ATTENTION Curators, please assist

Started by Shmuel-Aharon Kam (Kahn / שמואל-אהרן קם (קאן on Sunday, September 5, 2010
Problem with this page?

Participants:

Profiles Mentioned:

Related Projects:

This discussion has been closed by an administrator.
Showing 1-30 of 8939 posts

This Public discussion is being started by one of your friendly neighborhood Curators (See announcement: http://www.geni.com/discussions/6000000009728208335 and introductions: http://www.geni.com/discussions/6000000009792416945 ).

The purpose of this discussion is to enable you to easily post requests for assistance. Posting requests here will get you a much faster response than contacting us privately, as all of us (presently 24) can all see this discussion, as well as many other users who can assist as well.

Presently what we can help you with is:
* "killing zombies" - Zombies are historical profiles that are wrongly marked as living, and thus block merges. Please continue posting these requests here: http://www.geni.com/discussions/6000000008897331095
* Make public – if a historical profile is wrongly marked as private, we can ask the system to check if it should be marked public.
* Complete pending merges between ANY pair of public profiles.

A MAJOR caveat to our work is that we can only help with profiles/trees already connected to the so-called “big-tree” (really a forest) of almost 50 million profiles

If you found a duplicate tree that is NOT yet connected, we can’t help. You’ll have to go the usual root - initiate contact with the manager(s) of the other tree.

We most STRONGLY recommend that you ALWAYS first try and contact the other managers. You WILL get better and faster results that way (we are a last resort).

We also HIGHLY recommend that anyone who is serious about merging in the shared parts of our tree, join the Collaboration Pool. You will find there 500+ people there all wanting to help each other out. Read about it and JOIN here: http://www.geni.com/discussions/6000000008539319639

While initially we would like to concentrate on cleaning up the shared parts of the historical tree (such as European Royalty), we can and will assist anywhere needed. We will soon also start discussions that are for specific “areas” of the tree, for better focus.

What we can’t and WILL NOT do is:

* Do YOUR work for you. If you see duplicates in a tree, and can stack them, please do so, BEFORE you ask us to complete the merges. Do the maximum that you can. There are 5,000,000 of you users, and only 24 Curators... WE need YOUR help at least as much as you need us. This is a TEAM effort.
* Before you ask us to merge a "stack" of pending merges, do through the stack yourself, and check if all of the profiles belong there. It is NOT too hard, and you'll save everybody a lot of work and grief. You might find the following guide useful: http://wiki.geni.com/index.php/Resolving_Merge_Issues
* Connect new trees to the “big-tree”. If you want to do so, please have the courtesy to explain to the manager(s) of the unconnected tree, what this means and involves. MOST problems in the “big-tree” are due to people being dragged into it, against their will or without anyone explaining what this means.
* MOST importantly - We will NOT force ANY decisions on any one. If you have issues with the other manager(s), please try and work them out first.

Feel free to join the discussion, we'd like to hear your opinions as well.

Thank you,
The Geni Curator Team

One additional note:

While we VERY MUCH DO need your help, from time to time, as major clean-ups are done in specific areas of the tree, you will see many merge-issues around there, some of them even marked as "WORK IN PROGRESS", either in the profile name itself, or as a Curator Note at the top of the profile.

If you see these, then PLEASE do NOT RESOLVE these merge-issues, EVEN if they are in your merge-center. When heavy work is in progress, your attempts to help may actually get in the way.

I'd like to repost this list as well:

THINGS PEOPLE CAN DO TO HELP

1. Look for and respect the occasional "Work in progress" messages from Curators for some difficult areas - we're doing our best to fix these. They are temporary :-)

2. Avoid using Hot matches to merge - most bad errors seem to be initiated by these

3. Be VERY careful when linking profiles to be merged. The wrong merges and errors happening out there are absolutely incredible ... The most common error here on Geni is merging two people that are DIFFERENT. Remember to ALWAYS check parents, partners and children before merges.

4. Always open all the profiles involved in a conflict and read the information BEFORE linking them to be merged (learn to use multiple tabs if you don't already)

5. Open profiles and try to REDUCE stacks, checking carefully if the profiles stacked are really the same. There are a LOT of bad stacks out there with many different people linked to be merged. Unlink and/or ask for help. These "bad stacks" usually lead to a lot more bad linking. (Post the remaining merges to our upcoming "Curators needed" thread - will be up soon.)

6. Tell Curators what areas you are or will be working on to avoid chaos caused by several people involved in the same lines simultaneously. Important for the "busy" lines of the tree only, i.e where there are many managers and errors/issues.

7. Stick to areas that you know very well, and start reading/studying before entering a "new" area - educate yourselves :-)

8. Avoid (and delete) references to private websites and your own private Gedcom-notes as "sources" - use PRIMARY sources and where is exists: scholarly articles. A Rootsweb or LDS page is NOT a "source" - some of these are the cause for some very wrong connections out there.

9. Read and start Public Discussions for profiles where there are questions about connections etc. A lot of the time somebody can help. (Tip: profile names/areas etc in Topic, not just "Who are the parents?" - that way we know what area you are talking about without everyone having to read everything.)

10. Read our Naming Conventions, and leave Name conflicts to Curators/main managers (normally: keep data from Main). If a name looks weird you might want to ask why instead of changing it.
http://www.geni.com/discussions/6000000007295742137

@shmuel: "Complete pending merges between ANY pair of public profiles"

This philosophy changes from time to time, and it is important to get it right.

You are now saying that you want us (non-curators) to merge any and all historical profiles that we come across that we believe to be proper merges?

Even if we are merging into a profile that shows multiple parents? Multiple spouses? Wildly (or weirdly) different birth or death dates? Or wildly different birth or death locations?

If we do a profile on one of the above, should we notify you to clean up the mess?

Thanks for your hard work, and hours spent.

Please examine Thomas Thayer <Thomas Thayer>, b: 8/16/1596, d: 6/2/1665. He has two many parents, too many children, and too many siblings.

Please examine Captain Roger Dudley <Thomas Dudley>, b: c. 1540, d: 1586. He has too many parents.

This is a good opportunity for me to say this: the biggest problem for me is - when I try to clean up the mess in the big tree - that I very often can´t load relationship in a profil, so that I can get rid of wrong parents. I try and try again and again, but get a message, that Geni is out. Even if it goes through, I have seen, that after deleted a relation to a wife or husbond ( connected 10 times) - I get 20 connections. Why that? Am I the only one with theese problems???

@Mark Harold Melmed

I believe the best method for cleaning up is to get as many merges done as possible and then PROs can break the bad relationships.

Would you be kind enough to advise the correct parent relationships for your requests, and also include them in the "overview" tab of the profile for the future?

We're developing a "suggested template" for the top of the "overview" tab that goes something like this:

- Name, birth / location - death / location, why important.
- Child of
- Married to
- Children are

website references
discussion

The info is in the Mary Winthrop profile already, just needs a little rearranging, if you'd be willing to take that on.

Did you put in the information about John Colver? It's fascinating.

Please examine Ursula Alice Thayer (Dimery <Ursula Thayer>, b: 1568, d: 1628. She has too many fathers, & too many children.

Please examine Thomas Holbrook <Captain Thomas Holbrook, II>, b: 1618, d: 4/11/1705. He has too many parents, too many wives, & too many siblings.

Mark -

I took a look at the Winthrop gang. I am connected through Elizabeth Fones Feake Winthrop Hallett.

It looks like Mary Winthrop Colver is the daughter of John II, Governor of Connecticut, and I have adjusted the tree. The resolution John Sr's tree is an ongoing effort.

@Erica Isabel Howton: In most instances, I have no idea who the suggested parents should be. I obtained all of my historic profiles from another genealogist years ago, and uploaded them to Geni when I uploaded my entire gedcom file.

I am very unfamiliar with the people in the historic tree, and am very happy to let the curators sort things out.

Mark,

Understand and good to know.

Janet, thanks for snagging the Winthrops. I'll take on the Thayers, I seem to be some sort of third cousin.

If you have problems/questions etc related to the merge of Anglo-Normans: see this special Discussion for those:
http://www.geni.com/discussions/6000000009840568419

Mark,
I'm saying that we CAN merge any two profiles, this does NOT actually mean that we should, at PRESENT.

For now, while working on the "busy parts" of the tree, we are trying to reduce the "stacks" of merged profiles that are ALREADY entangled in the "big-tree". This will make the tree more usable, and reduce the number of conflicts, and thus give everybody a much clearer view of the tree. Some parts of the tree have 20! copies that are all PARTIALLY merged in. This creates a mess. With that mess reduced, merging in additional copies becomes MUCH easier. For example, there are presently at least 150 copies of Charlemagne. In theory, as a curator I could go on a binge and merge them all, but to do so, would cause so much damage to the tree, that would take months of work to fix... THIS is also yet another reason to NOT use tree-matches (AKA "Hot-Matches") for "celebrity" profiles.

That said, a LOT of you, will be having requests in smaller, more focused, LOW "traffic" parts of the tree. As such, merging in new copies in these areas, won't effect anybody else, except for the involved managers.

Benthe,
while this is not entirely relevant to your issue, it WILL improve performance - when working on difficult parts of the tree, reduce the number of generations displayed. This is done via the Preferences tab at the bottom of the tree. Sometimes you need to display the minimum of 1 and 1.

The multiple copies of the same connection is a known issue with Geni. They are working on it.

Mark,
if you seen a profile that has more than 20 managers, then obviously a lot of people have this profile in their tree. Some of the REALLY famous people, have HUNDREDS of managers (Charlemagne, mentioned above presently has 371 managers!). You can see this information when looking at the side-by-side view of a pending merge. I would suggest that if a profile has more than more 50 managers, then people should not merge them for now.

@Shmuel,
I always use 1-1-preference, so that surgestion did not help me (but thanks anyway!) . often I can`t load relationships where the problems are, and the o.k. - merge do not need it.

Speak plainer.

Hello, @Mark Harold Melmed. Mark, I understand your concerns. They are frustrations that we have all encountered and continue to encounter. One of the reasons that the new President of Geni, Noah Tutak, has begun to implement the new Curator system and other changes is so that we can begin to clean up the innaccuracies in the family trees that we share in common. We really appreciate people like you who are conscientious and concerned about getting these relationships correct. For example, I have Holbrook ancestors so I spent some time last night investigating the Thomas Holbrook case you presented. It appears that through faulty merging, at least two different Thomas Holbrooks have been merged over time into one single profile. This is the kind of careless merging we are trying to cut back on and hopefully end. Just because two people have the same name and lived in the same century does not mean they are the same person--but many people in their haste just stack them together and merge them.

One Thomas Holbrook was born 01 Mar 1593/94 Broadway, Somersetshire. England and died 10 Mar 1676/77 in Weymouth, Plymouth, Massachusetts. He was the son of William Holbrook and Edith Coles/Saunders Holbrook. He was married to Jane Powyes on 12 Sept 1616 in St. John's, Glastonbury, Somerset, England. Children were John, William (Capt), **Thomas (1620's-1697)**, Ann, Elizabeth and Jane.

He and his son Thomas seem to have been merged at least once or twice with *another* Thomas, son of John Holbrook (if the source below is to be trusted, though it may take further research to discover), who was born around 1624-25 in Weymouth or Dorchester, Massachusetts and died April 11, 1705 Sherborn, Middlesex, Massachusetts. Apparently this gentleman was married 4 or 5 times: to Experience Leland, Hannah Shepard, Margaret Bowker and Mary Rogers. For a profile on him with additional information, see page 60 of Who Begot Thee? http://books.google.com/books?id=e4MxAAAAMAAJ&amp;printsec=frontcov...

The bottom line is that if there is a family like this that concerns you, you can do all of us a service by researching it and becoming acquainted with the best genealogical and historical information out there. Much is online. I am not an expert on the New England families, but I know that there are several trusted sources of secondary information such as the New England Historical and Genealogical Register.

Curators Daniel Walton, Lori Wilke and Gene Daniel are also co-managers of this latter Thomas Holbrook profile, so perhaps they can add more expertise to some of these New England research issues.

Thanks, Mark, for bringing these to our attention. Together, perhaps, we can all work to get these accurate.

I guess there is a simple reason for the existence of more then 200 hundred profiles for Charlemagne. That is that it is difficult if not impossible to evaluate if a profile and tree is trustworthy, and if it is a good idea to connect your own tree to it. And in case of doubt, evryone will make a new profile.

The curator idea is very good, but what is also needed is a kind of certificate to put on a profile and tree, convincingly enough for everyone. I guess such people exist. ( I am a complete amateur)

Below is an online link to the above mentioned The New England historical and genealogical register.

The New England historical and genealogical register (1847) Volume yr.1912:

http://www.archive.org/details/newenglandhistor1912newe

(I'll stick a copy of this link in the "Internet Resources" thread also)

Nicolaas,
the reason for 200 Charlemagne is actually because at present, when people upload their GEDCOM files, Geni can NOT check if the profiles it contains are already present in the tree. The chance that any of these copies were entered by hand, one profile at a time, is extremely low. This is an extremely difficult problem to solve, if only because many of the VIP trees in these files are different from each, so matching them up with the existing tree is no easy trick. Secondly, at the time they join, most Geni users don't really understand the "big-tree" concept, so they want THEIR own tree, duplicates and all. Eventually these new trees get entangled in the main one...

Geni is actually working on giving us tools to "Lock" profiles and mark them as Verified, so that we will have a master copy that is vouched for. Having that, the tree should become a LOT more stable and reliable, and will also probably make filtering new GEDCOM files easier.

Shmuel,
OK, I see your point, but I did it manually, one by one. But it should be easy to check by GENI how these profiled were enetered.
I look forward to verified or certified profiles, that was basicly the point I tried to make.

I have a question that may be related to this subject. Why am I suddenly getting hundreds of request to merges on profiles that were last modified by someone whose account has been deleted. Is this another GEDCOM technical difficulty that hasn't been resolved yet?

Was sort of happy having my merge files down to zero (or close to - there is still a tree issue that I can't resolve because apparently the generation has so many variations of the same people that I time out - either that or it's the connection here in Chile), and now having all these merge requests that I can't resolve is disconcerting.

Anyone else having the same problem?

Ben - Geni was running a program this weekend to find some lost merges. They all turned up in all our merge centers as last modified by 'account deleted'.

If they are in your merge center, you can approve the merge. Just double check the dates, parents and spouses. If it doesn't seeem right, refuse the merge. if it looks good, approve the merge.

The merge process will remove that merge from your merge center and continue on it's merry automated way, although it may move to your tree conflicts or data conflicts.

If you don't want to deal with a merge you can forward it to one of the curators who is a co-manager, or delete your name from the manager list.

Thanks for all your help and questions.

Janet - appreciate the advice. Have started merging away again.

Really don't want to give up management privileges; I feel confident in my research and do have an interest in the stories of my ancestors being told correctly. My only hold-ups are in the technical issues - for instance, how do you "forward" something to a curator? (An example of something I would forward is where it says to view other merges first, even though the merge is one that you've already dealt with - it's obviously not a case of mis-merging as the personalities are the same, but the issue never seems to resolve itself, or leave my merge box, even after addressed a half-dozen times.)

Sorry if this appears to be all newby stuff. But I do appreciate the help.

To bring a profile to a curator's attention:

Open the profile, copy the web address, come to the discussion, paste the web address and say a few words.

This can also be done directly by a message to a curator, but I think the group prefers that they get posted here.

Glad to help, and glad you're helping too.

Cool Schmuel! I am so glad there will be verified profiles. From there an inverted permenant pyramid can be created so once we find our ancestors back to certain verified ancestors and viola! instant portion of tree added! I like that concept.
Oh, btw, I had 68 merge requests then it shot up to 2830 merge requests. I suddenly felt tired and wrote 7 more pages of my script. Thanks for the inspiration all!
Your distant cuz, Dez

Ben & Janet,
it would actually be better to "TAG" a profile instead of copying the page's URL. It's much easier to read, as such.

You tag a profile by using it's Profile-ID. This is the first long number that you will see in the profile's URL. Copy that into the discussion between double square brackets. Like this (only without the spaces) [[ Profile-ID ]]. Take President Obama for example:
* His main profile is at: Barack H. Obama, 44th President of the USA
* So his Profile-ID is 6000000001852266021
* Properly tagged you get Barack H. Obama, 44th President of the USA, which is much easier to read than the URL and shorter, and will always display the current name of the profile if it is changed.

The reason I would disagree on "tagging" rather than "copying" is that source material may vanish from tagged pages. Granted, this could be because of updates (as would be the case in most of Wikipedia), but some of it could vanish, either when a site folds or when access to it becomes by subscription only.

Further, to me it is much more convenient to be able to read the information on one site rather than having to wait for my connection to upload it all. (And with net neutrality likely going away, I think speed of transfer is going to become something of a major issue soon for all of us soon.)

To me, anyways, the biographies are of secondary importance. What is most important are the references to source material that often are attached to the biographies. That to me clarifies whether or not the information posted is accurate.

Hope that clarifies why I copy and paste instead of tag. -Ben.

Dez, that's hysterical.

Ben, I tend to agree about copy/paste even though it isn't as high tech and doesn't automatically update. Pros and cons for both methods.

Showing 1-30 of 8939 posts

Create a free account or login to participate in this discussion