Merging Qualifications

Question

We have grown so much, I feel it's time that we create "Merging Qualification" for newbies.  Maybe something like must be with us 2 / 3 months before they can merge anything.  I feel this is getting a little more important now.  What do you think?

Answer 1

I agree that a waiting period probably wouldn't do much good.  Someone coming here for the express purpose of vandalising Wikitree (which we've recently seen) wouldn't be stopped by a waiting period.  That's just something we're going to have to continuously monitor.

The "honest mistake," though, was a matter of someone completely misunderstanding what a merge was for.  Maybe we need to clear newbies to merge after they've shown that they understand what merges are, what they're for, when to merge, and when not to merge.  Rather than a specific time limit, maybe just an "OK, this volunteer understands merge basics" from staff, a Leader, or an Arborist.  Basics would be something like:

1. It is clear these two profiles represent the same person.
2. Conflicting information needs to be saved and noted in the merged biography if there isn't solid evidence that one piece of information is wrong.  Even then, it might be good to note the error and its source for future reference.
3. Sometimes conflicts are serious and it isn't apparent what is correct.  In those cases, we use unmerged match, indicating more research needs to be done.
4. Care should be taken when looking at the merge page that compares the data sections (Merging Smith-11111 into Martin-99999), selecting the correct data and preserving the conflicting data for notes in the bio section, if necessary.  Also, conflicting family needs to be fixed here (often involves proposing another merge, e.g., two different records for the mother, but those two also represent the same person).
5. Clean-up after the merge!  This is another long discussion and a bit too involved to go into detail here.  What can be safely deleted?  How do you consolidate sources/repeated information.  And more.

It seems like a lot, but merging shouldn't be taken lightly.  At least not until some Code Wizard on the Wikitree staff invents The Magical Unmerge that puts everything back as it was.

Comments

I'm fairly new but have a thought on a possible help to prevent newbies or anybody from making mistakes on merges.  Sorry I don't know the name of the exact page but durng the merge process we see a schematic showing the seperate items of each profile two colums with a third column of what will end on the merge.  This schematic table is awesome and something like this to use as a merge help would be wonderful to use in the review process before initiating the actual merge process.  It would be just a great worksheet for ease in working up a possible merge.  Something to consider.

---

Yeah, I didn't know what to call it, either.    But that's what I meant when I wrote, "...the merge page that compares the data sections (Merging Smith-11111 into Martin-99999)..."

I'm not sure exactly what you're asking for here, though - a way to see that page before proposing a merge?  That might be helpful - anything that would make conflicts to be resolved more clear.

---

You can see that page.   Hit compare, instead of merge and the comparison page is displayed.

If they are not the same, don't propose the merge.

---

Hi Maureen,

No, she wasn't talking about the compare page where the profiles are displayed side-by-side.  She was talking about the 3-column merge page, that shows the individual data elements, three on a row (profile-1, profile-2, result of merge).  It just makes it a bit easier to spot the conflicts, I guess.

---

Exactly, Fred - - thanks!   But Maureen, thanks to you to, 'cause if you  both hadn't shown interest, this wouldn't be progressing.  I want it some way to do this more easily, but I don't know what everything is called or where it is and I've been discouraged lately.  I really think this could be a help to many as the ability to look at the comparison in the beginning could cut the length of proof in many cases.  I just hope others give thought to whether this could be of help in preventing wrong merges.  I know it would make it easier for me to have this tool to use from the get-go.  - - "Betsy"

---

Is this something you see when you upload a gedcom?

---

It's what you see when you go to complete the merge - but only if it's approved.

Ideally you would see it when you hit the button to propose a merge.  Then you could set up how you think the merge should be done.

 

Another thing that might help might be, if the computer didn't fill in the third column with default choices where there's a significant difference.  The trivial merges would still be easy, but the bad ones would be hard work.

Default choices give people the impression that the computer knows what it's doing, it's not their decision, it's not their fault if it's the wrong one.

---

I completely agree with the idea of preserving data fields where it is not in "column c", so to speak. Just yesterday I did a merge where there was a notation in death place that I thought would be preserved in the Changes page, but wasn't. Luckily I had made a mental note of it and was able to add it to the Bio, but if I hadn't, that valuable information someone took the time to enter would have been lost forever.

---

Oh that.  If I go to that screen and I find that I've changed my mind, I close the screen. Then the merge doesn't go through.

Answer 2

I don't like this idea at all.  

there is no reason to limit people's activity on this site based on how long they have been here.

Comments

ALL THE MORE REASON TO "QUALIFY BEFORE MERGING".

---

Ditto that Jillaine!

---

Chris we tried that some time ago and I really really hope it will happen this time.

---

Regarding people being Malicious, once we discover this is happening, we can remove privileges.    Maybe?

---

I also agree with Jillaine....

---

previous question about merging badges

http://www.wikitree.com/g2g/59260/would-you-like-to-see-more-member-awarded-badges

---

Chris, I like the idea about "Earning Badges".

---

Yes Michelle, that is such an incentive.... And fun to get.

---

I look for waiting merges regularly and I see propositions that should not be accepted but are from both PM..... ggrrrrr

And yes, Wikitree needs some people that know what they are doing.  To many newbies on Wikitree and in genealogy research.   Genealogy is not suppositions,  it is facts.  

For Gedcom... I never trusted them on the Internet.... to many errors made by people that don't know genealogy.

Yes I would like to by an Arborist.  The sooner the test is available, the best for me :)

---

Guy I sent you a message a week or so ago about your Arborist request.

Answer 3

I completely agree with you.

Thank god we finally have the merge feed. It tells a lot!! I've found a lot of newbies merging unaware of what a merge actually is. If people don't understand what is merge is then why should they be merging?

2/3 months sounds good to me.

Maybe a specific amount of merges completed in the pm watchlist first?

After taking the quiz maybe there should be a requirement to watch the video before merging?

Comments

LET'S MAKE THIS HAPPEN QUICKLY!  So the "300" or so bad merges doesn't happen again.  We are going so much and so quickly that we need to get a grip on this now.

---

I would like to add to this to try and create a middle road.   So, would like to limit the ability to do merges when you are not the profile manager.   There are some new folks that had problems at first and created lots of duplicates.   I do not mind them cleaning up the duplicates they created, so where they are the profile manager on both profiles, I say, let them do it.

Every day I have one or more profiles that are merged by someone new that does not understand what they are doing, and I spend an hour cleaning up the mess.  New users should learn on their own profiles and not on the profiles of "others".

---

Robin,

Unfortunately, limiting merging to profile managers would not have prevented the close-to-1,000 (not 300) vandal-merges that were made. They were made by the vandal adopting orphaned profiles, thus becoming a profile manager and being able to complete the merges without the 30 day waiting period.

Also remember that-- especially after 200 years or so-- "your" profiles and "my" profiles become "our" profiles once on wikitree.

All that said, I think we need some way to slow down the merge mania -- innocent or otherwise.

---

Michelle, what quiz? And what merge feed? And how do I find the merge feed? I think the auto-approval system has contributed to the problem. I see to many "merges awaiting action" that never should have been purposes in the first place or that have serious conflicts which do not get researched. In spite of the cautionary message or the red "now clean up the profile" after-merge message, I still see profiles that are a post-merge mess! And as long as I am ranting, the reason "same person" for a merge proposal is not helpful. Tell me WHY you think it's the same person - what matches, and what doesn't. It's really not about time or experience, it's about commitment and responsibility to correctness.

---

Robin thank you, especially for your last statement, "It's not about time or experience, it's about commitment and responsibility to correctness.  I started WikiTree in November, 2015.  I am probably considered a "newbie" by most.  I'm working hard, several hours each day, on WikiTree profiles.  There are lots of things to learn, and lots of details on Wiki are hard to find... such as the quiz, the merge feed, the merge video mentioned in the answer from Michelle.

I'm working with PGM, Military & War, and with Sourcerers, primarily.  I started working with them even before all my family was entered, because I think the mission of WikiTree is valid - to collaborate with connecting the human family on one tree. I am thrilled when I find unsourced files of people who can be added to Military and War because they were a part of the American Revolution, or possibly to PGM because they emigrated to New England before 1640.  When I add sources, biography, footnotes to profiles I feel a sense of accomplishment - I am honoring these people, and future generations will see this tribute and acknowledge the greatness of our human family.

Additionally, I want to say that in my work on WikiTree, I come across many, many profiles where the pm has been on WikiTree for years and years - with a biography that includes only unsourced gedcom information.   

In October I was welcomed graciously and generously to WikiTree.  A few people came along side me and helped me get started.  Is it true that the more deeply one gets into WikiTree, the hierarchy is revealed? Do some of the "old-timers" want to keep the group small? I hope not.  The purpose of WikiTree will not be accomplished with that mindset.

Answer 4

I don't agree on time but I would agree on knowledge.  

Who's better .... a new genealogist with 3 months experience or a genealogist with 10 years experience but just started on Wikitree ?

Comments

Guy,

Sometimes, it’s really not a matter of how much experience a person has doing genealogy research.  It’s a matter of understanding what merges are on Wikitree and what they are for.

Not long ago, we had a new member merge his parents’ profiles into his.  He apparently thought he was “connecting the family!”

Again, intentional vandalism is one thing, innocent, well-intended errors are another.  We just need to make sure people understand the most basic concepts of merging profiles on Wikitree before we “turn them loose.”

---

That is about what I was saying... people should know what they are doing before merging.

---

Yes, Guy.  But this leads me back to things I posted shortly after joining Wikitree.  I've worked on computers since they were big units with platter disks - - nothing like today.  That said, we worked very hard to have GOOD instructions - clear - - concise - - sequencial.  That could help much for someone beginning to use merge along with everything else.  Maybe we don't need to think of "restriction" so much as another committee/group to do Process Management.  Ok, done with the mini-rant.  Back to just discussing merges.  

NOTE:  I'm not knocking the people that created the instructions in the past.  I understood processes i worked on but the fellows following me sometimes didnt's get what I thought was clear.  Everyone who as WORKED on Wikitree since the beginning has my appreciation.  I just think we may need to make sure it is EASY to learn how to do Wikitree.

---

Agreed, Elizabeth.  I have told more than one newbie, "Wikitree is big and complex, and it has a bit of a learning curve."  That's not a knock on Wikitree; a lot of systems have been ruined by overdoing "simplicity."  Far too often, the answer ends up being to remove the functionality that's making things "too complex."

A full frontal attack on setting up a tree-structured set of Wikitree tutorials would probably yield some improvement.  There are lots of Help pages, video tutorials, etc, already, but not really a "Start Here" with links to the next things to look at, depending on each individual's interests and knowledge.  I don't underestimate how difficult that would be to put together.  It would be a VERY big project.  But I think the results would make it worth the effort.  As I said, there's already A LOT of helpful information on Wikitree.  Sometimes, though, finding that information is VERY challenging.

Answer 5

While reading the entire discussion thus far, I was able to relate to the comments by those with dual experience- in genealogy and WikiTree - with a possibly unique reaction.

Less than 2 months ago, when I joined WikiTree, I was brand new to BOTH.  I immediately embraced the "one-tree" concept, thinking it was axiomatic for all genealogy.  From (sometimes heated) G2G discussions since, I have learned that experienced genealogists often have a great deal of difficulty grasping this idea.

Also, I was totally clueless about all the nuances of merging considerations, which I now believe I understand pretty well (then again, if I have an inflated opinion of what I understand, I might be one of the offenders here instead of helping).

I would like to propose a solution to the problems addressed in this discussion by combining a couple of the thoughts expressed by everyone else here.

How about allowing "probationary" merging by ALL members - new and old - until some conditions have been met.  Here's my idea of how it would work:

1. All members are initially "probationary mergers".
2. Some group - perhaps a subset of arborists - has oversight over any merges proposed by probationary mergers.  Any merges proposed by probationary mergers has to be approved by a member of this group before the merge can be completed.
3. This group would serve as mentors for the probationary mergers.  When they do not approve of a proposed merge, they contact the person who proposed it and educate them about merging.
4. Some criteria for "graduation" from probation might be to complete some number of merges or perhaps simply upon recommendation of a merge mentor..

Comments

I am Pre-1700 Certified, are you?  I had to answer 14 Questions to earn my badge. I am proud of that badge and fairly sure that I understand the system. Are you following my drift on this issue?

Why not a similar set-up for merging. Experienced genealogists would have no problem, but if you got 3 or more wrong, (you set the criteria), It triggers a note to a mentor (possibly a group effort like the arborists) who can step in to correct the problem.

Becoming an arborist requires a certain number of merges, why not the same for a "Merger Badge".

I would suggest a set of pre-arranged merging exercises be included to expose the newbie to the vagaries of the system.

These last two at the discretion  and supervision of the aforementioned mentor.

---

I think it's important to ask, why does it become necessary for newbies to propose merges in the first place?  

My strong feeling is that the reason is directly related to their initial gedcom upload.  As Robin noted earlier in the thread, newbies sometimes end up with tons of duplicates where they are the PM on both profiles.  Newbies also have a habit of uploading "dirty" profiles, meaning the names aren't quite right or the dates are off.  Since they're not yet aware of the matching process or how it works, they don't know to look carefully to make sure an ancestor isn't already here or don't know how, so they go forward with the upload and only catch it later that a profile already exists for that ancestor.  

I suppose I could be wrong, but if I'm right, my suggestion is that any restriction on newbies should be on their gedcom uploading ability, not on merging.  In the process, I think a big chunk of the problem of faulty merges goes away.

---

Kyle, that's one question I'm fully qualified to answer, as a very recent (maybe still) newbie here!

It wasn't "necessary" for me to propose merges very soon after I joined WikiTree ... but I did.  Here's why:

1. I was very excited to be joining my family to the world tree and couldn't wait to find people that were connected to my family, so I went looking on surname lists to see if I could find them.  (Yes, I know now that there are other ways to find matches, but that is the first thing I did after entering a couple of family members).
   
    
2. One of the reasons I searched to see if any of my relatives are here is that I saw all the stuff about not wanting duplicate profiles, so I was trying to be careful not to enter a new family member who might already exist here.
    
3. I armed myself in advance with the knowledge of what I'm supposed to do if I find people who match those I already entered - somewhere on WikiTree, I read that I'm supposed to merge them.

Well, WikiTree was saved from any disasters I might have created by being clueless about what I was doing, but only because none of my family was here.

Please note that this had absolutely nothing whatsoever to do with a gedcom - I was clueless about what that was, too.

 

---

My experience is that many (most?) newbies are not as committed to arming themselves in advance with knowledge as you are, Gaile.   

I was there myself a year ago, and I tried to upload a gedcom with a couple thousand people in it before I had any experience on WikiTree at all.  Luckily, by then the gedcom screening and limiting process had already begun and the gatekeeper (Eowyn, I think), told me to cut it down to something much smaller.  I tried to do it, but couldn't figure out how to make my gedcom smaller technologically.  Only after realizing that I couldn't fix the problem technologically did I figure out that I had to approach things differently with WikiTree.  I suspect many of our newer users do know how to fix their gedcoms to fit the guidelines, and still go ahead with these smaller uploads before they really figure out what WikiTree is all about.

---

I'm a newbie, and I've needed to merge. Why? Simple: I've been at genealogy for over a decade now, starting from a big database built from work by grandfather & great grandfather. As I enter my own branches back and search for duplicates, I've found several cases where my ancestor is already on the system twice: once with the parents, another with the spouse.

Another example is when there's alternative name spellings. I accidentally missed a duplicate with an alternative spelling. I realized it pretty quickly, and then had to ask for a merge.

When you're new, especially if new AND experienced, you're likely to be entering new material, weaving some new strands through the fabric. I'd argue that it's almost inevitable that one will find needed merges along the way. Especially if, like me, you're starting with a large volume of material to weave in.

---

Gail;

I like your 4 points, but as I believe was mentioned before, being a member for a certain period of time shouldn't be the only qualification.  There seems to be a number of members that have been on WikiTree for quite a while that don't know how to merge properly.

Answer 6

I would agree with a lot of the comments below. Maybe not a time frame of when you can merge but perhaps an educational quiz like the pre-1700 or even maybe something like what they do for the pre-1500 badge where you have to show them a proper merge and cleanup.

As a mentor, I get a lot of questions about merges and many people just don't understand the procedure for them. To keep allowing the merges to go on and then send every Ranger, Mentor, etc in to try and clean it up and walk them through it is difficult. There are not enough of us to go around and then we feel bad because someone hasn't had a mentor yet or had someone get back with them in a decent time.

I love helping people out but maybe if we put it in layman's terms, maybe more would understand. With my net, I cannot watch the videos. My speed doesn't allow it. Maybe we can do a page where we show screen shots instead to show what we are talking about. Maybe we can implement the share a screen through join-me or skype so we can help people more.

All I know is I spent the last 48 hours cleaning up a HUGE mess in the Ostrander line and it's no where near done. Bad merges, incorrect LNABs, no sourcing, or gedcom clutter. It's getting crazy.

Answer 7

I think there is a difference between being a good genealogist, and understanding how this web site works.  Mergers help the wikitree run cleanly.  You can be a fantastic genealogist with wonderful records but if you do not understand how the nuts and bolts of Wikitree works, and how a merger incorrectly processed can cause lots of issues later, all your knowledge counts for nothing.  I think Chris's suggestion is a good balance between not discouraging people from proposing mergers and avoiding any bulk changes that would be cause real disruption for people not familiar with the merging process.

Personally, I was terrified the first few times I processed a merge!!  :-)

Comments

Some basic questions I have prior to merging.

1) Does the information in the vital fields match with the bio and sources?

2) Are there any additional comments someone placed on the public bulletin board or in the bio that may shed more light on the profile?

3) Do the sources listed only reference online unsourced public trees or have no sources?

4) When looking at the changes tab, is there someone currently working on finding new and better sources for the profile?

5) Was the profile part of a gedcom that contains only broken links or references only sources found on paid subscription websites?

6) Was the merge default approved due to a PM is MIA, or is it possible to contact them regarding any unclear data found on the profile?

7) Can I find better sources that substantiate the profile?

8) Do I have time to analyze any potentially conflicting information and note these after the merge is completed just in case.

9) Am I constantly aware that after a merge is completed previous information of a profile may be lost?

10) Will I consolidate all of the exact information and clean-up the profile as per http://www.wikitree.com/wiki/GEDCOM-created_biographies?

11) Are there more duplicate profiles tied to the merge that can be proposed for merging?

12) Have "Old Style (Julian calendar)" and "New Style (Gregorian calendar)"  issues been resolved and noted?

Although many folks with the best intentions complete merges, we all must realize that doing a "Bad Merge" is much more detrimental than not completing a merge due to it necessitates reconstruction of profiles and lineages.

---

David;

Your point 11 is important because if the answer is yes, what is the lowest profile # to give you the merge direction and final profile number.