For FamilySearch to Review - Hope it might lead to long term improvements
LegacyUser
✭✭✭✭
Gordon Collett said: This is a case study for the designers of Family Tree for them to evaluate for the purpose of improving the workings of the program. It does not really need comments from other users about this specific situation but they can certainly still be added.
First, I would like to thank you for the design of the Watch List and Change Log which work very well.
This scenario illustrates both the value of the Watch List and the inherent risks involved in an open edit system when users do not seem to understand how the system works. It also shows that more ways to inhibit inappropriate changes in information could be useful.
Gyri Andorsdatter KJZF-YZ3 was imported from New Family Search in April 2012 with birth and death information as well as both parents from an extracted birth record along with completed ordinances. Using parish records, I later added her christening and burial dates, improved her birth and death dates, and set all the correct places properly. She was born 10 Sep 1820 at Kike, Fjelberg, Hordaland, Norway and died at three months of age. Sources, three in total, were added for all this information. I also worked on her extended family and added information and sources to many of them.
For some reason, another user added a husband to her by the name of Szabó Ferenc with a death date for him of 1944. He then changed Gyri’s name to Békési Ilona, deleted her birth information, deleted her christening information, deleted her burial information, and changed her death information from 10 Sep 1820, Kike, Fjelberg, Hordaland, Norway to 1976, Újkígyós. I did not have Gyri’s name on my watch list so was not notified of this change which sat unnoticed for about a year and a half.
Recently, however, he deleted her relationship to her parents with the reason statement of “d.” He then proceeded to delete the parent-child relationships for all of Gyri’s siblings to their parents and for all of Gyri’s uncles and aunts on both side of the family to their parents. He then deleted the marriage relationships for Gyri’s parents and her grandparents. This left all of Gyri’s grandparents and their children and Gyri’s parents and their children as isolated individuals in Family Tree with no family relationships. Fortunately, he did not change or delete any vital information on any of these people. I was notified of these changes because Gyri’s parents and grandparents are on my watch list.
Using the Change Log for Gyri’s parents and grandparents, I was able to reverse all this and put the families back together in less than two hours. I restored Gyri’s name and all her information, removed her as a wife, and created a new Békési Ilona as Szabó Ferenc’s wife.
The only information that would not easily restore was marriage dates and and places. I did have to add those back in manually for the parents and grandparents. All couple and parent-child sealings restored themselves and displayed properly after reconnecting all the family relationships.
A couple of years ago, this same user had done this same procedure, not quite as severely, and I had to restore a bunch of information. It may have been with one of Gyri’s siblings because I can’t find the prior episode in anyone’s change log but I did recognized his name from the last time he did this.
I did send this user a message about the damage he had done. He has not replied yet.
I don't have any great ideas on how to prevent this type of behavior, but I am very glad it was so easy to reverse. But that was only because it showed up on the report of changes to people I am watching. What damage might be being done in Family Tree in areas where no one is watching?
First, I would like to thank you for the design of the Watch List and Change Log which work very well.
This scenario illustrates both the value of the Watch List and the inherent risks involved in an open edit system when users do not seem to understand how the system works. It also shows that more ways to inhibit inappropriate changes in information could be useful.
Gyri Andorsdatter KJZF-YZ3 was imported from New Family Search in April 2012 with birth and death information as well as both parents from an extracted birth record along with completed ordinances. Using parish records, I later added her christening and burial dates, improved her birth and death dates, and set all the correct places properly. She was born 10 Sep 1820 at Kike, Fjelberg, Hordaland, Norway and died at three months of age. Sources, three in total, were added for all this information. I also worked on her extended family and added information and sources to many of them.
For some reason, another user added a husband to her by the name of Szabó Ferenc with a death date for him of 1944. He then changed Gyri’s name to Békési Ilona, deleted her birth information, deleted her christening information, deleted her burial information, and changed her death information from 10 Sep 1820, Kike, Fjelberg, Hordaland, Norway to 1976, Újkígyós. I did not have Gyri’s name on my watch list so was not notified of this change which sat unnoticed for about a year and a half.
Recently, however, he deleted her relationship to her parents with the reason statement of “d.” He then proceeded to delete the parent-child relationships for all of Gyri’s siblings to their parents and for all of Gyri’s uncles and aunts on both side of the family to their parents. He then deleted the marriage relationships for Gyri’s parents and her grandparents. This left all of Gyri’s grandparents and their children and Gyri’s parents and their children as isolated individuals in Family Tree with no family relationships. Fortunately, he did not change or delete any vital information on any of these people. I was notified of these changes because Gyri’s parents and grandparents are on my watch list.
Using the Change Log for Gyri’s parents and grandparents, I was able to reverse all this and put the families back together in less than two hours. I restored Gyri’s name and all her information, removed her as a wife, and created a new Békési Ilona as Szabó Ferenc’s wife.
The only information that would not easily restore was marriage dates and and places. I did have to add those back in manually for the parents and grandparents. All couple and parent-child sealings restored themselves and displayed properly after reconnecting all the family relationships.
A couple of years ago, this same user had done this same procedure, not quite as severely, and I had to restore a bunch of information. It may have been with one of Gyri’s siblings because I can’t find the prior episode in anyone’s change log but I did recognized his name from the last time he did this.
I did send this user a message about the damage he had done. He has not replied yet.
I don't have any great ideas on how to prevent this type of behavior, but I am very glad it was so easy to reverse. But that was only because it showed up on the report of changes to people I am watching. What damage might be being done in Family Tree in areas where no one is watching?
Tagged:
0
Comments
-
joe martel said: Thanks Gordon for your detailed sequence of events. One of the first rules of FT's open-edit model is to be able to reverse bad changes as easily as it was to mess it up. For the most part it sounds like you were successful. Though this is a good goal it always isn't possible, nor as easily as we wish - some would want one button to reverse the 40 previous ones. I'd like that too and if we were talking about independent person objects that might be possible, but the aspect of relationships between multiple persons, across generations makes it complex. So the fault of not being able to restore the relationship is a symptom of that complexity. Coming up with a non-complicated user interface for complex systems will always be difficult.
To me the one aspect that could alleviate users' pain is to somehow prevent messes from happening. That means a better mechanism to recognize the change will be wrong, and having a computer predict that. Humans and computers excel and fail at different skills. So it is a two part problem - facilitate the humans, and computers recognize bad patterns. FSFT is in its rudimentary stage of development - in supporting collaborative work around collaborative data. So I don't have answers but do appreciate ideas to go the next step to better quality information, and a system that puts heart into the brain - better understanding the fact of our ancestors, along with getting to know them beyond those facts.0 -
Adrian Bruce said: We have seen total hijacking of profiles several times before in GetSat but I don't think that I've ever seen one as radical as that. I struggle to decide what the hijacker (for want of a better term) was thinking of. My best guess is that Gyri was simply the last profile that he'd looked at (no idea why) and that to him, Gyri's profile was simply a piece of paper that could be reused for his purposes. Maybe he thought that everything he entered was his, and his alone, and by the same token, he couldn't possibly be upsetting anyone else because their data was theirs and theirs alone, so he couldn't affect it.
Not sure how we can fix this without getting inside their head..... Well, not properly fix it.0 -
Juli said: Gordon, do you want me to write up something in Hungarian to send to this user? I'm trying for a charitable interpretation of his actions, and the best I can come up with is a language barrier. (He can't figure out how to navigate the "add new person" dialog, so he resorts to hijacking existing profiles.)
Unfortunately, the civil death registers for Újkígyós, Békés, Hungary are only online on FS up to 1971, so I can't tell whether Mr. Szabó and Ms. Békési were a real couple or not. (Without a location for Mr. Szabó's death, I have basically no hope of finding it; it'd be the equivalent of looking for a Frank Smith in Indiana. I did check: no such death was entered in Újkígyós in 1944 -- but there was a death entered of a baby with the same name as the apparently-clueless user.)0 -
Gordon Collett said: Thanks for the offer. Please do. That could be a great help if he just needs to understand, as do so many people, that there is only the single database for Family Tree as Adrian discusses above. He probably has done very little in the tree as it looks like he has created a couple of duplicates for the people he is trying to enter.0
-
gasmodels said: In light of the issue Gordon presented in detail and the continuing issue that many of us have with other users who taken actions that are really almost incomprehensible, it is clear that unless some action needs to be taken to reduce these type of activities or this will be an ongoing issue forever. It appears that no amount of sourcing or other documentation prevents actions of this type. Joe has suggested that one alternative is to develop some automated response from the system to indicate that the activity is likely to be incorrect and this might inhibit the user from completing the change.
I am going to suggest an additional approach. Others have suggested that all new users need to undergo some sort of certification training but it appears that nothing along that path is forthcoming. Why not consider have a select group in support review these kinds of situations and make a personal contact with the user to explain what they have done and how it damages the system. This could be done in a way so that it is not a "slap on the hand" but more of a help with understanding how to use the system. We now rely on one user talking to another by messaging and in my mind this has been quite unsuccessful as a global solution. I think it is time for the department to take some direct action to help alleviate these damaging changes particularly when they are repeated by the same user. Just an alternative that would be relatively easy to implement ( maybe finding the right support individuals would require some effort) but would directly attack the users who do not understand how the system works.0 -
Brett said: Gordon
As an aside ...
I have NOT looked at the "ChangeLog" in the aforementioned case of yours; but, ...
Question:
Did the other User/Patron "Add" a Spouse/Husband, BEFORE; or, AFTER, "Changing" the Birth and Death records?
Because; IF, the User/Patron "Added" a Spouse/Husband BEFORE "Changing" the Birth and Death records; then, this is a case where "FamilySearch", the "System", should NOT allow the addition of a Spouse if the individual/person Died as an Infant/Child - which I have come across a number of times (for well 'Documented"/"Sources" Infants/Children).
Whereas, of course; IF, the User/Patron "Added" a Spouse/Husband AFTER "Changing" the Birth and Death records; then, in such a case there should be no such impediment in "FamilySearch", the "System".
Just a thought.
Brett
ps: I not so long ago, found a individual/person, that was "Documented"/"Sourced"; but, not extensively so, where it had been indicated that they had Died as a Child, where only the "SP" was "Completed". Through thorough research, I noticed that the "Death" record that had been referenced was incorrect; and, actually related to another "Unrelated" Child, who had died young from a place that was in a nearby County. Whereas, in this case I corrected the individual's/person's Death (life span) and extensively "Documented"/"Sourced" the "Changes". But, in my case, I at least have some experience in Genealogy/Family History and the "System" - as we all know, NOT all Users/Patrons have such experience.
.0 -
Gordon Collett said: In the change log, it looks like the husband was added to her first, then all her information was changed.0
-
Juli said: Kedves [username]!
Immár másodszor is "eltéritette" valamelyik rokonom profilját az FS-en, és valaki egészen mást próbált belőle csinálni. Arra gondolok, hogy talán nincs tisztában a FamilySearch családfájának felállításával és használatával.
Az FSen -- az Ancestrytől, MyHeritagetől, és hasonló internet-szolgáltatásoktól eltérően -- nincsenek egyedi családfák. Ez mind egy nagy közös fa, amin bárki módosíthat bárhol bármit (bejelentkezés után). A cél, hogy minden elhunyt személyről, aki valaha élt, legyen egy -- és csak egy! -- profil.
Az év elején igérték, hogy rövidesen lesz magyar fordítása az FS felületének. Addig is talán jó lenne angolul tudó rokon vagy szomszéd segítségével használni a weboldalt, a fenti célt szem előtt tartva.
Köszönettel,
[signature]
Dear []!
This is the second time you have "hijacked" one of my relatives and tried to turn her profile on FS into someone completely different. I think that maybe you are not clear about the setup and use of FamilySearch's family tree.
On FS -- unlike on Ancestry, MyHeritage, and similar internet services -- there are no individual family trees. This is all one big shared tree, on which anyone can modify anything anywhere (after signing in). The goal is to have one -- and only one! -- profile for every deceased person who ever lived.
At the beginning of this year, they promised that a Hungarian translation of FS's interface would be available shortly. Until then, perhaps it would be good to use the site with the help of a relative or neighbor who speaks English, keeping the above-mentioned goal in mind.
Thank you,
[]0 -
Gordon Collett said: Thanks. Did you send this? Or was your intent that I copy and send it?0
-
Brett said: Gordon
In that case, that is were we need that impediment - "FamilySearch", the "System", should NOT allow the addition of a Spouse; IF, the individual/person, Died, as an Infant/Child (very young).
Of course, the impediment could be easily lifted by any User/Patron; IF, the Death record, was subsequently "Changed", to indicate that the individual/person was, in fact, OLDER, than just that of the Dead, Infant or very young Child.
That impediment would NOT totally stop the situation; but, would just put an extra level of protection; where, the User/Patron, had to make a conscious "Change" (effort); BEFORE, "Adding" a Spouse to an Infant/Child, who died, at a very young age.
As I was once told by a "Programmer" ... "You cannot make a 'System' FOOLPROOF; because, fools are so ingenious!" - so true ...
Of course, like everyone, I class myself as one of those fools ...
Just my thoughts.
Brett
.0 -
Juli said: The latter: I have not sent it.0
-
Juli said: I have a T-shirt: "Any fool can use a computer. Many do."0
-
Brett said: Juli
I like that.
Brett
.0 -
Paul said: In line with the criticism aimed at Facebook and Twitter, I am surprised there has not been more comment on the fact that FamilySearch appear unwilling to apply any resources to what amounts to attacks on its Family Tree product. Not "attacks" in the form of criticising the software itself, but in the sense of individuals causing regular damage, either by accident or design.
Of course, FamilySearch's resources are minute compared to those large organisations, but surely where there is frequent abuse of the system by certain individuals there must be the provision for either warnings or, in some cases, sanctions - including complete blocking of any participation.
The issue of damaging work is one that affects many of us - causing hours, even days, of corrective work, when we should be spending our time enhancing the contents of Family Tree - not putting it back to how it was before a careless (perhaps even malicious) individual messed things up.
Where damage is accidental, the programmers could play their part by NOT offering us ridiculous "possible matches" or "record hints". So much of my "corrective" work is down to this factor alone, with inexperienced users merging / attaching every suggestion given by the "loose" algorithms.
With the open-edit nature of Family Tree must come some degree of regulation to control bad behaviours. I can't imagine the true extent of the damage being done on individual IDs I have worked on. I have just over 1,000 persons on my watch list and am constantly having to "revisit" them to undo erroneous changes. I must be missing many, many other pieces of damaging work on those IDs I do not have on watch.
Please, FamilySearch management, take this matter very seriously, or the project status will start to decline instead of moving in the desired direction.0 -
Paul said: BTW - I have sent numerous messages to those who have made crazy changes / additions / merges. A common response is on the lines of, "Thank you for letting me know". I have never received a justification / explanation of why they made the error(s) and, quite commonly, do not receive any reply at all.
Like many users, I am trying to do my best to get rid of as much fiction as possible, but do believe FamilySearch could do just a little more to help - surely they don't want the current havoc to continue? Examine some branches in the Tree you do not give regular attention to and I believe you will see just how bad things are.0 -
Tom Huber said: A while back I suggested a number of enhancements for the merge and the source linker... These same enhancements could be used when an edit of a record takes place that has problems when compared to the existing data.
Something like this would not be easy to implement, but I have hopes that one day it may be part of the system.0 -
Gordon Collett said: Ron Tanner mentioned in his Q&A 2/13 that the new merge routine can be used in the beta platform now ( beta.familysearch.org ). I haven't looked at it yet.0
-
Jessie Hearle said: New merge from beta.
These two women were merged by another person, they are aunt-neice
The merge was very destructive as the other person didn’t just unlink what she perceived to be incorrect relationships, she stripped all attached sources from them as well.
I hope this merge routine is implemented soon.
It won’t prevent hijacked records, but it’s a huge improvement.
0 -
Paul said: One minor criticism here - again FamilySearch chooses to call everybody concerned our "ancestor". Comparitively few of the individuals we deal with are likely to be ancestors - most are relatives, albeit some quite distant ones.
The word "ancestor" has been dropped / replaced elsewhere, so hopefully it will be changed to "relative" before this new feature reaches the production version of FT.0 -
Gordon Collett said: It would probably be good to start a new post with comments about the new merging process. Something along the lines of what will immediately appear when it is released into production but without the cries of "why didn't you warn us about this!" I've got a couple of minutes, so I think I will.
"Your Ancestor" could probably just be replaced with "Original Person." I do a lot of merges in the cluster research I am doing in Stord, Norway, for people my wife is not related to at all to assemble families from extraction records while working to find all the people she is related to.0 -
W David Samuelsen said: Regulation does exist with bad behaviors already because in one case, it was super bad bad that several cousins joined forces and reported one person to "Abuse" section.
The section took care of that guy. (Banned for life due to the language as well )0 -
A van Helsdingen said: I was recently looking at the Upload Guidelines and Policies, and noticed point 3: Cross-Dressing is not permitted in uploaded photos.
So FS considers uploading photos with cross-dressing to be bad enough to require prohibition, while taking a laissez-faire approach with vandalism in the FSFT, which ruins thousands of profiles, causing more work for genealogists, presumably upsets the ordinance program of FamilySearch's sponsor the Church of Jesus Christ of Latter Day Saints, and wrecks the reputation of the FSFT. But it is the former that is prohibited while action is rarely taken against the latter.0 -
ATP said: On the other hand, when you really think about it, this free site would not even exist if ordinance work was not major doctrine of the Church of Jesus Christ of Latter Day Saints creating and informing its existence.0
-
A van Helsdingen said: How is that relevant to my point that FS would be better off using its limited resources that come from volunteers to improve the quality of the FSFT by reducing vandalism and sloppy edits, instead of policing cross-dressing in uploaded photos?0
-
ATP said: I emphasize "free" because, before the microchip revolutionized computer technology and "free" research became available, it could cost real money searching for primary records and paying for them in various state archives and libraries and county courts and other sites holding local historical records, not to say the cost of the travel. Anyhow, just a thought.0
-
David Newton said: We could say to kiss and make up, but unfortunately that's prohibited as well.0
-
ATP said: David Newton,
Great irony! LOL!0
This discussion has been closed.