671@ FAM                            

Updated 6/3/2015

Because they are not attached to an individual, I can find no way to manually delete them from the database from inside FTM.

I have tried several tricks to fix this but to no avail. Compacting the file does not fix the problem. Nor does the export of the Tree to a different FTM file rather than directly to a GEDCOM file.  Years ago there was a workaround for repairing a tree, But this no long works.

At this point in my project, I  think my only choices are; live with the problem, edit the GEDCOM file and manually delete each GEDCOM entry, or I could write a program to do those deletions.

Those two latter options would require a load of new databases with the corrected GEDCOM and a manual reload all of the associated media. None of this is to my liking. I hope that this post will result in someone offering a much better solution.

How is this caused? In my case and in this recent project I often used several of the reports that allow selection and deletion of groups of individuals. I suspect that one or all of these deletions are done without regard to the family and/or marriage records of those individuals. If this is true it demonstrates some very lax testing on their part. This seems to be a continuing theme.

I cannot explain the presence of the unattached records in my original file. This file has been in effect for nearly 15 years and has undergone various updates as I have implemented new releases. It could stem from data entry errors I have made in entering new information, deleting miss-entered individuals, or as the result of merging GEDCOM files, with these errors in them, into my tree. But they do exist and, in my mind, FTM should not allow them to exist.

But, perhaps I am wrong and FTM did intensive testing and discovered this and other errors and they made a conscious business decision to ignore the issues.

Is this specific issue a major problem? Not really. The tree will still be processed, reports and charts will not be affected. However, processing time will be longer as a result and, more importantly, the errors will be perpetuated as we generate GEDCOM.

What this issue does demonstrate, once again, is the lack of attention to detail that we repeated see in the product. This software is not free and many of us repeatedly purchase upgrades in the hopes that the new release will be an improvement over previous releases.

I continue to be disappointed and frustrated.