I’d like to ask for advice on how to reduce the amount of manual work involved in this process.
I have a main tree that I created and maintain in Gramps. I also have a GEDCOM file that I received from an acquaintance, which has no connection to Gramps. I need to compare these two trees and possibly supplement my tree with missing people from the external tree. What’s the most efficient way to do this? I understand that this needs to be done manually to avoid turning my tree into a mess, but both trees contain several thousand people, and manual synchronization could take weeks, if not months. I’m not interested in all the records from my acquaintance’s tree—just certain surnames. However, this still amounts to about 1,500 people out of 6,000 in his file.
Here’s what I’m currently doing:
I’ve set up a virtual OS. Both the virtual machine and the host have Gramps installed. My main tree is on the host, and on the virtual machine, I’ve opened my acquaintance’s GEDCOM file. On the virtual machine, I search for all the people by surname that I want to compare with my own database and add them as needed. Then, I search for them one by one in my database. If they’re not there, I create them. On the virtual machine, I tag the reviewed people as “added.”
Perhaps someone has ideas on how to partially optimize this work. For example, I’d greatly benefit from the ability to copy people from one program to another or something similar. I understand that this could lead to some inconveniences, like duplicate places, but I can merge them manually later.
I often use RootsMagic Essentials for this, because it can search duplicates fast, and works well in Windows and Wine, with some tweaks, so that I can run it next to Gramps, without a VM.
There is no automatic merge, unless you pay for RootsMagic, and 1,500 is a big number for such a task. The duplicate search is so much faster however that it helps a lot, especially when I import my own tree and the other one in a single RootsMagic DB. RootsMagic is also a bit smarter than Gramps, because it eliminates duplicate birth dates and places, etc., on every merge, so that there is much less need to clean up after merge.
Since I run only one copy of Gramps, I would run a report (tag report) that just gives me a list of names of those in the main tree and load it into a spreadsheet. This can be dismay right beside Gramps. In Gramps I would bring up the other tree and just walk down the list of people eye balling what is in the main list and the other tree. When I found a person I wanted to add, I would add a tag to that person. (you can add tags to multiple people at one time)
When done, I would create a gedcom based on the tagged persons and then import it to the main tree.
That is my manual process.
I think this is impossible for me, because I use places hierarchic. It will be lost if I will export gramps data to RootsMagic and then import it again. Right?
That’s right, but if the other tree is in a GEDCOM file now, there is no hierarchy in that either, so you will need to use the place update tool anyway, and I always use that after importing new branches from FamilySearch with RootsMagic.
The reason that I mention that program is that it makes it far easier to explore what I have, and find out where the duplicates are, between my own and that other tree, and because I don’t need a VM to run it next to Gramps. And once I know which persons are worth importing into my main tree, I often still import the GEDCOM into a new tree in Gramps, so that I can make a selective export, in Gramps format, that I later import and merge into my main tree. I do that, because Gramps has better filters, and gives me better ways to clean up the new data before import.
And in cases like these, I always try to focus on gateway persons, which are end-of-line persons in my own tree, and connect to lots of new ones in the other.
Merging hundreds of persons that already exist in both trees is hard work, so I can’t really recommend it.
I always use this, after import, and cleaning. And the latter is often needed when I download branches from FamilySearch, where verified places already exist in a format separated by commas, but where some are poluted, because people often add "Of, " in front of a place when a record says that someone’s “Of Amsterdam”, which means that he or she cam from Amsterdam. And in all such cases, I remove the "Of, " part before I generate the hierarchy.
Generating the hierarchy after import does create duplicates in the place tree, but they can be merged automatically by one of the Isotammi tools, which I also use, but can’t explain in one single message.
What I can show is how RootsMagic displays duplicates:
This is a screen that I created after merging a GEDCOM sent by a cousin into a RootsMagic tree that I often create as a copy of my main tree. And this screen pops up in less than a minute when I start the Merge Duplicates tool in RootsMagic Essentials, and gives me a quick insight into the persons that are worth merging into my main tree, or which can be ignored, because their vitals and relatives are already the same as what I have.
This screen comes from the free RootsMagic Essentials 9 running in Wine in my Linux Mint 21.3.
