Problem - People merging duplicates merge the older PID# into the newer PID# and break everything.

Gordon Collett

Excellent set of instructions. You may want to re-post them as a new entry rather than having them buried at the end of this ancient thread which starts out as a discussion of problems with merging which have been fixed through two or three major updates to the site and hundreds of minor updates. Its title is not applicable at all anymore. With the current merging system is really does not matter which ID is preserved in the merge and a "wrong" order of merge does not break anything.

Regarding your concern with duplicates and all the ones added by "the system" in 2012, a bit of historical background is needed to understand this. FamilySearch, as the Genealogical Society of Utah, starting collecting genealogical information as submitted by patrons of its library in 1894 (see: https://en.wikipedia.org/wiki/FamilySearch) Through the years this information was collected and indexed in various formats including Family Group Sheets, the Temple Bureau Index (TBI), the International Genealogical Index (IGI), the Ancestral File (AF), the Pedigree Resource File (PRF), and the Vital Records Index (IGI). Also, starting in the 1970, they started the extraction program, now known as the indexing program, in which volunteers went line by line through original records and copied out vital information.

Some of these formats allowed for linking of families, some allowed just individuals with either parents or spouse but not both, some came from primary sources, some came from the complied research of patrons, some allowed sources to be included, some stripped out any sources when put into that particular database. Checking for duplicates was a bit haphazard at best.

In order to address this problem of multiple duplicates in multiple databases and to allow descendants to correct and properly source information about their families, much of this information was compiled into the New FamilySearch (NFS) database in 2009. That system worked pretty well but had some significant short comings.

In the spring of 2012, all the databases mentioned above were thrown together to create the Family Tree database we now have to work with. The system did not create any of those records you see dated April 2012, it merely copied them from the TBI, IGI, PRF, AF, VRI, and NFS databases and many of the extracted record databases which were at that point part of the IGI into one huge database, leaving it up to the people who know and are will to properly re-research all these families, in other words, us, to properly combine duplicates, fill in holes, and expand the database with all the people missing from it, all with sources to establish the correctness of what we do.