The Wrong Combination from NFS

I got the following comment concerning the now imminent separation of New.FamilySearch.org (NFS) and FamilySearch.org's Family Tree (FSFT). Here is the comment from my dear friend, Anonymous:

The problem I would like to see addressed is that of instances where multiple individuals in nFS have been combined into one individual. If the summary is the only data transferred to FSFT, and if the listed ordinance date is the result of the multiple individuals being assessed together, how does one know that the ordinances work has been attributed to the correct individual? How does on know that the vital dates and places are attributed to the correct individual? How can one go about separating these combinations if they cannot even be seen in FSFT? I have not seen this topic addressed anywhere. I personally cannot get my head around how this could possibly be correct information. Do you know the developers' answer / explanation?

I probably have to back up a little ways to start to explain all this.  The answer to all of the above questions does not lie in either NFS or FSFT. The answer is also fairly difficult to understand unless you have worked extensively in both programs, but I will try to make it as plain and simple as possible.

First, the issue of combining individuals in NFS. I assume by this, the person is talking about combining individuals who are not all the same person, i.e. a mistaken combination. If the person is asking about the situation where multiple copies of the same person are combined then the answer is trivial. Starting with the second situation, the trivial one, the information that survives in FSFT can be edited to show the correct information. The only issue with the ordinance information is whether or not the ordinances have been done. The ordinance dates that end up showing are really irrelevant to any issues or concerns. The only real concern is whether the ordinance has been done or not and that fact is indicated in FSFT.

Now, to the issue of wrongly combined individuals in NFS. Here there are two main points. One, there is only one place in FSFT for each individual to exist. That is, only one individual can occupy a node in FSFT at a time and the information for that node is arbitrary to some extent. That means that FSFT does not determine the need for ordinances based on the information supplied, except to determine if sufficient information is available. Once the ordinance is completed, the information about the individual can be changed and it will not affect his or her ordinance status. Completed means completed. So let's suppose A and B are wrongly combined in NFS. A is the record that survives in FSFT. That means that the node or point on the family tree occupied by A is filled with information from both A and B. But since only one individual can occupy a node at a time, the extra information from B is either thrown away or corrected to reflect the correct information for A. This is where the information has to come from someone outside of the program, with the correct information in their own database. A is then correctly represented in the FSFT.

Now what about B? B's information is not lost altogether as assumed by the commentator, but rather the node or point on the FSFT is either empty or has some other information about B. So now, the person working with FSFT will find that empty node and fill it with information from their own database. I have actually seen this work where the FSFT had the wrong name for the person with missing information when the name was corrected, a duplicate entry was found with all the missing information. The question above presupposes that the combined nature of the entries from NFS will somehow be preserved in FSFT. This is a false assumption. Only one person can occupy a node or point on FSFT at a time. Changing the information or a lack of information does not change the structure of the tree.

Let me go through this again. Let's say A name is Frank Jones and he is incorrectly combined in NFS with Franklin Jones (another person). I am the descendant of Frank and so when I go into FSFT, I see some incorrect information about Frank and correct it. For me, that is the end of the story. The fallacy in the question above is that combining the two individuals will somehow ignore all of the individual's associations. In FSFT, the incorrect individual will show as an alternative husband or alternative name. So there would likely be one marriage record with Frank showing no ordinances and one with Franklin showing ordinances. It seems there is a slight possibility that the work for Frank will show as done when it is not. I am not aware if there is a failsafe provision in FSFT that takes care of this seemingly rare situation. But one of the two is going to come up with needing ordinances.

Now what happens to Franklin? His node or position on the FSFT is either empty or filled with a named individual. If he is the one with his work done, then the node may or may not show that work, depending on wether or not the combined information is present in more than one place.

Now what happens to the merged individuals? They can be unmerged at any time evidence shows that this is needed. None of the information in FSFT is ever really lost. This is really no different than a more simple mistake of recording the ordinance work done for the wrong person.

Whether or not the "ordinance date has been attributed to the correct individual" is not even an issue in FSFT. The issue is not whether the date is correct but whether the work has been done. If it has been done, then the date is immaterial from that point on. The date is only interesting for historical or record keeping purposes but serves no further function. If the ordinance work shows as needed in FSFT and there are no available duplicates then the work will be done, regardless if there is some hidden or missing combined record with some other individual in the tree. This appears to me to be a very slight chance of repeating the ordinance work compared to the huge amount of duplication allowed by NFS.

We don't really care if the dates etc. are "attributed to the correct individual." You, as a family member can go in and change any incorrect dates and any other incorrect information. But the key here is having a master copy (your own database) of the correct information. If you find incorrect information in FSFT, you can correct it. It would also be nice to let the rest of us know that the information is really correct by having a source for every change.

There is no need to separate the combinations. The FSFT program itself, by its structure, corrects the problem. A and B are two different nodes or point on the FSFT and can be dealt with individually. There no need to worry about the data coming from NFS if you have your own master copy of the "correct" information.

So now the topic is addressed. If this isn't clear enough, I will do a video explaining it in more detail. Let me know.