Marriage merger

That's odd ... many times I have seen where merging does suggest or provide the option to merge the couple at the same time. I don't know why the behavior would be different - perhaps there was a duplicate of one and not the other?

@Julia Szent-Györgyi Cool .. I guess I'm just the lucky one ... or maybe it's another feature (creating profiles and selecting the already created couple that I am thinking of - no reason to create a duplicate if there is a correct profile already). Anyway if i am recalling incorrectly - thanks.

I have also come across this. Merging two records for the husband often creates the suggestion to merge the two records for the wife (which you then have to go and do separately), but apparently it makes the suggestion only if her names and dates are similar/exact -- FT does not give you the opportunity to merge the two wife records (as does, say, my Legacy program at home, which goes into "Finding Surrounding Duplicates" mode, and also retrieves potentially-duplicate children! and all in the same pop-up window). It's surprisingly easy to end up after merging two husband records to have wife Sarah with children Thomas 1847- and Jane 1849- and wife Sally with children Ann 1853- and John 1855- and not realize that Sarah and Sally are also the same person.

I think that Brett has assumed that the OP wants a blanket "Merge All At Once" feature, which I'm pretty sure isn't what he meant! In Legacy, each "Potential Duplicate?" pair that comes up requests me to compare the data for the two records before I merge them (or not, as the case may be). I agree that in FT it would be handy to have a "Potential Duplicates?" window pop up!

I understand @Brett . 's concerns that "Many users have trouble with just one merge at a time". But I think it would be an improvement if there was some support in the system to "stage" multiple merges, execute them all at the same time after cross-checking tham all, and leave a link between them that shows in the log that can be followed both ways.

I've been doing some work of some lines in a cetain rural area in the 19th century, where it seems like close to 100% of the people were entered in AF (perhaps from controlled extraction) or at least IGI well before the Tree went live. However, each child is in there with their christening date and parent names, and the marriage record has their marriage date and parent names, but no other dates or connections, and most of the tree entries haven't been touched since 2012. So there might be tree entries:

• John Smith (XXX-YYY1) and Mary Doe (XXX-YYY2) married 1875, with two just-parent records connected to each
• April (XXX-YYZ3) christened 1876, daughter of John Smith (XXX-YYZ1) and Mary Doe (XXX-YYZ2)
• Beth (XXX-YYW3) christened 1878, daugher of John Smith (XXX-XXW1) and Mary Doe (XXX-YYW2)
• Charles (XXX-YYQ3) christened 1880, son of John Smith (XXX-YYQ1) and Mary Doe (XXX-YYQ2)

Each record has a suggested source from the IGI, which says "somewhere on this film", which is a batch with hundreds of images, but at least they're generally chronological by date. Then it'll have another suggested source from indexing, linked to a particular image. There might be civil birth-records for those children as well.

And when I go look at the sources, I see that not only were they recorded in the same parish but they all were from the same neighborhood, in a small town, with the same godparents. And the civil birth-records give consistent ages for the parents, and place of origin, and grandparent names.

So for now my workflow is to attach all the direct sources to each record, and then start merging, typically trying to merge into whichever record was touched earlier in 2012. So the family above would be three husband merges, followed by three wife merges (later followed by additional merges for the children's marriages).

But writing the reason for each merge is repetitive. I'll admit that I've written little more than "seems to be husband in same couple" or "seems to be wife in same couple" sometimes. It would probably be better to have a summary of the merge proposition for the whole family staged in an offline file, and copy and paste it into the reason statement for each merge.

And sometimes the case for a merge is less clear cut; maybe the wife was christened as Mary Jane, daughter of Robert Doe, but the adult records sometimes refer to her as Mary Jane (no surname), sometimes Mary Doe, and sometimes Jane Doe. Or the groom and bride reported ages of 21 and 18 when they got married, but their own seemingly-matching birth records would have made them 20 and 21, or 24 and 16. Or the civil birth-record for a child says her mother was 25, and the death-record a year later says her mother was 30. Or a child's birth record (reported by a man the mother was living with) says the child's father was unknown, but the death-record a year later gives the father as the same man who reported the birth. It would be nice to be able to cross-reference the same discussion of discrepancies like that , and their resolutions, from all the merges they're relevant to.

I'm not disagreeing with you, Brett! You do not want to see a "merge all at once" feature, that is clear! But again, I don't think that is what @Joshua David Liddicott is asking for.

If you merge two records for John Smith, each of whom has a wife, it would be convenient if the system would then pop up a window saying "Possible Duplicates: Mary Jones YYY-YYYY and Mary Jones ZZZZ-ZZ" and then lead again to the "Compare" page. (This is "staging" them, as @davidleelambert suggests.) A user would not have to accept the system's suggestion if it is clear (or if there is even a possibility) that the two Marys are different people.

I too have spent many hours separating records that have been incorrectly merged, and completely agree that a feature that says, "You just merged John Smith and John Smith, shall I merge the wives for you too?" without making you look at and compare the records for both wives, is not a good idea!

But @Dennis J Yancey also makes a good point ....

(There is already a feature similar to this in place, though I'm afraid I can't remember the exact situation where it comes up. If I remember correctly, it is in adding a child's record to one parent -- not at the usual place, on the father's page, though -- and the system pops up a box asking "Do you want to add to father John Smith, or to John Smith and Mary Jones?" apparently assuming that you are already sure that the mother is Mary Jones. "Add to Couple?" I believe is the phrase used.)