Merge experience presentations
Presentation by Kathryn Grant and Robby Parker
https://www.familysearch.org/rootstech/session/improving-tree-integrity-the-new-merge-experience?cid=rt_copy
Comments
-
Thank you. Very interesting, especially this concept van the original unique person as defined by (for example) a 2012 entry (or a later date on first entry of a person, of course). Are there (future) plans to make this more explicit in the merge process? (For example, putting the change log in chronological order instead of reverse chronological order?)
Also thinking about the potential implications for the Quality system.
1 -
@FrankLittle thank you for the feedback. I have noticed that we are needing more, and better "steps" before merging. Kathryn has shared more info on merging. Please watch if you have time and let me know what you think.
I am moving this down to Discussions, so that we can have (hopefully) some "best practices" steps and methods, shared by our more experienced users.
Duplicates in Family Tree, Part 1: Why They're There and How to Find Them
Duplicates in Family Tree, Part 2: How to Resolve Them
0 -
Happy to respond here, Rhonda. One reflection from the start (Part 1), the PIDs are very helpful, since they are unique. I note them in my own copy of my personal family tree research, and log any changes or merges with the current PID. Even without any other support, this is a valuable point which other systems do not offer.
1 -
Wrt creating new people, as indicated by the PIDs beginning with P… (this years): one reason this happens in my experience is because the new PID is suggested by the system if it does not find a suitable person. Especially true when adding a marriage partner. This may not be because the vital information is too sparse.
By the time I've decided to add someone like this, I've already established to my own satisfaction that it's both a real person and they did marry. Dealing with persons in England and Wales it's usually someone found as a named partner mostly as "wife") in a UK census (1841-1921), and confirmed by a civil registered marriage (a free source is FreeBMD or one of the paid platforms). It'll be backed up if at all possible by newspaper notices (for the date, but on a good day also the parents and places). With luck, I'll already have the relevant birth and pre-marriage history from census or other data. I'm actually pretty reluctant to start adding people into the tree on my own cognizance if at all possible.
But I notice that the system isn't so good at finding people without baptism or church wedding data. That's mostly when the self-made PIDs come into play.
1 -
And one surprise for me: Find Similar People. Never thought to click on this.
What I think it's useful for in particular is to find and exclude similar people for the purpose of disambiguation.
(Though I must say some of the finds are rather off-the-wall/ Why is Charles Wharton (1799-1860) similar to Nanny Wharton (1790-1844) = and yes I know they are brother and sister, so I guess the answer is—same parents.)
1
