Feedback on the new merge process
I tried the new merge process and here are my recommendations:
Include a way to edit a person's information from the merge window, like adding a middle name to an individual. The old merge process had this feature.
Include a slide button for the sources. I wasn't sure the source would be added once I merged the records. I had to go back to the person's page and check the sources to make sure the new source was there. The old merge process had this feature.
This process feels like we don't know what is really happening because the add buttons are gone. Overall, I prefer the old merge process.
Comentários
-
I agree! especially about the visual of information sliding from left to right.
1 -
I agree! I teach FamilySearch classes at our Centre and this is especially important for those not confident about the merging process. It is really nice to have the visual of exactly what will remain after the merge takes place and gives a new researcher confidence to proceed.
The "merge reasons" provided in the old version also helps those new to merging understand what information they are looking for if they are unsure, and gives them confidence to leave a reason as they know what it should look like.
Having the 3 questions on the old version, and the add buttons also gives "rushed" mergers some pause to evaluate whether or not the individuals are actually the same person or just have similarities in information. It may take a few minutes to review information, however, if the objective is for a more "accurate" record for each individual, the time is well spent.
I feel that the suggestion that "These 2 people have no significant inconsistencies. However, you should find proof that they represent the same person before merging them" may be "too quickly" bypassed and create even more errors in merging than already takes place. This is especially true in cultures where similar names are used repeatedly even within families.
AI is fabulous, but is often misunderstood as verification. This may be a disaster for accuracy in merging.
I feel the old version was near perfection for all levels of users!!! Those experienced, know how it works and can verify and proceed efficiently, and those more hesitant are given enough information and guidance to have the confidence to move forward with necessary merging.
0 -
I do like the warnings, but maybe it should be limited to the less obvious things (eg. 2 different places for the US 1920 census).
I don't feel like it helps to point out differences in birth dates. (And if you do, you need to automatically throw out any dates from the US 1900 census.)
0 -
Sorry, not a fan of the new experience. Way too basic and I forsee many erroneous merges of PIDs who are clearly different on the person page but only summarized on the merge screen. I also found that facts were not merged from the left side into the right side and there is no mechanism to force this. Situation: I placed the more complete profile on the right but there were facts I wanted to bring across from the left except no mechanism was available. I proceeded with the merge but when I reviewed the results the facts from the left had not migrated (as I hoped) to the surviving PID on the right, so I had to restore the merge and switch the places to maintain the fact. This will be a problem for two PIDs with separate facts on each side. Which facts will I choose to not merge across when I choose the PID to survive? Please return the detailed merge function. thx 🙂
0 -
Yes @JLG43, I agree that there are not enough choices in the new merge about what do keep or leave.
I think we could accomplish the same goal (quality checking) by just adding a few warnings to the existing merge function.
1 -
@CherylMillerBlack I really felt confident using both the detailed SourceLinker and the detailed Merge function as they were so similar and once I knew one, I knew the other. They made me pay close attention to every line of detail. I'm hopeful they will reconsider this new Merge Analysis tool and go back to old system that wasn't really broken at all. 🙂
1