I just noticed I have a slew of duplicate people in RM 11. I had dealt with this in prior versions, using auto merge find the people for me, and manually merging the few people I found on my own, but my database was pretty clean as of the last use of RM 10. I imported the RM 10 database into RM 11 and I’ve been using RM 11 since shortly after its release. Yesterday discovered I have a ton of duplicate people in my database that now need to be merged. Has anybody else experienced this? I’m also seeing duplication of birthdates in individuals (i.e. the birth date appears as two or three separate entries, some with all the information and some with only the year.) I ran the database tools and got no error messages.
IF you made a copy of your RM10 database (.RMTREE) file and moved it to a separate folder for RM11… opening that copy from a new installation of RM11 would convert it WITH NO DUPLICATION.
Typically, what you describe is the result of Manual Merge operations done after GEDCOM import or dragNdrop between populated databases. Same with RM10.
It may only be a matter of terminology, but I would assume you converted your RM10 database to RM11 rather than imported your RM10 database into RM11. Converting from RM10 to RM11 will not generate any duplicate people. It will keep any duplicate people that were already there in RM10, but it will not create any new duplicate people during the conversion process.
Assuming you did convert from RM10 to RM11 rather than importing from RM10 to RM11, the symptoms you describe suggests that some sort of import has taken place in RM11 since the conversion. The import could have been a GEDCOM import, a drag and drop, an import from FamilySearch using FamilySearch Central, or some sort of download from Ancestry using TreeShare.
So I would ask: what was your process of getting from R10 to RM11, and have you done any sort of import since them?
As kbens0n already mentioned, if your original R10 database is still there, you could copy it and convert it again to see what the status of your duplicate people are. If it’s still there and if you convert it in place without first copying it, it will no longer be an RM10 database after you convert it.
If memory serves, I just pointed RM11 to the v10 rmtree file and opened it. I don’t recall using any import function and definitely have not imported any Gedcoms. I unfortunately didn’t save the original RM10 file although there might be a backup of the RM10 file on the date I purchased RM11. I’ve been around since RM7. Although I’ve occasionally found a few duplicates in my database I’ve never had them to this extent.
I don’t see how you get duplicates when opening the same database each time in RM.
Hope you have the RIN displayed.
A screenshot may help to see what you see.
In that case, this is a great mystery. When you opened your RM10 database with RM11, RM11 converted the database to RM11 and it became unusable with RM10. What you might call normal use of RM does not produce duplicates. The duplicates would have to be imported, or I suppose the duplicates could even have been typed in again. But that seems unlikely.
If you have your last RM10 backup, that could be a useful diagnostic tool. But be careful with a restore. The file name will be the same as the database that is now RM11. If you do a restore using RM10 and if you restore into the same folder, you will be writing on top what is now your RM11 database. As a result, you will lose everything you have done since you installed RM11. You can avoid the problem by restoring to a different folder, but that can be trickier than it sounds.
For those kinds of reasons, you may well be better off simply continuing to work through the duplicates in RM11. I assume you are merging the duplicates. But I don’t understand why the duplicates keep re-appearing. As I already said, normal use of RM doesn’t create duplicates.
Here’s what I’m seeing. The people are completely duplicated and have separate RIN’s
In the interim, I let RM do an auto-merge and it got rid of most of them. Whatever happened must have been idiosyncratic to me since nobody else is reporting seeing this. The auto-merge seems to have fixed the duplication of people although now I have to delete a whole bunch of duplicate facts for people. For what it’s worth, rather than learning how to use sources, I keep my source information for each fact in the notes for each event. The facts for all the duplicate people have blank notes but they do have all of the same facts as the original. I have no idea if that’s some kind of clue to those of you who really understand the program. To me it would sound like a GEDcom of the database was reimported and that Gedcom file had been made without including the notes but I don’t recall ever importing a Gedcom into my main database.
If it wasn’t a GEDCOM import it could have been a FamilySearch import. The FamilySearch import doesn’t include notes or sources.
I do use FamilySearch to add selected pieces of data on an individual one at a time and I’ve seen that type of duplication but this is birth, christening, marriage, death and burial on almost everybody. Is there something else when using RootsMagic 11 with FamilySearch that could have caused this? If so I really want to avoid that. Over time I’m going to have to fix almost 10,000 people.
Can you show a screenshot of what is duplicated? Might be a clue there
I don’t think there are any other means of duplication.
When it comes to merge, there are a few particulars. AutoMerge only works for EXACT matches w/no differences (all the same facts must be present and filled in with matching typed character data, ETC.) and DOES NOT produce any duplicate events/facts. Manual merge for EXACT matches also DOES NOT produce any duplicate events/facts.
So, when You say “going to have to fix almost 10,000 people”, it makes me think You should examine these duplicates for Date Last Edited to check if dates were two different time periods for any particular pair of duplicates. You might should fire up RM10 and Restore your most recent Backup of this database and check for the possibility of this duplication having occurred prior.
Duplicate facts will happen in merge if the places have duplicates in the place list. Even if the place name is identical you have to merge the duplicate places so its not considered a different place and keeping both facts.
Looking at how high the RINs (record indentification numbers) are in the screenshot, it suggests an import of some sort happened in the file. RM on its own can’t duplicate or even merge people. All of that is user driven to initiate. Sort the People List view by Rec# and look at those with the higher numbers. Then check the Date Edited on them. If the dates are all over the place then it was a GEDCOM import. (That’s assuming the GEDCOM is from another RM file.) If the dates are all the same then it was an import from FamilySearch.
Below is a screen shot of what I’m seeing. As I mentioned in another post, I keep my sources in the notes section. This birth entry has the source in the notes; the birth entry below it doesn’t. He was married twice. In the second marriage there’s one entry with just the year present and no note; the one below it has the full date and a note. I had a bunch of people with just a year for an event and then the full date in a 2nd and note or in some cases a 3rd entry for the event below it. I can explain the duplication of events because I let the program do an automatic merge when I discovered how many people were duplicated, and that got rid of most of the duplicate people, but I have no idea why the people were duplicated. I only noticed this since updating to RM11 and having opened the RM10 database with RM11. Since I’d never had a problem with an update before, I didn’t make a copy of the RM10 database although there are RM10 backups available, which I always do on shutting down the program. Since nobody else has seen this, the duplication of people must be idiosyncratic to something I did. I didn’t import any GED files in recent memory although I’ve generated a small branch GED for a cousin recently. I did not however import that into this database.
“AutoMerge only works for EXACT matches w/no differences (all the same facts must be present and filled in with matching typed character data, ETC.) and DOES NOT produce any duplicate events/facts. Manual merge for EXACT matches also DOES NOT produce any duplicate events/facts.”
Auto merge didn’t behave that way for me. It merged people and duplicated their facts:
“So, when You say “going to have to fix almost 10,000 people”, it makes me think You should examine these duplicates for Date Last Edited to check if dates were two different time periods for any particular pair of duplicates. You might should fire up RM10 and Restore your most recent Backup of this database and check for the possibility of this duplication having occurred prior.”
Since everybody’s already automerged I can’t easily go back and see if duplicate people had different last edited dates, but as you can see in the image above after merging two identical people, it still says the last time I was in this person was March 2024. I had never noticed that date line until you asked about it.
I’m running out of energy tonight to reinstall RM10, but I do have a backup file from Sept 15, the day before I got RM11.
Those weren’t EXACT matches to begin with (essentially two different people because of the Notes mismatch You mention above) and RZamora1 explained that.
“Sort the People List view by Rec# and look at those with the higher numbers. Then check the Date Edited on them. If the dates are all over the place then it was a GEDCOM import. (That’s assuming the GEDCOM is from another RM file.) If the dates are all the same then it was an import from FamilySearch.”
The dates occur in sequence. I can see a progression of my work based on the Rec#. There will be several new people all having one date and then it progresses by a day or a few days and there’s another small number of people who were added.
I’m looking at parts of RM I’d never used before.
I think I figured out what happened and it’s something I did wrong. Lying in bed this morning I remembered that I had made a Gedcom file of the entire database to put on a thumb drive for my son in September shortly after upgrading to RM11. I would normally test the gedcom file before giving it, and probably used “Import” instead of “Create a New Roots Magic File”. I wish I’d remembered that before posting here.
Makes perfect sense. Thanks.