I have a large database of over 300,000 names accumulated over 30 years. I am currently comparing entries in my RM7 database to FamilySearch using RM10 since it can no longer be done in RM7. However, I would like to be able to see the new matches in my working RM7 database. If I auto match to FamilySearch in RM10, make a new gedcom in RM 10 and re-recreate my database in RM7, am I likely to lose important information. I’m assuming something will be lost, and should I take the chance of doing harm to my many years of research to be able to visually access the new FS matches. I mostly use basic features in RM7 that export without difficulty to online shared databases. For example I omit the extra place details since they do not transfer well. I have few custom facts and and have not made changes to sentence structures within RM7. Will going from RM10 back to RM7 change the way RM7 features work? One downside will be to have to wait the number of hours it will take to do the auto matching. The first time I matched the entire database took 24 hours.
I’d first emphasize that doing Backups and storing them in two different locations (or emailing yourself an additional copy) should be a way to restore back to any historical time in your database evolution.
Second, since the least hassle / safest method is to export that RM10 GEDCOM and then import to a newly-created RM7 database… there’s isolation from your master database.
RM7 features cannot change, so they’ll always act / produce as they always have. It sounds like you’re after FSID changes? File>Compare might turn those up for you to dragNdrop to your Master across and ontop of each individual (for merge) so a new Backup can be created.
Focus on selection of only folks with FSID’s for export to the GEDCOM for working on Compare back in that new RM7 database.
Thank you for the information. I was not as clear as I might have been. I want to be able to use the newly created RM7 database as my master and am hoping not to lose important information in the export/import process. What I am doing is trying to find birth dates for some 40,000 people of my 307,000 that have no birth dates from my matches on FamilySearch. I am not looking for changes on FamilySearch, just looking for possible new information to add to my database one person or family at a time.Some days I find from 50 to 100 to add. In a couple years time I have found about 10,000 of these. Thus, trying to compare the current master database and my replacement one is not a reasonable option for me at this time. Since I continue to do this editing on a daily basis, I have to keep the latest days work. I am paranoid about backups. I do this every few hours! and try to keep external ones on a flash drive.
Apologies, a full export from RM10 followed by import into a new master RM7 database sounds straightforward and certainly RM7 will still retain and show all data that its features provide for currently.
I gave you bad information. RootsMagic 10 FamilyTree matches do not get assigned a FamilySearch ID fact, so narrowing selection would be tedious to manually mark only those “matched” manually.
Just a thought @kbens0n --is there any way, that @mscheffler could make a group based on changed date
I cannot give an absolute answer, but I think there is a very high probability that under the circumstances you describe, you will not lose any important information.
There are known losses in transferring data from RM to RM using GEDCOM. An excellent description of these losses may be found at GEDCOM & DnD transfer losses It does not appear to me that any of the known data losses with GEDCOM transfer are likely to affect your process with respect to important data loss. But review those losses for yourself. Also, do be sure to specify the Extra Details (RM specific) option on your GEDCOM exports from RM10. And be sure the review the other GEDCOM export options very carefully. I’m sure you are already doing that.
If you are at all able to run SQLite, Tom Holden developed an SQLite script to revert an RM9 database to RM7. I believe it would still work with RM10. I tested it very thoroughly during my RM7 to RM9 conversion. I moved from RM9 to RM10 a couple of weeks after my RM7 to RM9 conversion. The timing of my RM9 to RM10 conversion was because that was when RM10 was announced. Otherwise, I would have gone straight from RM7 to RM10 if RM10 had been released a little sooner. But as you can tell from how long it took me to convert, I was a very slow and very reluctant convert to RM9/10.
Tom’s script doesn’t have any of the GEDCOM data losses, per se. But it obviously is not able to revert new RM features such as multiple color sets back to RM7. In the case of the reusable citations in RM8/9/10, Tom’s script reverts them back to RM7’s copied citations. And in fact, GEDCOM itself does not support reusable citations, so RM10’s GEDCOM export reverts them back to RM7’s copied citations.
I totally get why some RM7 users are reluctant to convert to RM10. I think the RM7 user interface is much superior. I used to complain about how clicky it was, but I think the RM10 user interface is much clickier and much slower to use than was the RM7 user interface. But I eventually converted anyway. The reasons were that RM7 no longer supported the FamilySearch and Ancestry interfaces, and that RM9 and especially RM10 have many great new features. I never had any interest in converting to RM8.
I would take the approach of finding what you want in v10 from FS.
Then since you already have a duplicate database of 10 in 7, find the corresponding person in 7 and just update them.
On Monday update v10, Tuesday copy the specific data to 7. Wed update v10, etc OR whatever you want.
This way you don’t have to worry about anything being lost doing gedcom after a bunch of changes every time.
If it takes 24hrs to do the gedcom, it would be faster my way. Good luck.
The gedcom files are created quickly. However, it takes many hours for RM10 to automatch to FamilySearch for the 307,000 name database.
That is on my newest laptop with 16 gig of RAM. So far it has auto matched 9500 names in 7 hours. 220,000 names were previously matched in my master database before RM7 could no longer get the information from FamilySearch. Judging by the status bar of the progress I am guessing it will take another 7 or 8 hours.
When automatch is complete and I can make a new master RM7 database from the RM10 export with the updated matches, I can get back to editing and looking for missing dates. I don’t keep the two databases in sync which would take too much time as data entry in RM10 is exceedingly slow. RM7 is such an easy program for data entry, has much easier to read screens, meets my needs, and runs so much faster, that if it were not for having lost the FS feature, I would ignore version 10.
Maybe I am missing the obvious, but I see no other way to get the info from FamilySearch to RM7. For finding new information person by person I look up the RIN from RM7, compare the same number to FamilySearch on RM10, and type what new information I want from FS (in my preferred format) that I want back to my master R7 database.
The GEDCOM will contain the FSID so I’m not sure why you are needing to match the FSID in RM10. Are you selecting the option “Extra details (RM Specific)”, though I don’t even think that necessary to preserve them. You should only need to open FamilySearch Central in RM10 to have it update the ordinance statuses.
The first color-code set in RM10 will be preserved going between RM7. You will lose your groups, and Tasks (research logs, and To-Do items).
I don’t understand. I do NOT want the FSID and have never put FSID numbers in my database. I certainly am not interested in ordinances
All I want to do is look at the matches between specific individuals where I am lacking specifics and their supposed matches to see if there are birthdates, death dates, and/or marriage dates that seem reasonable. This worked really well in RM7 before changes were made so we could not access the FamilySearch information within RM7.
Now I have to keep a duplicate of my RM7 database in RM10 to see the matches. If there is information in the match in the FM10 database that I want to use, I then go to my RM7 database and type the particular date or dates in. I enter by hand any information that I take from shared databases of any sort.
It makes for a lot of work, but I see no other way to see the matches. There are so many errors in the FamilySearch entries, I would never want to download batches of information. However, if I find an approximate date for an individual where I am lacking one, I may add it if the information appears carefully entered.
The FSID is what matches a RM person to a FS person. You can’t have one without the other, even if all you want is hints. I guess I don’t understand what you are finding missing and causing you to run AutoMatch then. The GEDCOM will absolutely include them.
I’m wondering if mscheffler has been opening a completely unmatched (over 300,000) person RM7 database for conversion by RM10 and then running FamilySearch AutoMatch against all of them, each time… instead of just newly-introduced or newly-updated (with more discerning info for FS matching) much lesser number of unmatched additional folks?
I must admit I am confused. I imported a gedcom from my master Rm7 database that I have been working on for many years into RM10. The RIN numbers are the same in each database. Then I started the automatch utility on the RM10 database so I can bring up my view of the family and compare to the FS family match side by side. The process does not work if the automatch has not been run. I do not do anything to the RM10 database other than look at it.
If the match has say a birthdate I do not have I go to my RM7 database and add it there. I search the RM10 database using the RIN number that is in the pedigree view on RM7.
I am currently waiting for FS automatch on RM10 database to complete. It has been running 12 hours so far. I then plan to make a gedcom from this RM10 database and put it in a new database on RM7. That will become my NEW RM7 master database. I want it to be exactly the same as the current one, except the FamilySearch icons will show up in pedigree view. That helps me in deciding which families to go back and compare a second time for discrepancies.
My primary concern is that RM10 will not have changed my database other than to add the FS symbol to the far right of the name in pedigree view.
I was originally asking if when I do this, I will lose anything from my database. I want the two RMdatabases to be exactly the same except the new one will show the match symbols for Family Search.
I neglected to mention the RM10 database and my working RM7 database are on two different computers.
In re-reading thru this I think this is what you are doing …
1 - Gedcom from 7 to 10 (personally I would have Imported from 7)
2 - In 10 run FS Automatch.
3 - Look at the results in 10 (side by side) and then add missing info to v7. (I would have both version running on just 1 computer not 2)
4 - Gedcom FROM 10 to NEW 7 database. (since you did not add info to 10 it is not in the new 7)
I would enter the dates into 10 (even if you don’t like it) and then gedcom that to 7.
It sounds like you are repeatedly running the Automatch. How often do you do that?
Selecting based on changed date is the optimum approach for import into RM10, whether via RM7 GEDCOM import into RM10 -or- by dragNdrop to a clean new RM7 database for direct conversion by RM10. And the FamilySearch AutoMatch operation would be much quicker.
I have just reread the entire thread. I’m not sure I understand your process 100%, but I think I’m pretty close to understanding. Let’s see how close I am to understanding.
- You export a GEDCOM from your master RM7 database and import it into a new and empty RM10 database. At that point, your master RM7 database and your new RM10 database are as close to identical as possible in most every way that matters. Even so, I agree with the suggestion to do a direct import from RM7 into RM10 without using GEDCOM. Even though your RM7 and your RM10 are on different computers, you still could copy your *.rmgc file from your RM7 computer to the RM10 computer, the same way you now copy your *.ged file from your RM7 computer to your RM10 computer. Then RM10 could import the *.rmgc file.
- You do an automatch of your newly created RM10 database. Other than that, you don’t type anything new into RM10. The practical effect is that people in in the RM10 database who were not already matched are now matched. In database terms, what that means is that an FSID has been added to the previously unmatched people. I don’t use automatch, so I don’t know how it deals with people who are already matched and whose FSID has changed at FamilySearch. For that matter, I don’t know how it deals with people who are already matched in general. It surely would run a lot faster if it only tried people who were not already matched. It may already be doing that, but I don’t know for sure.
- You export a GEDCOM from your RM10 database and import it into a new and empty RM7 database.
- You find people in your new RM7 database who are newly matched, who don’t have a birth date, and who have a birth date in FamilySearch. I confess I still don’t understand how you do that. But you seem to have a method that works well for you.
- You type the birth information from FamilySearch into your new RM7 database.
- Your new RM7 database becomes your new master database.
Is that about it?
To “thejerrybryan” You got it! Generally I do add more than the birth date from the match if what is on the match seems looks credible. The automerge of my new RM10 database has now been running some 18 hours. Looking at the progress bar, I expect it to finish within the next hour or so. If my new database on RM7 looks good on import, I do not expect to repeat the process anytime soon! There is some information on the Automatch feature on pp. 313-4 of the RM10 book.
For those of you not familiar with automerge, I occasionally use it to share new information I have uncovered or to correct some really obvious errors on the Family Search Family Tree. I usually only send corrections if I have vital records sources so as not to introduce any new errors.
Surely a typo, intending automatch?
I can’t really think of anything to help your situation. The GEDCOM back and forth between RM7 and RM10 is not really the issue. Even if I were doing this myself and even if I used SQLite to get the data back and forth between RM7 and RM10, the excessive time associated with the automatch seems to be the problem, not getting the data back and forth. I don’t see any way that SQLite could help with the automatch.
I personally much prefer the manual matching process. I do not have the experience with automatch to know if it’s really faster than manual match or not, given that I would always want to check matches after the fact anyway and type my data in by hand anyway. So I only manual match and I only manual match individuals of immediate interest. But for an RM7 user, doing the manual matching forces you actually to become an RM10 user, at least to a certain extent. After RM7 would no longer connect to FamilySearch, I had to quit matching altogether until I upgraded to RM9 and then shortly thereafter to RM10. I am now back to manual matching using RM10.
Was the automatch this slow when you were still able to do it using RM7?
I imported my RM10 gedcom and now have a new automatched RM7 database up and running. I don’t remember how many years back I had done an automatch on RM7. I usually just manually matched as I worked on individual families one at a time. I’m sure the automatch was quicker, however.
Everything I do with RM10 takes time. RM7 is quicker and I have a long time use of adding data so it is done more easily. Also the screen views are easier on older eyes.
I don’t think I will repeat the process I did today anytime soon. I had created an RM10 database with an import right after FamilySearch quit working with RM7. The RM10 database becomes somewhat obsolete over time if one adds too many names to RM7. I had done a lot of adding new information and editing over the past few months, thus making me want to update to have everything up to date.