Jump to content


Photo

Importing GEDCOMS from Ancestry.com

gedcom ancestry.com sources

  • Please log in to reply
23 replies to this topic

#1 RassilonRegenerated

RassilonRegenerated

    Member

  • Members
  • PipPip
  • 7 posts

Posted 05 July 2014 - 12:19 PM

Greetings. I am new to roots magic and so far I love it but I am having a bit of frustration dealing with syncronization with my Ancestry.com tree. Basically Ancestry exports a GEDCOM and then I import it to RM6 and that is suppossed to be simple.  But the reality is there are two problems and I wanted to air them here as many may be having the same issues but may not even realize they are occurring:

 

1. It is silently dropping data from the import and loosing valuable info. Some of the blame for this is clearly on the part of Ancestry.com for having invalid GEDCOM files but part really has to be laid on RM6 for not raising alerts. I have just finished a difficult merger of two systems and cleaned everything up an now face having to redo it again as there was lots of important data that was dropped without warning.

 

2. I does not import and preserve the _APID tag from Ancestry.com. This is a custom tag from ancestry that they use to identify what sources are attached to a person. This is vitally important if you dont want to get a ton of "hints" that you have already reviewed and end up attaching duplicate sources to your people.

 

So explaining this I am going to go into some of the details of GEDCOM. Basically GEDCOM is a text file with one item of data per line. Adding a line break kind of srews things up, and that is all the fault of Ancestry for adding extra line breaks, but really I would have expected RM6 to at a minimum return an error saying "There were errors on import, check the log file!" As it is I had to find out about the log file from web searches. It is located in the same directory your new RM database is located with the name of yourgedcomfile.LST. Any time you import a GEDCOM look through this file to see what data you are loosing because RM is not going to warn you!

 

The problem I run into is that particularly sources have line breaks in the data itself. For example here is a snippet from my GEDCOM, particularly a source added by Ancestry.com

 

2 SOUR @S-396213722@
3 PAGE "About the Western States Marriage Record Index." About the Western States Marriage Record Index. N.p., n.d. Web. 14 Mar. 2014..
3 NOTE http://abish.byui.ed...recordID=529095
3 DATA
4 TEXT Details For Marriage ID#529095
Groom Last Name: STUCKI
Groom First Name: Golden Autumn (21)
Groom Residence: Salt Lake City, Salt Lake, Utah
Bride Last Name: SALMON
Bride First Name: Mary Edith E
5 CONC vans (19)
Bride Residence: Salt Lake City, Salt Lake, Utah
Place: Salt Lake City, Salt Lake, Utah
Date: 04 Oct 1933
County of Record: Tooele
State: Utah
Volume: C
Page: 213
Notes:
Salt Lake T
5 CONC emple


Now that is clearly broken. Everyline shoud have started with a number followed by a tag. The extra line breaks are going to break things. and they do. Here is from RM6 import log:

 

Unknown info (line 3151)
    Groom Last Name: SALMON

Unknown info (line 3152)
    Groom First Name: John Homer (33)

Unknown info (line 3153)
    Groom Residence: Salt Lake City, Salt Lake, Utah

Unknown info (line 3154)
    Bride Last Name: EVANS

Unknown info (line 3155)
    Bride First Name: Margaret Phoeb
    5 CONC e White (28)

Unknown info (line 3157)
    Bride Residence: Salt Lake City, Salt Lake, Utah

 

etc...

 

And the problem doesn't always stop there. The broken source throws off the parsing genreally and often you also loose the following fact/source before it finds its place again. In case you are now realizing all the cool data you are loosing from Ancestry on import I offer below how I believe the file should be imported and I plan to write a perl script to fix it that I will post here if anyone is interested. End result should look like:

 

 

2 SOUR @S-396213722@
3 PAGE "About the Western States Marriage Record Index." About the Western States Marriage Record Index. N.p., n.d. Web. 14 Mar. 2014..
3 NOTE http://abish.byui.ed...recordID=529095
3 DATA
4 TEXT Details For Marriage ID#529095
5 CONT Groom Last Name: STUCKI
5 CONT Groom First Name: Golden Autumn (21)
5 CONT Groom Residence: Salt Lake City, Salt Lake, Utah
5 CONT Bride Last Name: SALMON
5 CONT Bride First Name: Mary Edith E
5 CONC vans (19)
5 CONT Bride Residence: Salt Lake City, Salt Lake, Utah
5 CONT Place: Salt Lake City, Salt Lake, Utah
5 CONT Date: 04 Oct 1933
5 CONT County of Record: Tooele
5 CONT State: Utah
5 CONT Volume: C
5 CONT Page: 213
5 CONT Notes:
5 CONT Salt Lake T
5 CONC emple

 

So that is the first issue. RM team please consider warning users when there are errors on import and notify them where to find the file.

 

 

 

Second issue is the _APID tag. This tag does not seem to be documented publically but you can find a bit about it here:

http://ancestryforum...osts/46c21606a9

 

an example of how it looks in a GEDCOM is like a line like this:

 

3 _APID 1,7857::52344

 

I do not know exactly what the numbers mean. but this event attached to you various sources is what Ancestry uses to keep track of what sources you have reviewed and attached to your people. When you consider that you cannot update a tree through import but only upload changes to a new tree, this becomes vitally important to actually find new data to review and avoid having it present again the sources you already have attached to your people. Unfortunately RM6 silently discards these tags as well. From the import log you have possibly thousands of these:

 

Unknown info (line 3829)
    3 _APID 1,7884::168068278

 

Now I understand the RM does not have any function to make use of this tag but I am hoping there is some function hidden somewhere that will allow it actually custom map it somewhere so that when RM6 exports back to a GEDCOM it can be preserved properly to allow better functioning with Ancestry.com. I would be happy to talk with and experiment with any RM6 developers on this issue. Since I don't see that heppening soon. I am noodling on some ideas to capture the _APID tags inside another tag as text somehow then form yet another script to massage it back to the correct form for Ancestry.com. I may post back here with details.

 

So after this very long first post here I would ask, Has anyone dealt with these issues already and worked out a solution? Can anyone at RM respond specifically with any ideas around importing and preserving the _APID tags?

 

Thanks for reading

 

 

 

 



#2 TomH

TomH

    Advanced Member

  • Members
  • PipPipPip
  • 6254 posts

Posted 05 July 2014 - 01:27 PM

A well-documented report, RR. But it's not the whole picture in trying to sync with ancestry.com. There have been quite a few prior discussions in the Forums but I do not recall these particular issues. Some of my own limited experience can be found at http://sqlitetoolsfo...ag=ancestry.com . There are so many issues that it has become the rote answer from RootsMagic that we users should lobby Ancestry.com to issue an API for third party developers to interface with.

I'm intrigued by your research into the _APID tag and would be pleasantly surprised if there were a way to round-trip it back to an Ancestry tree and preclude re-hints.

Tom user of RM7550 FTM2017 Ancestry.ca FamilySearch.org FindMyPast.com
SQLite_Tools_For_Roots_Magic_in_PR_Celti wiki, exploiting the database in special ways >>> RMtrix-tiny.png app, a bundle of RootsMagic utilities.


#3 Laura

Laura

    Advanced Member

  • Members
  • PipPipPip
  • 4276 posts

Posted 05 July 2014 - 03:17 PM

Just how do you want RM to warn you that it created a list file on gedcom import, perhaps, on a popup screen which appears after a gedcom import is finished?   I would not like to need to close a popup screen if there was no list file created.

 

The Importing a gedcom file article, item 6, in Help. does tell us FM may create a list file and where to find it.

 

I wonder how far Bruce may be able to go to adjust gedcoms made by Ancestry or other third party genealogy progrsms when importing into RM without running into copyright issues?



#4 TomH

TomH

    Advanced Member

  • Members
  • PipPipPip
  • 6254 posts

Posted 05 July 2014 - 06:47 PM

Some prior work I did on roundtripping between Ancestry.com and Rootsmagic is described at http://forums.rootsm...th-ancestrycom/

Tom user of RM7550 FTM2017 Ancestry.ca FamilySearch.org FindMyPast.com
SQLite_Tools_For_Roots_Magic_in_PR_Celti wiki, exploiting the database in special ways >>> RMtrix-tiny.png app, a bundle of RootsMagic utilities.


#5 zhangrau

zhangrau

    Advanced Member

  • Members
  • PipPipPip
  • 1518 posts

Posted 05 July 2014 - 06:53 PM

I don't have my data stored on Ancestry, but some of my corresponding researchers do, so this will become an area for me to get better at . . . .

 

I gather that having a copy of FTM helps with the GEDCOM issues, by introducing another step to the sequence:

Ancestry -> FTM - > GEDCOM -> RM

and then

RM -> GEDCOM -> FTM -> Ancestry

 

I don't have FTM currently installed on my laptop, and I don't recall how old my registered copy is - I suspect I'll need to upgrade in order to get "best" results.



#6 TomH

TomH

    Advanced Member

  • Members
  • PipPipPip
  • 6254 posts

Posted 05 July 2014 - 07:50 PM

I have not experimented with RM -> GEDCOM -> FTM -> Ancestry and am unaware of any others so I cannot say whether it might be better. The reverse direction has the major advantage of bulk transfer of source images to RM.

Tom user of RM7550 FTM2017 Ancestry.ca FamilySearch.org FindMyPast.com
SQLite_Tools_For_Roots_Magic_in_PR_Celti wiki, exploiting the database in special ways >>> RMtrix-tiny.png app, a bundle of RootsMagic utilities.


#7 RassilonRegenerated

RassilonRegenerated

    Member

  • Members
  • PipPip
  • 7 posts

Posted 05 July 2014 - 08:00 PM

@TomH

Thanks for your awesome web resources. I actually very closely read the posts from your first reply before I composed my post and it gave me several hints. I have not read the reply from your second but skimming it, it looks interesting and I will tackle it when I have some more brain cells.

 

@Laura

I Dont think an notification telling you there where errors and where to find the report is at all unusual expectation for a program.

 

@zhangrau

Your mention of RM -> GEDCOM -> FTM -> Ancestry is what I was thinking as well. I must have interaction with FamilySearch with is not there yet in Ancestry (but they are working on it) and I really would like to use ancestry, and I pretty much insist on having a well working desktop software for geneology. So its a tough set of demands. Ultimately I am unwilling to spend more money for something that _might_ work and what I have read is that setup adds a step and really does not solve the problems.

 

@Everyone

I am pleased to report I stepped away from the BBQ today long enough to try out some of the theories I mentioned above and I am pleased to report they worked great! I was very happy. I can now import a GEDCOM from Ancestry.com with no errors and all data coming in. I can modify this file in RM6 and then export it as a GEDCOM and return it to Ancestry.com (as a new tree) and have it recognise all attached sources (no re-hints). Modify it some more there then bring it all back and merge with RM again. (OK merging back is still a bit of a chore and as always keep an eye out for duplicates including duplicate sources. Read below for the directions.

 

OK so this involves complicated changes to a text file and while not tough for anyone used to programming it can be a little intimidating. In trying to make this simple I chosen not to use any programing language or scripts although that might be very easy for someone who is familiar with Perl or Python. Instead however I use a text editor called Vim. It is based on an older program called vi and I will call it vi going forward because everyone does. This is a very powerful text editor (as will be demonstrated) and available as both a command line program or a graphical program (Gvim) for Linux, OS X, and even Windows. The graphical version should be fine for this task and I will give the exact commands. It is a free program.

 

Step 1. Get rid of unexpected line breaks that are causing error imports. As explained above this very much breaks the GEDCOM standard.

 

To do this, open your GEDCOM file you exported from Ancestry.com in vi then execute this exact command (just type it in after opening the file or even cut and paste it:

 

:%s/^\(\d [A-Z_]\{3\}\)\@!\&0\@!/5 CONT /g

 

Thats a pretty cute command isn't it! Basically it is doing a very complex search and replace by looking at the beginning of each line of your GEDCOM file and finding any line that does NOT begin with a digit followed by a space, followed by either at least three CAPITAL LETTERS or an underscore (_) unless the line starts with a 0 which is a bit different. Basically it is looking for any new line that is not in GEDCOM format. There is a slim chance it will miss a line it should catch, if it does you will either need to imporve my search or edit that line by hand. It's a fairly slim chance. It is possible it may tell you no matches found. If so rejoice you don't have this issue with your GEDCOM file. When it finds a line that matched this criteria it adds "5 CONT " to the beginning of the line. This is where I lucked out. I was afraid I was going to have to use a programing language to determine what digit (specifies depth in GEDCOM) was needed at the start of the line. But looking through hundreds of errors it is always a 5. I have a reasonably complex file I was working with and I hope this will continue to be true. I am like 80% confident on it right now. So adding this test on these lines corrects it to being a valid GEDCOM. Importing it to RM6 now has no errors except the _APID errors. All the data comes in perfectly. From the test case ancestor I was using I had 6 sources attached and in the broken format I was successfully getting 1 source with partial data. Now I got all 6 with detailed comments containing all the import text from the example above. I was thrilled to get this far and if you use Ancestry you probably want to at least try this much.

 

Now to the second problem, the _APID tags and getting Ancestry to preserve which sources are attached....

 

Step 2 Still got that file open in vi? If not reopen it. execute this command:

 

:%s/^3 _APID/3 NOTE Ancestry Tag: 3 _APID/g

 

This is a much simplier search and replace. It looks for "3 _APID" at the beginning of each line. When it finds it, it replaces it with "3 NOTE Ancestry Tag: 3 _APID". What this is doing is taking this Ancestry.com specific tag and encapsulating it inside a detailed note on a source. The end result is you will have sources in your program that have a details note saying something like "Ancestry Tag: 3 _APID 1,1788::77473660". You need these if you want to sync sources with Ancestry so dont delete them. The rest of the source citation should look pretty standard.

 

Now that you have made both these changes to your file save it and close vi.

 

Step 3 Import. As always its best to import to a new file then move over the data you need with the awesome drag and drop feature.

 

Admire the beauty of what you have created. check the LST file and find there are no errors. None! Check some sources and find your _APID notes (Maybe use find everywhere)

 

Make some edits and export to GEDCOM for return to Ancestry.com.

 

Step 4 Preparing for return to Ancestry

Looking at the GEDCOM now you will find your preserved _ADIP tags on lines like this:

 

3 NOTE Ancestry Tag: 3 _APID 1,1788::77473660

 

In a minority of cases you may have other detailed notes on the same source and it turns out like this in export:

 

4 CONT Ancestry Tag: 3 _APID 1,1788::77473660

 

Before returning it to Ancestry we need to deal with both these cases. I did not find any other permutations but obviously I have not been experimenting long.

So open your GEDCOM in vi again and run these two commands:

 

:%s/3 NOTE Ancestry Tag: //g

:%s/4 CONT Ancestry Tag: //g

 

Save and close.

 

Step 5 Upload. Upon returning your file (to new tree) in Ancestry.com you will see all your previously attached sources in all their glory, no rehints. (OK you may get rehints about other people's trees but no rehints about historical records).

 

Anyway I hope that is helpful. Feedback is appreciated.



#8 TomH

TomH

    Advanced Member

  • Members
  • PipPipPip
  • 6254 posts

Posted 05 July 2014 - 09:34 PM

Wow! RR, your handling of the _APID is a huge step forward in syncing with AFT. Well done!

And thanks for the tip on vim - I cannot recall using it. My tool of choice for years for regular expression search and replace was Ecobyte ReplaceText, formerly BK ReplaceEm.

Tom user of RM7550 FTM2017 Ancestry.ca FamilySearch.org FindMyPast.com
SQLite_Tools_For_Roots_Magic_in_PR_Celti wiki, exploiting the database in special ways >>> RMtrix-tiny.png app, a bundle of RootsMagic utilities.


#9 RassilonRegenerated

RassilonRegenerated

    Member

  • Members
  • PipPip
  • 7 posts

Posted 05 July 2014 - 10:06 PM

That's a new one to me but looking at it I am sure it would work as well with a similar regex.
I rarely use windows for anything however. I run RM6 on linux under wine currently.

#10 Laura

Laura

    Advanced Member

  • Members
  • PipPipPip
  • 4276 posts

Posted 05 July 2014 - 10:43 PM

I don't have any problem with RM telling us it created a list file on gedcom import.

I just don't want it to tell us it did not create a list file.

I am still interested in what form and where you would want RM to warn you that a list file was created.

#11 RassilonRegenerated

RassilonRegenerated

    Member

  • Members
  • PipPip
  • 7 posts

Posted 05 July 2014 - 10:58 PM

I think a popup with a dont show this message again check box is the pretty stand industry solution.



#12 TomH

TomH

    Advanced Member

  • Members
  • PipPipPip
  • 6254 posts

Posted 06 July 2014 - 06:35 AM

RM generates a LST file for all GEDCOM imports whether or not it has a problem with it.

Tom user of RM7550 FTM2017 Ancestry.ca FamilySearch.org FindMyPast.com
SQLite_Tools_For_Roots_Magic_in_PR_Celti wiki, exploiting the database in special ways >>> RMtrix-tiny.png app, a bundle of RootsMagic utilities.


#13 TomH

TomH

    Advanced Member

  • Members
  • PipPipPip
  • 6254 posts

Posted 06 July 2014 - 10:14 AM

RR, if you are planning to produce reports from RM with Source or Citation Notes, you might want to use privacy brackets:

 

3 _APID 1,1788::77473660

becomes

3 NOTE {Ancestry Tag: 3 _APID 1,1788::77473660}

 

When you export from RM, you will have to export private notes in which case all other private notes you may have made will also be exported and Ancestry will faithfully reproduce them with the privacy braces - that is maybe not a good thing. So another operation on the GEDCOM, after converting the private _APID note back to a tag, would be to delete all private notes from the GEDCOM. That could be tricky given that a private note may span multiple lines. Alternatively, perhaps a SQLite query could be used to unprivatise the _APID note before export.

 

I usually use PSPad for editing GEDCOMs. It has a regex line editor with separate Find and Replace fields and slightly different syntax from vi so this combo does the conversion of the _APID to a (private) NOTE: 

Find:       ^(\d) _APID (.+)$

Replace: $1 NOTE {Ancestry Tag: $1 _APID $2}

 

and this reverts it, whether a NOTE or a CONT, to a _APID:

Find:       ^\d [CN]O[NT][ET] {Ancestry Tag: ([^}]+)}$

Replace: $1

provided it is all on one line.


Tom user of RM7550 FTM2017 Ancestry.ca FamilySearch.org FindMyPast.com
SQLite_Tools_For_Roots_Magic_in_PR_Celti wiki, exploiting the database in special ways >>> RMtrix-tiny.png app, a bundle of RootsMagic utilities.


#14 RassilonRegenerated

RassilonRegenerated

    Member

  • Members
  • PipPip
  • 7 posts

Posted 06 July 2014 - 09:37 PM

I implied in my regex but did not explicitly state, the _APID is always a 3 _APID on import. After Exporting from RM it comes out as a 3 NOTE or a 4 CONT but should (i am pretty sure) always be restored to a 3. So I think your regex ^\d [CN]O[NT][ET] {Ancestry Tag: ([^}]+)}$ might fail in a minority of cases.



#15 TomH

TomH

    Advanced Member

  • Members
  • PipPipPip
  • 6254 posts

Posted 07 July 2014 - 05:43 AM

The original level of the _APID, typically if not always 3, is stored in the NOTE value imported to RM. No matter what level the RM export NOTE or CONT is at, the original level of the _APID is restored.

Tom user of RM7550 FTM2017 Ancestry.ca FamilySearch.org FindMyPast.com
SQLite_Tools_For_Roots_Magic_in_PR_Celti wiki, exploiting the database in special ways >>> RMtrix-tiny.png app, a bundle of RootsMagic utilities.


#16 TomH

TomH

    Advanced Member

  • Members
  • PipPipPip
  • 6254 posts

Posted 08 July 2014 - 07:45 PM

The original level of the _APID, typically if not always 3, is stored in the NOTE value imported to RM. No matter what level the RM export NOTE or CONT is at, the original level of the _APID is restored.

The Citation _APID is typically 3 but the Master Source _APID is typically 1. So the search/replace algorithm I used covers both cases and I see in the imported database the Master Source _APID in the Source Comment, e.g.: {Ancestry Tag: 1 _APID 1,1904::0}, and the Citation _APID in the Detail Comments, e.g.:  {Ancestry Tag: 3 _APID 1,1904::4006540}.

 

I have not yet done the upload cycle to Ancestry so I don't know if it is necessary to transmit the Master Source APID, given that it is implicit in the Citation APID and you seem to have done so successfully without..

 

Now that I see that the APID as a NOTE ends up in RM Comments, not Notes, the use of the privacy braces is moot as there is no privacy control over comments (in fact, I don't think there is for Source and Research Notes either, yet; only for person and fact notes). However, they (or some other delimiter) may be useful to separate the APID string from any other text that may appear there.


Tom user of RM7550 FTM2017 Ancestry.ca FamilySearch.org FindMyPast.com
SQLite_Tools_For_Roots_Magic_in_PR_Celti wiki, exploiting the database in special ways >>> RMtrix-tiny.png app, a bundle of RootsMagic utilities.


#17 TomH

TomH

    Advanced Member

  • Members
  • PipPipPip
  • 6254 posts

Posted 03 August 2014 - 01:36 PM

...Basically GEDCOM is a text file with one item of data per line. Adding a line break kind of srews things up, and that is all the fault of Ancestry for adding extra line breaks, ...
 
The problem I run into is that particularly sources have line breaks in the data itself. ...
3 DATA
4 TEXT Details For Marriage ID#529095
Groom Last Name: STUCKI
Groom First Name: Golden Autumn (21)
Groom Residence: Salt Lake City, Salt Lake, Utah
Bride Last Name: SALMON
Bride First Name: Mary Edith E
5 CONC vans (19)
Bride Residence: Salt Lake City, Salt Lake, Utah
Place: Salt Lake City, Salt Lake, Utah
Date: 04 Oct 1933
County of Record: Tooele
State: Utah
Volume: C
Page: 213
Notes:
Salt Lake T
5 CONC emple


Now that is clearly broken. Everyline shoud have started with a number followed by a tag...

I just observed the opposite problem, line breaks in the transcription are removed and the line contents concatenated in the Ancestry export. The transcription is exported in a TEXT block. Line breaks in the "Other information" field of a source are exported correctly within a NOTE block (I must retract: some are exported correctly, seemingly empty line breaks are but consecutive non-empty lines appear to be concatenated - behaviour is confounding).

 

So Ancestry is processing content it sends to a TEXT block differently from how it does for content to a NOTE block. GEDCOM supports CONC|CONT tags for both. Your search and replace procedure works for those cases where it outputs bare new lines with no CONT tag but I see no way to correct Ancestry's inherent deparagraphing.

 

I wonder if your example still exports as it did. Has Ancestry tinkered again or is it a difference between how our respective newline characters were generated. Mine was by manually adding text through the web editor whereas yours was undoubtedly generated by Ancestry likely a copy/paste from the referenced URL which page I copied from using Chrome and pasted into the citation editor with concatenation resulting in both on export. I wonder if browsers make a difference.


Edited by TomH, 03 August 2014 - 02:28 PM.

Tom user of RM7550 FTM2017 Ancestry.ca FamilySearch.org FindMyPast.com
SQLite_Tools_For_Roots_Magic_in_PR_Celti wiki, exploiting the database in special ways >>> RMtrix-tiny.png app, a bundle of RootsMagic utilities.


#18 DeaneHunter

DeaneHunter

    Member

  • Members
  • PipPip
  • 17 posts

Posted 13 October 2014 - 02:23 PM

....just wondering if there is any more to this discussion?  It seems to have been abandoned.... would maybe be nice for one of the participants to make some final comment(s), maybe summing-up, or pointing to a continuation, or something similar :unsure:  I was happily reading along and then started looking for the next page, or something! I am nowhere near at the level of these participants, but do have RM6 first, and Ancestry.com, and FTM 2014, and would very much like to have my Ancestry trees copied into my RM6(with full detail). I did do a sync to FTM and then a Gedcom download, and then imported that to RM6. But since have added to the Ancestry.com trees, so I am interested in this topic of discussion.  Thanks,



#19 TomH

TomH

    Advanced Member

  • Members
  • PipPipPip
  • 6254 posts

Posted 13 October 2014 - 05:15 PM

Deane, I got on a different track addressing the conversion of TMG databases to RootsMagic when the announcement was made about TMG's end of support. I do hope to get back into the Ancestry issues before long as I, too, have previously downloaded data in limbo en route to my main database and further additions in my Ancestry tree. Not a good state to be in...

The Ancestry-RootsMagic relationship is difficult. I don't hold out hope for a simple, easy answer. Best to keep pushing Ancestry.com to produce an API to which RootsMagic could interface.

Tom user of RM7550 FTM2017 Ancestry.ca FamilySearch.org FindMyPast.com
SQLite_Tools_For_Roots_Magic_in_PR_Celti wiki, exploiting the database in special ways >>> RMtrix-tiny.png app, a bundle of RootsMagic utilities.


#20 RassilonRegenerated

RassilonRegenerated

    Member

  • Members
  • PipPip
  • 7 posts

Posted 13 October 2014 - 08:21 PM

I am afraid my final answer is a rather bleak one. I stayed with this topic a bit further and was working on a transform script but after a while I realized several other problems with the exported data. Some are referenced above but the main one that put me over the edge was the duplication and truncation of source citations. Source citations are everything in Geneology and are the main reason to use Ancestry.com. The idea that complete and useful citation data was not being returned made the whole exercise pointless.

My end result, and the one I recommend to others now, is that I keep my database in a very clean state in RM and export from there to Ancestry and other services. I do not import back from any of these services. In the event I have new facts, people or sources, I add them to RM completely manually. Yes that means I manually make a source citation, extract all the data from the source by hand and associate with the correct person and facts. It is a slow process but it is far less time consuming than untangling a large database that has been chewed by ancestry.com and remerged. Unfortunately, if you want a clean, consistent, well sourced database, there is no shortcut to manual importation of the data. As fun and fast as Ancestry makes it to find and attach sources, there is no getting around the fact that a quality product to leave as a legacy for your family is going to take a lot of time to compile. This is a horrible conclusion as it means that in spite of all our advances in technology, we have still not figured out how to make genealogy softwares work together with a unified format.

But again I must stress, clean and clear, detailed sources are needed in genealogy or the tree is mostly worthless. The Internet has gotten too wild with amatures publishing lists of names and facts with no research or documentation to support any of the assertions. Services like Ancestry are great for research, it is inexcusable that they fail to have an easy export of the research once done.