Introducing the Geni Consistency Checker

Private User , Thanks for the "look at." I am going to contact the profile owners as I don't manage this profile, but for males, I have never put in a "birth surname" - only for females who married and I had their "married name" along with the birth surname. If I do not know one or the other, I leave that one blank. I imagine a number of others do this.

This article (published in "The Finnish American Reporter" August 2000) helped me understand some of the Finnish naming issues, but it may not explain all: http://www.genealogia.fi/emi/art/article216e.htm .

If you look at Heikki's father Jacob Johansson Huhtala and his mother Maria Caisa Henricsdotter Kangas , it is just pretty hard to determine what name goes with who and where in my opinion. I think adding "Huhtala" to Heikki's surname at birth is likely to prompt a different kind of inconsistency since it is not a name on either his father's or mother's profile. Just my confused thoughts.

The New Geni feature with inconstancies is making Icelandic trees filled with inconstancies since we in Iceland dont use surnames.

@Guðbjörn Ívar Kjartansson :
The same goes for Norwegian names before 1900. Frozen surnames became more and more common from the mid-19th century, in urban areas. Not until 1923 was this enacted for all by law.

Before that, it was patronyms that applied. Son and daughter were given a patronymic name after the father, e.g. "Ole Andersen" (Ole, son of Anders) or "Kari Andersdatter" (Kari, daughter of Anders).

Surname-like names were used, often in the form of farm names, but had more function as an address, e.g. "Ole Andersen Dalen" (Ole, son of Anders, in the Valley) Many Norwegian genealogists choose to set the farm where a person was born as a surname. Therefore, it is also common for children of these to not have the same surname as their parents. Children may also be born on different farms.

In the period when frozen surnames became more common, and eventually fastened by law, it was also common for siblings to choose different names. Typically, people in the countryside froze the farm name where they lived, while those who moved to the city froze a patronym as surname.

Similar name traditions are also common in the other Nordic countries, and in several places in Eastern Europe.

As genealogists, I believe these nuances are important to protect. I'm afraid users here and there may change profiles to "clean up" in "inconstancies". I see a tendency for this when I get hits on MyHeritage profiles, where one often sees the use of sur-names and maiden names far down in the Middle Ages.

This doesn't make sense!

Private User

In Finland, old surnames have gradually become established by various nicknames. All old nicknames are also possible as last names. The surname can be based on 1) house name (or village name), 2) host name, 3) excess name, 4) professional name

Even in the 19th century, the same person may have been known by three different names, such as Matti Kestilä, or Leisto, or Junnila, in 1824.

Also land ownership has affected the surname many occasions more than the family ties. Comparable to this is the western Finnish way that the owner of the house took the surname of the house and if he changed the house the surname also changed. Even in these cases, two names, both old and new, may have been used.

There are also differences between the surnames in Eastern Finland and Western Finland. Also women kept their name and did not take the name of their spouse until the 20th century.

So, for Finland, that part:"child does not have the same last name as father" needs to be corrected in this program.

On the one that I looked at, the system was giving an error message because the birth surname field was blank.

If you can advise an exact calculation that is always correct, they should be able to code it so that the error does not occur. For example if birth country = Finland AND birth date before xxxx then don’t check birth surname.

If there is no hard rule and you are sure that the profile is correct, you can select ignore.

Arne Åsbjørn Drangeid said, "As genealogists, I believe these nuances are important to protect. I'm afraid users here and there may change profiles to "clean up" in "inconsistencies". I see a tendency for this when I get hits on MyHeritage profiles, where one often sees the use of sur-names and maiden names far down in the Middle Ages."

I wish I could give a thumbs up to that statement.

Private User, I am not familiar enough with many in my tree to hit the ignore button, especially if I don't manage the profile. Google translate only helps me so much. I am going to contact some profile managers to ask for assistance, but I have hit ignore for all on several where "married too young" appeared when I knew it was correct on profiles I managed or was familiar with to the point of adding evidence/sources and the inconsistency does not go away always even refreshing it.

Susanne Floyd, there were severe performance issues this weekend. This could have had some influence on ignoring inconsistencies.
Also the coding for some of the inconsistencies have changed not so very long ago. So those inconsistencies have to be recalculated and that is taking some time.
If you ignored an inconsistency for which the check conditions where changed the ignore probably will not be remembered until you ignore the latest version.

Guðbjörn Ívar Kjartansson : For the Icelanders, it should really be simple, since you still have a system of consistent, real patronyms: Leave the "surname" and "birth name" fields blank, and fill in all the first names in the "name" field, and the partonym (like "Kjartansson") in the field for "middle name / patronym"...

Thanks Job Waterreus and Private User for your patience with me trying to explain this. I know that there are kinks with any new thing. You all are doing what you can and amazingly, I think it will get better. Arne Åsbjørn Drangeid has already figured out what to do for our Icelander friends. I guess we just keep checking it out. :-)

Ewww. That means redoing ALL the Icelanders, ALL the pre-Tudor Welsh, ALL the pre-modern Scandinavians, etc. etc. etc.

Why not just modify the checker instead of forcing everybody to slog through years and years and years of work?

Private User, If you can advise an exact calculation that is always correct, they should be able to code it so that the error does not occur. For example if birth country = Iceland AND birth date before xxxx then don’t check birth surname.

Iceland wouldn't require a "birth date" conditional - they STILL use patronymics, not surnames.

Sweden had mix-and-match naming systems until late in the last century (patronymics, "farm names", "army names", generic locatives, and a scattering of consistent surnames), and didn't legally mandate surnames until 1966(!).

It is by now well known that the Welsh didn't consistently use surnames until Henry VIII Tudor forced them to (c. 1536).

There's actually a pretty good article on Wikipedia going over who did use patronymics, for how long, and which peoples still do: https://en.wikipedia.org/wiki/Patronymic

So maybe a surname mismatch should trigger a query, "Is this correct?", and if the answer is Yes then set to Ignore.

There also appears to be a problem for the name check between parents and children.
Example: https://www.geni.com/inconsistencies/for_profile/6000000001958304355
The child has in the also known as the value of "Kole" which is the same as the father has for last name as well as surname at birth.

Have just added a husband for a profile. Up springs an inconsistency alert:

AAAA and her spouse BBBB are 35 years apart.
Tip: Correct the birth date of AAAA (born circa 1898) or the birth date of BBBB (born estimated between 1868 and 1928).

have substituted AAAA for wife's name, BBBB for husband's name.
Aside from that, is exactly what they show me -- an approximate date in the wife's profile, husband who was just added had no dates yet in his profile, they provide the Geni approximation of his DOB, which includes the year of his spouse's approximate birth - and this is called a 35 year difference - and I wasted my time clicking on the inconsistency alert to see that -- Mike Stangel is that really how you feel it should be working?

Inconsistency alert was on her profile, and in lower left corner of her profile in Tree View, not his, if that is relevant.

Private User & Mike Stangel, I saw a number of similar warnings (also with one profile with estimated birth) and most (if not all) of them with 35 years as difference.
There is also something very wrong when you use before like these:
Eeke Pieters and Pieter Jarrighs Jarchs

Corrected and added second child and children to this profile, Anders Abrahamsson

He married the second wife at age 69, she was 29, then he had two children, one at the age of 72, the next at age 75. Of course I got message of inconsistency, he was too much older than his wife and he got children at a too high age, but, it's still according to the church book.

So, what's the meaning of this checker, I found this answer in quote below to be the one that most corresponds with the minds of the ones that has set the levels of the checker to be fitting.

"Morality police, is one of many names used to describe groups of people whose job (often self-appointed) is to enforce standards of moral behavior and religious adherence among the general public."

You should feel ashamed!

Hi Private User, I don't believe the intent is to impose a moral standard on anyone.

I have had the same inconsistency and sometimes it is as you have shown a perfectly valid union between an older person and a younger person. But more often I have found that the younger spouse has been incorrectly attached to the wrong generation (ie an uncle and nephew with the same name) or to the completely wrong spouse or a merge has merged completely different people or the incorrect dates have been entered. Without this consistency check, I haven't noticed the error.

In the cases where it is correct (someone older had a younger spouse), I have added a simple biography on the profile overviews and marked it as correct for all.

Here is an example of a relationship that the consistency checker has shown me needs to be investigated

John Freebody and his spouse Sarah Connally, (1898-?) are 94 years apart.
Tip: Correct the birth date of John Freebody (born 1803) or the birth date of Sarah Connally, (1898-?) (born November 08, 1898).

The spouses are 94 years apart in age. The wife was born 68 years after the marriage and 8 years after he died.

I also found this article https://en.wikipedia.org/wiki/Age_disparity_in_sexual_relationships which seems to suggest that for US, Australia and UK, only about 1.3% of marriages/relationships have spouse age differences greater than 20 years. Sorry I haven't read it in depth to see if other locations are widely different or whether it it different historically - it does say " The pattern was also confirmed for the rest of the world, with the gap being largest in Africa." but I am not sure which pattern it is referring to.

I don't know what age Geni flags as a "please review" but from memory I think it is around 35 years difference in age - but we did ask for the list to appear with the largest age difference at the top.

Suzan Martin Wish I could give a thumbs up on your comments. You and I are connected through a different group from those mentioned who migrated South and I think from my research and family lore they may have been Quakers. I am learning so much reading these discussions.

Where you know that the information is correct, you can select Ignore and Hide for all.
Or
Select ignore and hide for me

I'm confused; I've clicked override and ignore all on the married too young. --- Still showing up in "Yellow as a warning". This is annoying for an OCD person. How can I reset this and get this to work correctly?

FYI! Back in American History, they married young, probably too young for GENI standards, but this was a fact. Even the higher class, I may also have some Hillbillies along the way, but who was the genius who decided to put the Marriage equation in the GENI mix? You are discriminating against women during this era, by saying they were too young to be married, I take offense. It's bad enough, when they married at age 20, in the 1700s, on the marriage bond, one of my great-grandmothers was called a Spinster! I consider that rude. Haha. Her sisters were not called Spinster's because they were younger.

Every one of my maternal and paternal lineages migrated to America in the early 1600s, so every we have quite a bit of warning, at least I've gone back over most of my lineage and found other mistakes not related to the marriage factor. I had over 200 on the first day, at least 90% all married too young. They were all from the 1700s...

Do we have directions about how resetting after we override and ignore to get the yellow warnings off?

Overall it's a wonderful feature, I'm finding and repairing many mistakes, so thank you for developing the software. I know it takes time when rolling out new software to work out the kinks.

Just don't mess with Suffix.

Suzan Martin,

I'm not sure about what you mean by Still showing up in "Yellow as a warning".
Can you give an example profile.

Job Waterreus I agree the wording on the Missing Birth Surname for Adrienne Kaan could be better. In that case it's true that he has no birth surname but since his last name doesn't match his spouse, it's not fair to assume that's a married name. In that particular case we have some fuzzy name matching that would have guessed "Caan" is similar to "Kaan" but even so that doesn't mean we would ignore the inconsistency, because for all we know it's a typo instead of a legitimate change in spelling. I'll have to think about how we can soften the language on that inconsistency to mean "there's nothing in his birth surname field and his last name, which doesn't match his spouse, is kinda like a parents' but not exactly."

Private User I believe we fixed the scenario regarding the circa 14-year-old. Let me know if you are still seeing cases like that where the dates are too strict or described improperly.

Private User we adjusted the surname matching algorithm to allow for partial matches such as Spanish naming conventions, and we've re-run all Missing Birth Surname checks. Let me know if you're still seeing incorrect inconsistencies and please post an example or send it to me via Inbox message if you prefer.

Private User I'm not sure I understand the issue.. can you post an example of a "false" inconsistency?

Private User we've added the 5 suffixes you suggested. Those should clear up in a couple days.

Sorry I've run out of time to respond, I'll pick this back up as soon as I can.

Here's a very very good example of why the Suffix field is ABSOLUTELY NECESSARY:

This person Sir Richard de Camville, of Warwickshire

is MUCH too easily mash-merged with THIS person:

Sir Richard de Camville of Oxfordshire

They ARE NOT the same person, they DO NOT have the same wives, brothers, children, etc., and they DID NOT live in the same place.

This is such a difficult and knotty problem that even Cawley(!) got it all bollixed up.

The line of Camville of Oxfordshire came to an end with a third consecutive Richard who had no living brothers or uncles - just a sister Isabel, who married Richard d'Harcourt. Isabel inherited the manor of Stanton from her brother, and her husband claimed it jure uxoris - and it has never since been out of Harcourt hands, thus it is known to this day as Stanton Harcourt.

The line of Camville of Warwickshire continued for several more generations - and the Gerald/Gerard de Camville who married Nicolaa de Haye belonged to THAT family.

"Bennett Phillip de Matas may have a suffix incorrectly added to the name field.
Please check the following, English (default): de, de
Tip: Go to the profile of Bennett Phillip de Matas and move the suffix to the correct field."

Out of the blue, 19 new inconsistencies, like the one above, have appeared on my home page. Two things have confused me. First...why only 19, when the form of the surname that appears in the most instances is "de Matas"? Even my own profile has not shown up as an inconsistency. Secondly, "de Matas" is the correct form of the surname. There is no suffix.

In other words, there is no correction to be made to any of those 19 profiles. As I queried before, ignoring these erroneous inconsistencies does not resolve the issue. They will remain there. I would prefer it if these errors could be removed instead of having to get personally involved.

I have finally figured out what I was doing wrong,, I wasn't clicking "re-check" - good grief I'm so sorry you guys. worker error, reminds me back in the early 80s everyone in the office would tell me the computer was broken, user error.