I keep my main tree in Gramps where I can store virtually everything I
need and publish as I see fit. It represents what I determine to be the
evidence-based ground truth.
I also have an account on MyHeritage from where I can solicit
improvements and corroboration from a wide range of fellow contributors,
some very trustworthy via DNA matches and others a bit less.
As things begin to clarify with time and as DNA matches permit I can
place more trust in segments of the MyHeritage tree which I should
âimportâ with all their citations and data into my Gramps. These
segments are often not entire clean lines of ancestry or lines of
descendancy but fragmentary parts of those.
The ideal, I suppose, would be to âtagâ an identified segment from the
tree at MyHeritage and (as if by magicâŠ) only export the tagged
elements and reference data, and none of the attached tomfoolery it
references like myriad attributes for I canât tell what. But it limits
me to just broad exports like âdescendants of XYZâ or âancestors of
ABCâ, thereby bringing in lots of other spurious data and individuals to
the output Gedcom file.
The import and the surgery begins! Then the fix-ups.
I canât be the first to wish for a better work-flow for this aspect so I
wonder if someone has already developed a better approach to harvesting
âgoodâ segments of an online DNA matched tree ( or any arbitrary
segments of an online tree for that matter) for easy import to Gramps.
Things I have considered so far:
(a) wholesale import into a new Gramps-tree first followed by tagged
export out of that Gramps tree into my truthy-tree. Followed by fixups
done by a gramps tool (as yet undeveloped)
(b) sed and awk scripts to automatically âsanitizeâ the Gedcom exported
from MyHeritage before importing into my truthy-tree. Folowed by manual
fix-ups. (my current work-flow!)
Any time you intend to merge work by others into your database, it is always a good idea to first import the other family tree into a new empty tree within Gramps. Here you can make any tweaks to ensure it will match how you store and organize your work. Then from the new workâs backup you can import into your tree.
As with any work like this, backup, Backup, BACKUP!! Make sure you have a backup of your work in case you regret your actions.
And then it a matter of merging. And it is more than merging people. Merging places, sources (citations ?) will often need some attention.
My only other piece of advice (or my personal preference) is to keep the database coming into your tree small. It will make things easier knowing what needs to be merged. Tags are a helpful tool which can be set for all imports. Again it can help identify what needs to be merged.
Not having used MyHeritage, I cannot offer options for downloading from the site. If you cannot pre filter the download, you can use filters and exports from within its Gramps tree you will make to pare down the actual banches or families you want.
I canât be the first to wish for a better work-flow for this aspect so I wonder if someone has already developed a better approach to harvesting âgoodâ segments of an online DNA matched tree ( or any arbitrary segments of an online tree for that matter) for easy import to Gramps.
I can very well imagine that you are the first one. Let me eyplain why I think so:
DNA genealogy is a very new field and is still viewed with suspicion by many researchers. This has to do with the fact that new things always have a hard time at first. Unfortunately, it also has to do with the fact that the black sheep of the industry first recognised the (financial) benefits of DNA. They gave the DNA thing a bad touch.
Gramps is one of the most powerful but complicated and challenging genealogy tools. MyHeritage is at the other end of the that scale. I know that, because I wasted many years of my research time on MH and ended up at Gramps.
Summarising data in a scientific context must always be done conscientiously. This must either be done by hand or the results of scripts must be confirmed by a large number of samples. Additionally, MH and Co. are not reliable sources. Therefore, either way they are not agreed for the scientific context. With the exception of exact DNA matches, perhaps.
If you are not interested in the scientific context, I would like to know the reason for your effort. Leave the complicated Gramps behind and transfer all the data to MH, enjoy the supposed benefits and rejoice in seemingly new relatives every day.
Please note that MH can change the import/export functions or adjust the format at any time. MH is not interested in (serious) exchange with other tools. Why should they? Itâs the only way a paying member can become a non-paying member. The higher the hurdle, the better.
Please allow this personal comment: It may be that you can trust matches in MH based on DNA. But any other match on MH (and Ancestry, etc.) must be marked as âuntrustworthyâ. If you are interested, I can tell you some stories from my very small family tree that show how problematic and unfortunately often wrong the fast matches are.
Iâm sorry I canât help you. I hope that I could at least explain to you why I believe that there is no one else here with exactly this problem.