Mergathon needed ?

Question

Should Wikitree consider having a MERGATHON ?

As of the end of May 2023 there are over 8200 merge requests that are in default approval status and just waiting for any Wikitree member to come along and process them ...

Jan 2023 current start of list =
https://www.wikitree.com/index.php?title=Special:BrowseMatches&type=pending&order=dateup&canAct=1

May 2023 current end of list =
https://www.wikitree.com/index.php?title=Special:BrowseMatches&limit=20&start=8200&type=pending&order=dateup&canAct=1

During May 2023, I updated about 200 entries from this list ... but obviously more help is needed to bring default merges up to date.

Comments

If you use wikitree +, you can use ApprovedMerge to find all approved merges. By adding a location, you can find them in a country or location for a smaller list.

---

do you know if there is a list of magic words that work on wikitreeplus ?

Otherwise do you know the magic words for approved merges with duplicates between global tree and unconnected ? Thanks

---

The Help page for Wikitree + has a section with the Magic Words.  

I am not sure what you are asking about.  Wikitree + isn't going to look at profiles to determine if they 'should be' in a proposed merge

The following are the words that can used to check on the Proposed Merges. 

• ApprovedMerge: Limits results to the profiles that are waiting to be merged.
• pendingmerge: Limits results to the profiles that are waiting for approval.
• unmergedmatch : Limits results to the profiles that have unmerged matches.

Following are words that are checking connections

• connected: Limits results to profiles, that are connected to the global tree.
• PublicTree: Limits results to profiles, that are connected to the global tree only with public relations.
• PrivateTree: Limits results to profiles, that are connected to the global tree through private relations.
• unconnected: Limits results to profiles, that are not connected to the global tree.
• unlinked: Limits results to profiles, that are not connected to any other profile.

'ApprovedMerge' returns 16,191 profiles, which should be approximately 8,095 proposed merges that are approved. If I add 'unconnected' to that query, it returns 1,572 profiles. Using 'connected', it returns 16,193 profiles

Just for further info.  

PendingMerge returns 22,495 profiles, 

UnmergedMatch returns 44,893 profiles.  Gillian had asked the Arborists to work on the Unmerged Matches in 2022, so that group was reduced quite a bit last year.

---

No I wasn't asking about whether they "should" be merged. I took the words "duplicates between global tree and unconnected" from the Data Doctors Weekly Report which is a category that is regularly tabulated, so I thought there would be a magic word for that. I will look into it more. Thanks for your help! 

---

N, does this WikiTree+ query give what you are looking for - https://plus.wikitree.com/default.htm?report=srch1&Query=Suggestions%3D106+ApprovedMerge&MaxProfiles=500&Format=&SortOrder=LNAB&PageSize=50?

---

Thanks for that suggestion :)

Answer 1

Bad merges are really hard to correct - they cannot be "undone".

To do merges right takes time and consideration.

Thons tend to encourage speed and quantity, so I would be hesitant to say that having a Mergaton would be a good idea.

Instead - maybe a month-long recruitment drive for the Arborists Project? If nothing else it would increase awareness of the project, which has excellent info available at [Project: Arborists], including a link to the project's [Helpful Information for New Arborists].

Comments

This is the reason why I even only suggest really obvious merges. In the moment when there is a little bit of contradiction (not talking of birth year 1853 or "before 1860") I stand back.

Same goes with sourcing. Where there is something unclear if that is the correct family, the profile sticks in the list.

---

Agreed Liz - when I go down a page hoping to complete some merges, I encounter many that are in need of research and sourcing before they can be merged or rejected.  This takes time and not a task for a Thon.  

There are dozens of new duplicates created weekly so it is not surprising that the list grows longer in spite of our efforts to shorten it.

---

Liz,

I agree with your sentiments wholeheartedly

---

Having to resolve conflated merges because of people completing a proposed merge is a big problem. Too many people 'assume' that every proposed merge should be completed without rechecking dates, locations and family connections.

---

OK understood :(

I always participate in the thons, but never have really high numbers, because I always stress quality over quantity, but I am aware that not everybody has my viewpoint.

Maybe they could have an "Arborists Weekend" where quantity is NOT stressed :)

---

Last year Arborists worked on Unmerged Matches, since those have to be reviewed and re proposed or just removed. We got that number improved.

---

I totally agree with Liz. For every 10 merges done by someone who isn't careful, we end up with 3 conflated and erroneous profiles that then need to be picked apart.

We've had a few members "going for the badge" that aren't carefully comparing and making smart decisions.

I would instead encourage those who proposed the merge to go back after 30 days and complete it if they are sure that it's valid, since they were already looking at the profiles in question and presumably have some familiarity with the families.

---

yes Jelena, I agree with your idea. Right now I am only merging very obvious dups. Because there are so many, many merges, I am passing for now on ANY merges that have questions. I leave questionable merges for those who are more knowledgeable about the family or when more time is available for research.

---

@ Linda - I spent 2022 working on the unmerged matches initiated by me ... I was surprised to find that they weren't all initiated by me. Some were merges that I had proposed that someone else had set to be an unmerged match. It took all year, but I went from hundreds to just 9 (really difficult pairs). 2023 had a few added that I haven't gotten to yet, so I'm up to 12.

Answer 2

I think that is a good idea. I try to do merges myself to help clear the list, so we definitley need more members to help keep up. Also when you do merges it create even more merges. A never ending cycle.

Answer 3

I don't even know how many times I've proposed merges for duplicate profiles. (Usually, it happens while I'm searching for somebody else, and the duplicates come up in the search results.) I just propose the merge and move on. Checking to see how often (or, more accurately, how rarely) the merges actually happen would just give me ulcers.

Comments

I do that too Greg, but one extra step I take is to regularly check

"Pending Merges Initiated by Me" at ...

https://www.wikitree.com/index.php?title=Special:BrowseMatches&type=pending&order=dateup&canAct=0&requested=  (ending with YOUR 8 digit profile#

Say if everyone checked that about once a week (or monthly), then by only doing a couple merges, they could be kept up to date :)

---

I agree N Gauthier. I check weekly.