Process that enables curators to merge isolated trees into the big tree [new c. Feb 2015]

Started by Private User on Thursday, February 12, 2015
Problem with this page?

Participants:

  • Private User
    Geni Pro
  • Geni Pro
  • Private User
    Geni Pro
  • Private User
    Geni Pro
Showing all 7 posts
Private User
2/12/2015 at 3:04 PM

On Feb. 10, Mike made clear mention of

"a process we're running that enables curators to merge isolated trees into the big tree, if there's a suitable match. This process helps everyone and the response has been overwhelmingly positive, in part because it allows a user to get into the big tree without having to pay for a Pro account to see the match / request the merge. The process of reviewing the matches was intended for curators, whom we've cautioned to scrutinize these extra thoroughly."

He also added, "As of today we've released some fixes that should remove these matches from your merge center and remove the false implication that the merge was requested by a Geni user, instead clearly indicating that the match was made by the Geni system and requires extra scrutiny."

[See Mike's actual full comment here: https://www.geni.com/discussions/142108?msg=1000812 ]

Private User
2/12/2015 at 3:26 PM

Mike's comments were inspired by several folks commenting on Merges reported as being requested by someone who had not requested them (folks who also had actually not been on Geni in a long time) -- note the folks Mike tags in link above, and go back about a page or so in that thread to read their comments on the issue, if you are interested.

Private User
2/12/2015 at 3:27 PM

Apparently such merges are now being made, and Mike says the response to the merges has been "overwhelmingly positive" - but he does not say what will happen if merges are made on the Assumption that folks would like to be connected to the World Tree, but it turns out that the folks on the formerly Separate Tree do not want to be connected.

2/12/2015 at 10:01 PM

Lois - if someone objects to their (previously standalone) tree having been merged into the Geni world tree, they need to contact Geni Customer Support.

Private User
2/12/2015 at 11:02 PM

Lois, I think we all should take a look on the history for Geni.

We can start with the press release from January 16, 2007: "Geni.com Launches to Create the World's Family Tree" http://www.prweb.com/releases/Family/Tree/prweb497693.htm

You can read more history here: https://www.geni.com/projects/Launching-Geni-s-World-Tree-A-Retrosp...

As you can see the primary goal for Geni is and the whole time has been a World Tree, so it should not be a surprise for anyone that they now and then starts a cleanup process to get rid of isolated trees by adding them to the world tree, and this is not the first time Geni opens up for both users and curators, to help with that.

I can agree that it was some bad design that inactive users "got the blame" for requesting, sometimes horrible, merges. This list was however only for the curators to see (and we still have an only visible for us world tree merge queue), but when it started to bleed through to regular users we alerted Geni staff that they should change the design, but as you know making changes in the code of an active web site without interrupting it might take time.

If you take a closer look those merge suggestions are now flagged with a big warning at the top telling that they are automatically generated.

Geni allows standalone trees, simply because most of us start with that unless we get invited into an existing tree. People who wants a totally isolated tree should keep in their own personal computer and not publish it online, but as a compromise the owner company site, MyHeriatage could be considered.

Private User
2/13/2015 at 2:28 AM

Now that they've gotten the worst bugs out, I'm in favor. Some good matches came out of it, as well as some really wacko ones.

Private User
2/13/2015 at 2:44 AM

If you see bad matches or even suspect some, just reject them once and for all before anyone else completes them. Not everyone are as good as you seeing that something must be wrong.

There is always better matches elsewhere in the tree and once the merge has begun and it happens that the merge you rejected happened to be correct anyhow it will be included as in a zipper-function, - or at least you will have better information then to cut bad connections by comparing the connections from the other tree.

Showing all 7 posts

Create a free account or login to participate in this discussion